DRAFT
(Not all Hyperlinks are Active)
Activity Based Costing
Understanding Process Definitions and Industry
Applications
Knowledge Management Solutions, Inc.
January 2001
Knowledge Management Solutions, Inc.
January 2001
1
DRAFT
(Not all Hyperlinks are Active)
Table of Contents
Overview ……………………………………………………………………………….
3
ABC Glossary of Terms ……………………………………………………………..
5
Articles and Case Study Synopses ……………………………………………….. 12
Pitfalls of ABC and How to Avoid Them (Three Parts of a 5-Part Series) …. 13
By Gary Cokins
Examples and Applications
1. Activity Based Costing –
Getting Closer to the Root of Cost Drivers in the Warehouse …………………….
2. Beyond Theory: The Practice of defining Activities …………………………………….
3. The fifteen defining characteristics of an ABC/M system
(and without them, there is no way that a vendor can claim they do ABC/M) .....
4. Government Uses of Activity Based Costing ……………………………………………
5. ABC at NASA’s Lewis Research Center …………………………………………………
6. If ABC is the Answer, What is the Question? ……………………………………………
40
43
49
53
58
63
Case Studies – Applications
1.
2.
3.
4.
5.
Banc One Assesses Private label Credit Services ……………………………………...
ABC Gathers Speed at Mobil ……………………………………………………………...
U.S. Airways – Implementation Lands $4.3M in Process Improvement Savings ……
United Technologies’ Activity Based Accounting Is A Catalyst For Success ………...
The Change is Forever: Activity-Based Costing in the U.S. Marine Corps …………..
Knowledge Management Solutions, Inc.
January 2001
70
74
83
89
95
2
DRAFT
(Not all Hyperlinks are Active)
1.1 Overview
The Planning, Programming, and Budgeting System (PPBS) is a very complex concept
and the fast paced environment of Pentagon policies and shortened timelines make
advocating for specific areas all the more difficult. Such is the case with advocates of
Base Operating Support (BOS) within the Air Staff BOS structure.
Many organizational management models for problem & process analysis have either
been proposed or are currently being used within business, industry, and government to
improve the effectiveness and efficiency in operations.
While the selection of a model to use could be hampered if an attempt to review the
pros and cons of all models were made, this selection was driven by prior experience
and the need for timeliness.
KMS combined two models – the ADDIE and the ABC. Experience with the Air Force
Analysis, Design, Develop, Implement, and Evaluate (ADDIE) model suggested these
process design and development steps were prudent for this endeavor. Similarly, the
Activity Based Costing (ABC) model was an obvious selection from experience within
the financial management arena and from an understanding of the need for outputbased metric development within the BOS PPBS process. Providing a logical plan,
regardless of the complexity, was critical and would allow all participants to reach the
intended program goals with reliable, validated and reproducible results. By looking
closer at both models, we can To understand the models, let’s briefly examine each.
The ADDIE Model is an acronym for Analyze, Design, Develop, Implement, and
Evaluate. It is a derivative of the Instructional Systems Design (ISD) process used
throughout business, industry, academia, and government.
The Air Force is a long time proponent of the Instructional Systems Design (ISD)
process for defining, developing, and implementing (in a very logical and organized
manner) complex systems and for assisting personnel in understanding their associated
intricacies. The understanding and adaptation of the AF BOS PPBS process toward
better method for planning, advocating, and forecasting requirements seemed to be a
similar cause.
Instructional System Development (ISD) is the official Air Force process for developing
education and training for Air Force personnel. AFPD 36-22 directs the use of a
systematic approach to Air Force instruction. AF Manual 36-2234 discusses the
background of ISD, explains the process, and provides guidance on specific areas of
the process. It also provides the necessary theory and philosophy for developing
effective, cost-efficient instruction and process model development. It is this process
development and logical approach that made the ISD process appropriate for use.
So, what is ABC?
The ABC Model of Activity Based Costing is an industry-accepted approach to
maximizing resource utilization and associated activity efficiency while minimizing the
costs of outputs. It is this focus on the costs of outputs that make ABC different from
the typical input-based business model. By concentrating on the outputs of each major
organizational activity, ABC portrays cost-distribution throughout the organizational
processes of resource utilization, activity implementation, and output production.
Knowledge Management Solutions, Inc.
January 2001
3
DRAFT
(Not all Hyperlinks are Active)
Organizations devote thousands of hours and many personnel at the discipline of its study and
analysis. We understand the short timeframes and fast-paced schedules associated with your
Product Element Manager (PEM) and Integrated Product Team (IPT) member positions.
Therefore, the purpose of this reference document is to briefly describe and define the
components of ABC and to provide a reference source for further study by you, as time allows.
The IPT meeting slides present a task sequence combining the ADDIE with the ABC
models. The first two major ADDIE steps are combined with the ABC methodology of
identifying specific outputs and costs.
As an IPT member, you’ll recognize these steps from the IPT slides.
Analyze
− Baseline Current BOS PPBS Process
− Define Major Processes and Key Activities
− Target Activities
− Determine and Define Products and Services (Outputs)
− Define Customers
− Identify Existing Qualitative Data (Performance Measures)
− Identify Existing Quantitative Data (Costs) by Activity
− Analyze Existing Activity Metrics and Processes
Design
− Link Activities to Processes
− Determine Steps (Critical Path) associated with each Activity Process
− Identify Cost Drivers (Standards)
− Identify Activity & Resource Drivers (Frequency and Intensity)
− Design a Cost Structure for each Activity Process
− Design Future BOS PPBS Process Model
§ Design Narrative Justification
§ Design Wiring Diagram
− Design Output-Based Metric Criteria
− Design Metric Presentation Format
− Design Metric Tracking Methods
The ADDIE model can be said to comprise the ‘How’ of the logical flow to the problem
and process analysis and the Activity Based Costing (ABC) model provides the ‘What’
for the details within the financial costing and BOS PPBS environment. This
combination provides a simple, yet extremely effective, method for analyzing the PPBS
process, defining a procedural direction, laying out the steps, and suggesting the
solution.
Included here are initial descriptions and definitions giving the reader an upfront
description and definition for critical terms. Selected articles are available in the
appendices from government and industry and combine application examples with case
studies.
Knowledge Management Solutions, Inc.
January 2001
4
DRAFT
(Not all Hyperlinks are Active)
Definitions – ABC Glossary of Terms
This glossary appears courtesy of the Consortium of Advanced ManufacturingInternational (CAM-I).
ABC Model
A representation of resource costs during a time period that are consumed through
activities and traced to products, services, and customers or to any other object that
creates a demand for the activity to be performed.
ABC System
A system that maintains financial and operating data on an organization’s resources,
activities, drivers, objects and measures. ABC models are created and maintained
within this system.
Activity
Work performed by people, equipment, technologies or facilities. Activities are usually
described by the “action-verb-adjective-noun” grammar convention. Activities may occur
in a linked sequence and activity-to-activity assignments may exist.
Activity Analysis
The process of identifying and cataloging activities for detailed understanding and
documentation of their characteristics. An activity analysis is accomplished by means of
interviews, group sessions, questionnaires, observations, and reviews of physical
records of work.
Activity-Based Budgeting (ABB) An approach to budgeting where a company uses an
understanding of its activities and driver relationships to quantitatively estimate
workload and resource requirements as part of an ongoing business plan. Budgets
show the types, number of and cost of resources that activities are expected to
consume, based on forecasted workloads. The budget is part of an organization’s
activity-based planning process and can be used in evaluating its success in setting and
pursuing strategic goals. (See Activity-Based Planning.)
Activity-Based Costing (ABC)
A methodology that measures the cost and performance of cost objects, activities and
resources. Cost objects consume activities and activities consume resources. Resource
costs are assigned to activities based on their use of those resources, and activity costs
are reassigned to cost objects (outputs) based on the cost objects’ proportional use of
those activities. Activity-based costing incorporates causal relationships between cost
objects and activities and between activities and resources.
Activity-Based Management (ABM)
A discipline focusing on the management of activities within business processes as the
Knowledge Management Solutions, Inc.
January 2001
5
DRAFT
(Not all Hyperlinks are Active)
route to continuously improve both the value received by customers and the profit
earned in providing that value. ABM uses activity-based cost information and
performance measurements to influence management action. (See Activity-Based
Costing.)
Activity-Based Planning (ABP)
Activity-based planning (ABP) is an ongoing process to determine activity and resource
requirements (both financial and operational) based on the ongoing demand of products
or services by specific customer needs. Resource requirements are compared to
resources available and capacity issues are identified and managed. Activity-based
budgeting (ABB) is based on the outputs of activity-based planning. (See Activity-Based
Budgeting.)
Activity Dictionary
A listing and description of activities that provides a common/standard definition of
activities across the organization. An activity dictionary can include information about an
activity and/or its relationships, such as activity description, business process, function
source, whether value-added, inputs, outputs, supplier, customer, output measures,
cost drivers, attributes, tasks, and other information as desired to describe the activity.
Activity Driver
The best single quantitative measure of the frequency and intensity of the demands
placed on an activity by cost objects or other activities. It is used to assign activity costs
to cost objects or to other activities.
Activity Level
A description of how elastic or sensitive an activity is to changes in the volume,
diversity, or complexity of a cost object or another activity. Product-related activity levels
may include unit, batch, and product levels. Customer-related activity levels may include
customer, market, channel, and project levels.
Allocation
A distribution of costs using calculations that may be unrelated to physical observations
or direct or repeatable cause-and-effect relationships. Because of the arbitrary nature of
allocations, costs based on cost causal assignment are viewed as more relevant for
management decision-making. (Contrast with Tracing and Assignment.)
Assignment
A distribution of costs using causal relationships. Because cost causal relationships are
viewed as more relevant for management decision-making, assignment of costs is
generally preferable to allocation techniques. (Synonymous with Tracing. Contrast with
Allocation.)
Attributes
A label used to provide additional classification or information about a resource, activity,
or cost object. Used for focusing attention and may be subjective. Examples are a
Knowledge Management Solutions, Inc.
January 2001
6
DRAFT
(Not all Hyperlinks are Active)
characteristic, a score or grade of product or activity, or groupings of these items, and
performance measures.
Best Practices
A methodology that identifies the measurement or performance by which other similar
items will be judged. This methodology is used to establish performance standards and
to aid in identifying opportunities to increase effectiveness and efficiency. Best practices
methodology may be applied with respect to resources, activities, cost object, or
processes.
Bill of Activities
A listing of activities required by a product, service, process output or other cost object.
Bill of activity attributes could include volume and or cost of each activity in the listing.
Bill of Resources
A listing of resources required by an activity. Resource attributes could include cost and
volumes.
Capacity
The physical facilities, personnel and process available to meet the product or service
needs of customers. Capacity generally refers to the maximum output or producing
ability of a machine, a person, a process, a factory, a product, or a service. (See
Capacity Management.)
Capacity Management
The domain of cost management that is grounded in the concept that capacity should
be understood, defined, and measured for each level in the organization to include
market segments, products, processes, activities, and resources. In each of these
applications, capacity is defined in a hierarchy of idle, non-productive, and productive
views.
Constraint
A bottleneck, obstacle or planned control that limits throughput or the utilization of
capacity.
Cost Center
A sub-unit in an organization that is responsible for costs.
Cost Driver
Any situation or event that causes a change in the consumption of a resource, or
influences quality or cycle time. An activity may have multiple cost drivers. Cost drivers
do not necessarily need to be quantified; however, they strongly influence the selection
and magnitude of resource drivers and activity drivers.
Cost Driver Analysis
The examination, quantification, and explanation of the effects of cost drivers. The
Knowledge Management Solutions, Inc.
January 2001
7
DRAFT
(Not all Hyperlinks are Active)
results are often used for continuous improvement programs to reduce throughput
times, improve quality, and reduce cost.
Cost Element
The lowest level component of a resource, activity, or cost object.
Cost Management
The management and control of activities and drivers to calculate accurate product and
service costs, improve business processes, eliminate waste, influence cost drivers, and
plan operations. The resulting information will have utility in setting and evaluating an
organization’s strategies.
Cost Object
Any product, service, customer, contract, project, process or other work unit for which a
separate cost measurement is desired.
Cost Object Driver
The best single quantitative measure of the frequency and intensity of demands placed
on a cost object by other cost objects.
Cost Pool
A logical grouping of Resources or Activities aggregated to simplify the assignment of
resources to activities or activities to cost objects. Elements within a group may be
aggregated or disaggregated depending on the informational and accuracy
requirements of the use of the data. A modifier may be appended to further describe the
group of costs, i.e. Activity Cost Pool.
Cross-Subsidy
The inequitable assignment of costs to cost objects, which leads to over costing or
under costing them relative to the amount of activities and resources actually
consumed. This may result in poor management decisions that are inconsistent with the
economic goals of the organization.
Direct Cost
A cost that can be directly traced to a cost object since a direct or repeatable causeand-effect relationship exists. A direct cost uses a direct assignment or cost causal
relationship to transfer costs. (See also Indirect Cost, Tracing.)
Enterprise-Wide ABM
A management information system that uses activity-based information to facilitate
decision making across an organization.
Hierarchy of Cost Assignability
An approach to group activity costs at the level of an organization where they are
incurred, or can be directly related to. Examples are the level where individual units are
identified (unit-level), where batches of units are organized or processed (batch-level),
Knowledge Management Solutions, Inc.
January 2001
8
DRAFT
(Not all Hyperlinks are Active)
where a process is operated or supported (process-level), or where costs cannot be
objectively assigned to lower level activities or processes (facility-level). This approach
is used to better understand the nature of the costs, including the level in the
organization at which they are incurred, the level to which they can be initially assigned
(attached) and the degree to which they are assignable to other activity and/or cost
object levels, i.e. activity or cost object cost, or sustaining costs.
Indirect Cost A resource or activity cost that cannot be directly traced to a final cost
object since no direct or repeatable cause-and-effect relationship exists. An indirect cost
uses an assignment or allocation to transfer cost. (See Direct Cost, Support Costs.)
Life Cycle Cost
A product’s life cycle is the period that starts with the initial product conceptualization
and ends with the withdrawal of the product from the marketplace and final disposition.
A product life cycle is characterized by certain defined stages, including research,
development, introduction, maturity, decline, and abandonment. Life cycle cost is the
accumulated costs incurred by a product during these stages.
Pareto Analysis
An analysis that compares cumulative percentages of the rank ordering of costs, cost
drivers, profits or other attributes to determine whether a minority of elements have a
disproportionate impact. For example, identifying that 20 percent of a set of independent
variables is responsible for 80 percent of the effect.
Performance Measures
Indicators of the work performed and the results achieved in an activity, process, or
organizational unit. Performance measures are both non-financial and financial.
Performance measures enable periodic comparisons and benchmarking.
Process
A series of time-based activities that are linked to complete a specific output.
Profitability Analysis
The analysis of profit derived from cost objects with the view to improve or optimize
profitability. Multiple views may be analyzed, such as market segment, customer,
distribution channel, product families, products, technologies, platforms, regions,
manufacturing capacity, etc.
Resource Driver
The best single quantitative measure of the frequency and intensity of demands placed
on a resource by other resources, activities, or cost objects. It is used to assign
resource costs to activities, and cost objects, or to other resources.
Resources
Economic elements applied or used in the performance of activities or to directly
support cost objects. They include people, materials, supplies, equipment, technologies
Knowledge Management Solutions, Inc.
January 2001
9
DRAFT
(Not all Hyperlinks are Active)
and facilities. (See Resource Driver, Capacity.)
Support Costs
Costs of activities not directly associated with producing or delivering products or
services. Examples are the costs of information systems, process engineering and
purchasing. (See Indirect Cost.)
Surrogate [item] Driver
A substitute for the ideal driver, but is closely correlated to the ideal driver, where [item]
is Resource, Activity, Cost Object. A surrogate driver is used to significantly reduce the
cost of measurement while not significantly reducing accuracy. For example, the
number of production runs is not descriptive of the material disbursing activity, but the
number of production runs may be used as an activity driver if material disbursements
correlate well with the number of production runs.
Sustaining Activity
An activity that benefits an organizational unit as a whole, but not any specific cost
object.
Target Costing
A target cost is calculated by subtracting a desired profit margin from an estimated or a
market-based price to arrive at a desired production, engineering, or marketing cost.
This may not be the initial production cost, but one expected to be achieved during the
mature production stage. Target costing is a method used in the analysis of product
design that involves estimating a target cost and then designing the product/service to
meet that cost. (See Value Analysis.)
Tasks
The breakdown of the work in an activity into smaller elements.
Tracing
The practice of relating resources, activities and cost objects using the drivers
underlying their cost causal relationships. The purpose of tracing is to observe and
understand how costs are arising in the normal course of business operations.
(Synonymous with Assignment. Contrast with Allocation.)
Unit Cost
The cost associated with a single unit of measure underlying a resource, activity,
product or service. It is calculated by dividing the total cost by the measured volume.
Unit cost measurement must be used with caution as it may not always be practical or
relevant in all aspects of cost management.
Unit of Driver Measure
The common denominator between groupings of similar activities. Example: 20 hours of
process time is performed in an activity center. This time equates to a number of
common activities varying in process time duration. The unit of measure is a standard
Knowledge Management Solutions, Inc.
January 2001
10
DRAFT
(Not all Hyperlinks are Active)
measure of time such as a minute or an hour.
Value-Adding/Non-Value-Adding
Assessing the relative value of activities according to how they contribute to customer
value or to meeting an organization’s needs. The degree of contribution reflects the
influence of an activity’s cost driver(s).
Value Analysis
A method to determine how features of a product or service relate to cost, functionality,
appeal and utility to a customer. (i.e., engineering value analysis) (See Target Costing.)
Value Chain Analysis
A method to identify all the elements in the linkage of activities a firm relies on to secure
the necessary materials and services, starting from their point of origin, to manufacture,
and to distribute their products and services to an end user.
This glossary appears courtesy of the Consortium of Advanced ManufacturingInternational (CAM-I).
Knowledge Management Solutions, Inc.
January 2001
11
DRAFT
(Not all Hyperlinks are Active)
Article and Case Study Synopses
(Synopses Will Be Incorporated When Completed)
Pitfalls of ABC and How to Avoid Them (Three Parts of a 5-Part Series)
By Gary Cokins
Examples and Applications
1. Activity Based Costing – Getting Closer to the Root of Cost Drivers in the
Warehouse
2. Beyond Theory: The Practice of defining Activities
3. The fifteen defining characteristics of an ABC/M system (and without them,
there is no way that a vendor can claim they do ABC/M)
4. Government Uses of Activity Based Costing
5. ABC at NASA’s Lewis Research Center
6. If ABC is the Answer, What is the Question?
Case Studies – Applications
1. Banc One Assesses Private label Credit Services
2. ABC Gathers Speed at Mobil
3. U.S. Airways – Implementation Lands $4.3M in Process Improvement
Savings
4. United Technologies’ Activity Based Accounting Is A Catalyst For Success
5. The Change is Forever: Activity-Based Costing in the U.S. Marine Corps
Knowledge Management Solutions, Inc.
January 2001
12
DRAFT
(Not all Hyperlinks are Active)
Pitfalls of ABC and How to Avoid Them
Common Pitfalls in Developing an ABC Model (And How to Avoid Them), Part 1
by Gary Cokins, CPIM for BetterManagement.com
When an organization’s activity-based costing (ABC) implementation project
falls short of expectations, it rarely has anything to do with ABC methodology or
the technology that now supports ABC. ABC is merely a refined form of
absorption costing, so there are no major adjustments in accounting theory. And
computations once restricted to cumbersome mainframe computers can now be
performed with flexible modeling ABC software on laptop computers and other
user accessible platforms. But regardless of where the blame lies, failed
implementations in the 1990s tarnished the image of the activity-based costing.
An entire ABC project will be at risk for failure unless project leaders address all
the causes for lack of acceptance of ABC above and beyond and above the
threshold necessary for success. The causes for failure range from poor design
of the ABC model architecture to underestimating the ABC data users’ level of
resistance to what appears to be a new accounting methodology.
Many failures result from the misconceptions by members of the ABC project team.
Regardless of the problem, however, almost all of the potential pitfalls must be
considered and addressed to achieve the minimal threshold level for success where
users can continue to reuse the data and request updates from period to period.
Five areas expose an ABC project to the risk of failure:
1.
2.
3.
4.
5.
ABC model structure design and architecture
Selection and collection of driver data
Implementation and data integration
Getting management and employee support
Applying the data to decision making
These categories of risk must be addressed in a balanced way. That is, it is more
important to pay some attention to all areas than to overemphasize one or two while
ignoring the others. This article is part one of a five part series and will address the first
category of risk – ABC model structure design and architecture.
Problems Related to ABC Model Structure Design and Architecture
In a short phrase, poor ABC model design will lead to poor results. Since there are so
many other behavioral change management obstacles to be overcome with ABC,
having a sub-par model design makes the task of overcoming organizational resistance
to change only that more difficult.
Precision Versus Accuracy
One common misconception about ABC model structure design and architecture is that
precision is synonymous with accuracy. In other words, if one collects imprecise
Knowledge Management Solutions, Inc.
January 2001
13
DRAFT
(Not all Hyperlinks are Active)
timesheet data from the workers, then all the costs will be corrupted and inaccurate. In
addition, if one uses imprecise activity cost driver data, the costed results of products
and customers are even worse.
ABC project teams are notorious for constructing very large models that exceed, at least
initially, what is needed to solve an organization's major problems. The root of this
problem can be linked to perceptions and skepticism about the use of estimates in place
of facts and about any errors resulting from them. Accountants seek to provide readers
of financial statements with precise information of unquestionable integrity, whether it be
at the summary level or minutely detailed. Accountants' training has taught them that if
the input data is wrong, then any output will be wrong, and therefore, any uses of that
output data will be adversely affected.
Consequently, when designing and implementing an ABC system, accountants naturally
presume a detailed and comprehensive level of data collection. They assume that if
they collect great detail from everyone in every area, then well after-the-fact they know
they can answer any remote question by rolling-up the data into summaries.
Imprecise inputs, however, do not automatically mean inaccurate outputs, and precision
is not always synonymous with accuracy. It is counterintuitive to many, but with ABC,
error does not compound, it dampens out. The real consequence of an error when
distributing costs is that some destination cost objects are overcosted while the
remainder are undercosted. It is a zero-error game, and, in the end, 100 percent of the
costs are always completely assigned for each and every assignment. Figure 1
illustrates how ABC produces diminishing returns on error for additional increments of
administrative effort.
Figure 1: Diminishing Returns Curve
Knowledge Management Solutions, Inc.
January 2001
14
DRAFT
(Not all Hyperlinks are Active)
How can error dampen out? With ABC, the costs are first segmented into activities and
then traced proportionately to reflect the diversity of the consumption effect of each cost
object placed on the activities. That is, all of the activity costs are reaggregated into the
cost objects after they have been segmented. Each activity's assignment to all of its
cost objects may have slight error, but each assignment is aimed at a bull's eye – some
of the assignment errors are plus and some are minus. They tend to cancel each other
out. In effect, the law of offsetting errors kicks in. Any error in undercosting or
overcosting has a canceling effect.
If you ask an accountant what he or she believes is the major source of accuracy in
ABC, resource drivers (e.g., employee timesheets) or activity cost drivers, the response
will usually be the former. Accountants tend to rationalize that if the data has errors
initially starting with the first inputs, then any error can only get worse from there on.
This is counterintuitive because as the activity costs are reaggregated into the cost
objects, the activity cost drivers more greatly influence the accuracy. To get better
accuracy, the accuracy of the assignments located closer to the cost objects should be
better than those farther away.
By using an existing ABC model, one can empirically test for the evidence of this
property simply by significantly changing the hours or percentages for several of the
employee timesheet input data (i.e., assume some employees deliberately falsified their
inputs). Recalculate the ABC model, comparing the new costs of products, standard
service lines, and customers to the old. On average, the dampening property of an ABC
model reveals that large changes in resource drivers result in relatively small changes in
the cost object costs.
The implications of error and accuracy are serious. Unfortunately, many organizations
that sorely need the insights and intelligence gained from activity-based information are
denied them because of the misguided beliefs and skepticism of a few people. Some
ABC systems fail because of accountants whose excessive desire for precision resulted
in an ABC model that was too massive, overdesigned, and exceedingly invasive on
employees' time and emotions. These ABC systems have been constructed with
designs that are far beyond the point of diminishing returns in accuracy. Too often in
building ABC systems, accountants do not work with the end in mind, but presume the
data collected must be very detailed and very precise everywhere.
Common Pitfalls in Developing an ABC Model (And How to Avoid Them), Part 1
ABC Versus Process Costing
Another common misconception is that ABC and process costing are the same thing.
Project teams may be led to believe that by linking the activity costs into the business
processes, accurate costs-to-serve customers are assured.
Knowledge Management Solutions, Inc.
January 2001
15
DRAFT
(Not all Hyperlinks are Active)
If project teams better understood the purpose for a project and its expected benefits,
they would probably attain better results with their activity accounting. One common
source of confusion is that some managers are looking for profitability analysis while
others desire process improvement. These are two different objectives. Although activity
accounting is needed to provide insights for both objectives, the orientation of the data
differs for each. Many ABC project teams do not appreciate this subtlety.
Both views of costs – the strategic view for understanding profits and the operational
view for understanding processes – require work activity data that is shared, common
data. These two views are needed because the general ledger or fund accounting
system is structurally unable to provide cost data that allows costs to be properly
additively flowed (i.e., process) or re-assigned (i.e., absorption).
The traditional ledger data focuses on transaction-intensive data that gets bucketed into
payroll or operating supply accounts; this provides little insight to work activities, and the
costs cannot be related to variable drivers. Another structural deficiency with the chartof-accounts is the vertical nature of the organizational chart and hierarchy, which uses
cost centers. Activity accounting resolves these weaknesses by translating the ledger
cost data into work activity information. It is at this point that the activity costs can move
in one of two directions, but not in both directions at once. On one hand, activities
belong to processes -- you can add their costs up across time. On the other hand,
activities are also consumed by products, service lines, channels, and customers. In
short, activity costs can be re-assigned.
This point often eludes the ABC project teams. If their objective is to provide profitability
analysis, then the driving force is to segment the diversity of the resource consumption
into the widely varying mix of products, standard service lines, and customers. There is
usually a broad mix of intermediate outputs of work which the products, standard
service lines, and customers themselves are consuming. This type of costing is ABC.
Alternatively, if the ABC team's objective is business process management, then the
attention shifts to understanding more about the activity costs themselves. This includes
not only what drives the activity costs and their individual levels of importance or
performance, but also how the activities interrelate across time to make up the business
process. For example, a rocket and a railroad train are both moving forward, but not in
the same direction, because one is moving vertically and the other horizontally. This
analogy applies to ABC and the process view.
Figures 2 and 3 illustrate the two different orientations – the cost object view and the
process view. In the cost object view, ABC concentrates on segmenting variation and
diversity about how the activities relate to the mix of outputs, products, standard service
lines, and customers (i.e., cost objects).
Knowledge Management Solutions, Inc.
January 2001
16
DRAFT
(Not all Hyperlinks are Active)
Figure 2: The Cost Object View
In contrast, the activity costs in the process view concentrate on how the activities relate
to each other in time, forming a sequence or network (a mix) traversing across
organizational boundaries. However, work activity costs in the business process relate
to each other in time.
Knowledge Management Solutions, Inc.
January 2001
17
DRAFT
(Not all Hyperlinks are Active)
Figure 3: The Process View
ABC is time-blind. Process costs are mix-blind. The work activity costs are shared and common
to both views as the starting point of their route for flowing costs. The activity costs are the
initial translation of the general ledger expenditures. After the work activities are costed, they are
then either added across time for the process view or disaggregated and reassigned with eventual
reaggregation into the products and standard service lines for the view of the mix of final cost
objects.
Common Pitfalls in Developing an ABC Model (And How to Avoid Them), Part 1
ABC project teams that have not fully thought through this distinction between the two
views may unconsciously blend both, thus adversely affecting both results. The
instinctive view of some ABC teams' is to string several activities in time sequence, then
interrupt that orientation by tracing the activities to their product, standard service line,
or customer cost object using a single cost driver. At that moment they effectively
abandon the much more sensitive diversity already captured in each activity cost and
revert back to an averaged allocation, which is contrary to the reason for implementing
ABC in the first place. Hence, the mix costs of products and standard service lines
become much less accurate than they would have been had each activity cost been
individually traced using its own unique cost driver. In the cost object view, activity costs
are each individually traced to their cost object based on diversity.
ABC project teams that initially want to focus primarily on the business process costs
may similarly misdesign their ABC models. These teams tend to interrupt their timebased cost flow structure to trace their partially sequenced costs into the unique mix of
Knowledge Management Solutions, Inc.
January 2001
18
DRAFT
(Not all Hyperlinks are Active)
outputs, products, standard service lines, channels, or customers. They never complete
the full end-to-end activity costs which belong to the business processes. With this
discontinuity, they cannot validly measure business process costs. Further, any form of
unused capacity analysis or constraint-based thinking is adversely affected because the
team has prematurely broken the process chain.
When confused ABC project teams build hybrid ABC models, their cost-flow design in
effect straddles both views, resulting in erroneous results. When the ABC team does not
declare whether its goal is to cost the diverse mix or to cost the end-to-end processes,
they split the difference and get neither. In short they get less accurate product and
customer costs, and they get incomplete process costs.
The lesson is this: with a single cost-flow network, one cannot compute the costs of
both the product and service line and the process. Each activity acts like a node in a
network. If one sequences activities in time, one cannot discern how the mix uniquely
consumes varying portions of each activity cost. Likewise, if one traces activities (from
potentially two or more processes) to the various cost objects, one cannot compute the
cost of each process. Users of commercial ABC software solve this problem with one of
two solutions:
1. They tag numerical attributes on their activities that cross-reference to each
activity's location on a process flow chart (e.g., in Microsoft's PowerPoint).
2. They export the activity data into a process flow chart drawing tool, into an
analytical process mapping tool, or into an even more powerful discrete-event
process simulation tool.
Level of Detail
Many ABC project teams believe that ABC systems require a “mud slide” of data, and
when the systems are complete, they are the world's largest spreadsheets. One of the
unanswered questions in the ABC industry is, “Why does it take so long for
organizations to progress from snapshot models to fully integrated ABC production
systems?” There are a variety of factors, including people's natural resistance to
change. A key factor, however, is the rate of organizational learning about ABC. If ABC
models are inadequately designed and constructed, then the learning and the buy-in, as
well as the time to produce usable information, will be slowed.
There are many challenges to overcome with an ABC project in order to successfully
implement ABC. Organizations do not need the additional headache of a poorly
designed ABC model. This is not an issue with the software itself, but rather what the
organization does with the software.
ABC information has properties that are useful for both strategic profitability analysis
and operational improvement, and ABC project teams often find it a challenge to design
a model for both uses. The models are often too large. In reality, ABC models can be
kept to a manageable size. The level of detail depends on the kinds of decisions made
Knowledge Management Solutions, Inc.
January 2001
19
DRAFT
(Not all Hyperlinks are Active)
with the new data, and the required accuracy. Figure 4 illustrates how the ABC cross
has an additional dimension of detail and disaggregation.
Figure 4: The ABC Cross
ABC project teams that understand that less error resides in the resource drivers than
the activity cost drivers will substantially compress the size of their ABC models. In
particular, they will reduce their data collection efforts. Instead of collecting time-effort
data from every employee (and in excessive detail), they will seek to collect data from
the minimum number of employee representatives. This means that ABC models can
be sized right, involve a minimum effort, and still provide very relevant and reasonably
accurate results.
ABC models should not be too elaborate or excessively detailed. Pareto's 80/20 law of
diminishing returns is heavily deployed in the design and construction of ABC models.
The amount of data and level of detail should meet the decision maker's needs. If the
model is too simple, it will not be sufficiently accurate; but if it is too complicated, the
extra administrative effort and cost to maintain it may exceed the benefits. It is important
to understand the trade-offs between the relevance, significance, accuracy, and
flexibility required to capture the diversity of resource consumption at timely intervals.
ABC’s Multi-stage Cost Assignment Network
In the early experimental stages of designing and building ABC pilot models, where the
term pilot implies the ABC model is intended to become a repeatable ABC reporting
system, many organizations followed the popular articles written for popular
management accounting magazines. These articles described ABC as a “two-stage
Knowledge Management Solutions, Inc.
January 2001
20
DRAFT
(Not all Hyperlinks are Active)
allocation” procedure. This also followed the popular ABC cross symbol used in ABC
training courses. The costing procedure is as follows: Resources such as people,
machines, materials, and supplies are assigned to work activities, and then each work
activity is reassigned to its product, standard service line, or customer based on its
unique relationship with an activity cost driver.
There are two major differences from traditional cost allocations: (1) department costs
are disaggregated more finely into the work activities to allow for better segmenting, and
(2) multiple activity cost drivers are used in contrast to a single total volume-based
driver, based on an input (e.g., labor hours) or on a final output (e.g., number of
completed units) measure.
As ABC systems became larger and more complex, the ABC/M data is applied to solve
broader problems. For these solutions, the ABC cost calculation usually required more
than the simple so-called “two-stage” cost re-assignment as indicated by the cost
assignment view of the ABC/M Cross. Figure 5 illustrates the expansion of the cost
assignment network from two-stages to a multi-stage network.
Figure 5: Simple vs. Expanded ABC
The expanded ABC Model includes intermediate stages of activities – i.e., activity
outputs that are inputs to successive work activities. Specific usage, not time-based
sequence, is the dominant factor for determining this cost assignment structure. These
intermediate input/outputs cannot easily be traced directly to final cost objects (i.e.,
products, service lines, types of channels, or customers) since there is no causal
relationship. As a result, intermediate activities are two or more stages removed from a
final cost object.
Knowledge Management Solutions, Inc.
January 2001
21
DRAFT
(Not all Hyperlinks are Active)
The need for multi-stage cost assignments, in contrast with the earlier use of a simplistic
two-stage assignment, has been due in part to assure accurate costing. A substantial
and material amount of organizational work activities support the more primary activities
that are in closer proximity to products and customer services. Organizations often refer
to this support related work as overhead. These support-related activity costs raise the
question, “how much of this activity is consumed by specific products or service-lines?”
But it is virtually impossible to answer this question because the work is simply too
indirect and remote from the products to detect or sense any differences. However,
these support activities can be traced in proper proportions to other activities that
require their work. Such support-related activity costs are eventually burdened into the
primary activity costs. These intermediary activities support the work activities that do
detect the variation and diversity of the products or service line. In sum, the ABC/M
uses multi-stages to trace all of the costs through a network of cost assignments into
the final cost objects.
In summary, many activity cost relationships are in reality indirect with respect to the
organization’s resources and final cost objects. As a consequence, the network-like
structure of the expanded ABC Model cost assignment system exists because of the
complexity in tracing the consumption of resource expenses to their final product or end
customer.
Figure 6: Expanded ABC Cost Assignment Network
Figure 6 shows a generic expanded ABC Model cost assignment network diagram. The
Knowledge Management Solutions, Inc.
January 2001
22
DRAFT
(Not all Hyperlinks are Active)
main differences between this diagram and the two-step ABC Cross are the presence:
(1) of intermediate stages of activity-to-activity cost assignments, and (2) of cost objectto-cost object assignments.
It is helpful to imagine the cost assignment paths of the generic expanded ABC Model
of Figure 6 as pipes and straws where each path’s diameter reflects the amount of cost
flowing. The power of the expanded ABC Model is the cost assignment paths and
destinations provide traceability from beginning to end – from resource expenditures to
each type of (or each specific) customer – the origin for all costs. For some readers it
may be useful to mentally and visually reverse all the arrowheads in Figure 6 to the
opposite direction. This polar switch reveals that all costs originate with a demand-pull
from customers – and the costs simply measure the effect.
Some people are initially intimidated by thinking about ABC as an arterial cost
assignment as presented in Figure 6; however, it makes logical sense the more you
work with ABC/M. Also, the ABC/M cost assignment network is related to an
observation that has become known as Metcalf’s Law:
The value of a network increases as the number of nodes increase.
In short, the majority of the support work activities could not be directly traced to
products and standard service lines, but they could be traced to the work activities that
consumed them. Now the cost accountant had the same step-down cost allocation
design, but at a much deeper level. Cost accountants were reluctant to use allocation
basis, which had little or no cause and effect, resulting in very large spreadsheets. In
some cases, some of the product-making and service-delivering work activities were
directly consuming (i.e., being allocated from) a support work activity and also indirectly
consuming that same work activity via another direct activity. This is because the
support activity we’re describing was also feeding another support activity which was
allocated, to the product-making or service-delivering activity. In other words, the
assignments were not simply columns-to-rows, but were coming from multiple levels.
One of the solutions was to begin employing the “macros” functionality of spreadsheet
software, making costing solutions much more like computer programming. The real
issue is that the ABC cost reassignment is actually shaped as an arterial network, as
shown in Figure 6 . The solution is either to have the organization's information systems
department develop a custom costing system, or to use commercial ABC software
products that have been designed as assignment networks and that include other ABC
functionality.
Part 2 of the Three parts Currently Available from Gary Cokins on the Web
When ABC implementation projects fall short of an organization’s expectations, it rarely
has anything to do with the ABC methodology or technology that now supports ABC.
Many failures result from misconceptions by the ABC project team. Regardless of the
problem, however, almost all of the potential pitfalls must be considered and addressed
Knowledge Management Solutions, Inc.
January 2001
23
DRAFT
(Not all Hyperlinks are Active)
to achieve the minimal threshold level for success where users can continue to reuse
the data and request updates from period to period.
Five areas expose an ABC project to the risk of failure:
1.
2.
3.
4.
5.
ABC model structure design and architecture
Selection and collection of driver data
Implementation and data integration
Getting management and employee support
Applying the data to decision making
These five categories of risk must be addressed in a balanced way. That is, it is more
important to pay some attention to all areas than to overemphasize one or two while
ignoring the others.
This article is part two of a five part series and will address the second category of risk,
selection and collection of driver and cost object data data.
Problems Related to the Selection and Collection of Driver and Cost Object Data
In a short phrase, poor ABC model design will lead to poor results. Since there are so
many other behavioral change management obstacles to be overcome with ABC,
having a sub-par model design, with poor or excessively costly driver data to collect,
makes the task of overcoming organizational resistance to change only that more
difficult.
Reusing General Ledger Descriptions
It is a misconception to believe that many of the activity descriptions in the activity
dictionary, such as “maintenance costs,” can be easily applied to ABC using the
identical descriptions that already appear in the general ledger chart of accounts. The
wording in the general ledger (or fund accounting system for public sector
organizations) is alien, confusing, and unfamiliar to most employees and managers.
This language is outside their comfort zones.
The best way to define activities is to use the grammar convention of an “action verbadjective-noun.” For example, in the maintenance department example above, activities
could be worded as “repair broken equipment” or “perform preventive maintenance.”
ABC has immeasurably more impact if employees and managers can relate to what
activity costs really mean. Activities should be expressed in words that help employees
to visualize people or machines doing the work. For example, the activity “invoicing” is
hardly as effective as “computing and generating customer bills.” People relate much
better to this kind of language; there is a greater feeling that they can effect changes to
the work. Also, it is much easier to score or grade ABC's attributes when activities are
expressed in words with action verbs and nouns.
Knowledge Management Solutions, Inc.
January 2001
24
DRAFT
(Not all Hyperlinks are Active)
Choosing Activity Drivers
Activity drivers should reflect the definition in the CAM-I Dictionary of Terms: “[A]n
activity driver measures the frequency, intensity or quantity that drives the time and cost
of an activity.” In other words, an activity driver should be a measure that is nearly
identical to the measures of the output of the activity or business process it is part of.
Cost drivers and activity drivers serve different purposes. Activity drivers are output
measures that reflect the usage of each work activity, and they must be quantitatively
measurable. An activity driver, which relates a work activity to cost objects, “meters-out”
the work activity based on the unique diversity and variation of the cost objects that are
consuming the activity. It is often difficult to understand whether use of the term activity
driver is related to a causal effect (input driver, such as “number of labor hours”) or to
the output produced by an activity (output driver, such as “number of invoices
processed” or “number of gallons produced”). In many cases, this is not a critical issue
as long as the activity driver traces the relative proportion of the activity cost to its cost
objects.
Older, less effective terms, such as first and second stage driver, continue to be used to
describe items similar to the currently more accepted terms “resource driver” and
“activity driver”.
A cost driver is a driver of a higher order than activity drivers. One cost driver can affect
multiple activities. A cost driver need not be measurable but can simply be described as
a triggering event. The term cost driver describes the larger scale causal event that
influences the frequency, intensity, or magnitude of a workload, and therefore,
influences the amount of work done that translates to the cost of the activities. As an
example, a sales promotion can be a cost driver for substantial increases in the
company’s work activities of the order fulfillment process. The amount of effort taking
orders, for example segmented by teenagers versus senior citizens, would require an
activity driver (i.e., number of orders placed due to promotion) to calculate the
proportional costs to customers in each segment. There would be a unique activity
driver for each work activity that belongs to the order fulfillment process.
Using the familiar ABC/M cross diagram, Figure 1 illustrates how activity drivers have
higher order cost drivers. The figure illustrates how activity drivers are lower order
drivers of cost drivers. ABC/M relies on activity drivers for tracing costs, and collectively
they are useful when combined with quality management (QM) problem solving tools for
identifying root causes.
For example, if an activity residing in an insurance company is “spot-check complex
claims,” the activity cost driver should be the “number of complex claims spot-checked.”
This way the average unit cost for each spot-checked complex claim can be trendreported over time, and more important, the particular client or customers that are
associated with these kind of spot-checks can bear the burden of the extra cost for that
Knowledge Management Solutions, Inc.
January 2001
25
DRAFT
(Not all Hyperlinks are Active)
work. The activity cost can be metered out to its cost objects in proportion to the
activity’s use.
Clearly there is a higher-order cause that has led to the spot-checking, such as
conflicting procedures; but understanding that connection should take place as part of
the ABC activity analysis. Problem resolution, if warranted, should follow the activity
analysis.
One of the reasons that employees are trained in total quality management (TQM)
analytical techniques, like Taguchi methods and fishbone diagrams, is so they can
apply their learning to the use of facts and measures. TQM training explains cause-andeffect relationships; that is, costs fundamentally are measures of effects. Employees
must master the inferences that can be gained by understanding the first-order activity
cost drivers (i.e., output drivers) even though they may have to test their hypothesis
using second- and third-order inferences.
Activity drivers should be closer to the symptom than to the root cause. When ABC
models are constructed this way, the calculated final cost objects are more accurate,
and there are less expectations that the ABC system is going to be the magic pill that
provides all of the answers.
An organization should consider many factors in addition to costs (and profits) as it
assesses what are better ways to improve performance. Figure 2 shows a staircase for
how raw data is converted into managerial information and used for decision making
and taking actions. A good question is where is the epicenter and most central location
for where the ABC/M data is on that staircase of data progression into actions?
Figure 2 shows a reasonable answer. The ABC/M data only provides findings and
observations. From this, the organization can discover more things that it never knew
and ask more questions and test hypotheses. Perhaps it may need to collect additional
or more granular ABC/M data to answer those questions, but the key point is that the
ABC/M data is serving as a stimulant to get people thinking about what options the
organization might take. Later on, and further up the staircase, ABC/M data can also be
used to test or compare the outcomes of the alternative actions, but ABC/M data is
initially used to aid in learning through discovery.
The message here is that although making good decisions is an ultimate goal, the
ABC/M data is strictly an enabler to be combined with other factors to attain that goal.
Managers must understand what the ABC reports, findings, and observations mean and
analysts must use the ABC data to test hypotheses and eventually reach conclusions as
shown in Figure 2. Only then should alternative options be formulated.
Keeping the Final Cost Objects Simple
Knowledge Management Solutions, Inc.
January 2001
26
DRAFT
(Not all Hyperlinks are Active)
The trickiest part of designing an ABC cost-flow assignment network is in the final cost
object module. All activity costs should ultimately reside in the two places where they
actually originate:
1. Customers or prospects and service-recipients
2. Business-sustaining cost receivers
There can be an entire chain of final cost objects that precedes the customers and
service-recipients. That is, the customers and service-recipients consume final cost
objects, such as products or service-lines, that have also already consumed activity
costs.
Figure 3 illustrates the ABC/M cost assignment network. Focus on the bottom module,
the final cost objects. Note how various final cost objects consume the other final cost
objects. The metaphor for this cost consumption chain is the “predator food chain” of the
animal kingdom where large mammals eat small mammals and small mammals eat
plants.
The left-most final cost object represents the model’s suppliers. The unique work activity
costs caused by one’s suppliers, such as processing their purchase orders or
negotiating deals, are attributed to those products that are purchased. The National
Association of Purchasing Management (NAPM) refers to this as the “total cost of
ownership (TCO).” This means the invoice price of the purchase does not reflect the
entire cost of procuring that product. As an example, just think about the differences
between technically sophisticated suppliers who use EDI, e-commerce, and bar-coding
in contrast to archaic suppliers who use error-causing faxes. Which type of supplier
causes more of a company’s workload and costs – apart from the direct material
purchase cost? Suppliers cause their customers different workloads independent of
volume.
The final-final cost object, which in Figure 3 is the customer, ultimately consumes all of
the other final cost object costs, except for the business sustaining costs. In effect, what
ABC/M does, as modeled in the cost assignment network, is reflect how the variation
and diversity of cost objects segment activity costs that have drawn on resource
expenses.
As an example, if one is determining the various costs on a military base of providing
civilian and military services, the final cost object is all of the soldiers. The soldiers are
the reason the cost structure exists. Some soldiers will be low-cost and others highcost, some soldiers sleep in old high-maintenance barracks, some are trained for army
tanks, which in turn require cost maintenance and parts. The possibilities for identifying
examples of diversity are endless, but that is the challenge for the ABC model
designers, who need to identify enough of the right kinds of diversity to make the
model's size manageable yet still relevant for decision making. In other words, the work
activity costs must first be traced via causality to the person or item causing the activity
cost to fluctuate, such as an army barracks. After all the activity costs are assigned to
Knowledge Management Solutions, Inc.
January 2001
27
DRAFT
(Not all Hyperlinks are Active)
their final cost objects, then the cost objects themselves should be logically attributed to
whatever is consuming them.
If there are substantial costs and sufficient diversity in another type of cost object, for
example the type of customer order (standard orders, specials, adjusted, international,
etc.), then the “order type” can qualify as its own separate and visible final cost object.
Another example might be the type of freight-haul trip, such as truck, marine or rail or as
less-than-truck load (LTL) versus full truckload. This type of final cost object would
serve as an intermediate repository to capture diversity of the type of work output.
After activity costs are traced into all of these final cost objects, then those costs are
further re-traced to the customers based on the mix of order-types consumed by each
customer. Hence, “All customers are not created equal.” ABC/M equitably traced all the
costs based on unique usage by its downstream final cost object.
Figure 4 displays three potential cost object types that could be isolated and assigned to
as an intermediate destination for activity cost accumulation prior to being re-assigned
to customers.
Note that without individual final cost object “types” being isolated, the activity costs that
trace into them would have been directly assigned to customers from the same activity
costs. The factor or basis (i.e., activity driver) would not likely be reflective of the
activity’s use; in fact, the activity would likely be pooled with other activities and
therefore be traced with the activity driver chosen to trace the more prominent activity
cost.
Improved accuracy comes from establishing an individual final cost object and thus
isolating each of them. Then, using a two-step cost assignment method, the activity
costs trace the workload to this type of cost object, and then the customer is shown to
be “purchasing” and consuming the output. The second of the cost assignments are
referred to using ABC/M lingo as cost object drivers (the term “activity driver” is no
longer applicable as the work activity cost was already accumulated in the final cost
object.)
For advanced ABC/M users, they may wish to view product profitability including
customer costs (e.g., to determine and publish prices in their price list catalog). Today’s
advanced ABC/M software allows multidimensional views of various combinations of
cost objects. With multidimensionality, a two-way bi-directional linkage replaces the
sequence of the predator food chain. Other dimensions can include geographical sales
territories, store locations, or specific salespeople. Multidimensionality is made possible
by computer technology called on line analytical processing (OLAP).
In summary, calculating costs with ABC/M allows re-assembly and assignment-tracing
for all the work activity costs to reflect how each customer, channel and market segment
consume the costs.
Knowledge Management Solutions, Inc.
January 2001
28
DRAFT
(Not all Hyperlinks are Active)
Approach to Full Absorption Costing
Managers often assume that eventually all costs must be fully recovered by service
lines, including customer or service-recipient related costs, distribution and warehousing
costs, selling costs, and all of the general overhead costs. Two issues arise from this
assumption. The first is whether standard service lines consume customers or vice
versa. The second concerns the traditional debates about full absorption costing (the full
cost recovery concept).
Many managers are accustomed to financial information that traditionally reports
product and service line costs. They are so familiar with this reporting format that their
minds are programmed to believe that the products and service lines are somehow
responsible for all of the work activities, including those such as making sales calls or
responding to expedited requests. However, these activity costs are somewhat
independent of the standard service line. In reality, each customer is consuming its own
unique mix of services (which have their own costs), and those costs can then be
combined via ABC with the unique costs-to-serve for each customer.
Figure 5 contrasts a traditional profit and loss statement with one produced for an ABC
system. It reveals that ABC profit margin contributions are layered like an onion skin.
Price covers all of the costs, but the sequence can be layered to allow for comparisons
of profit contributions at each layer.
Unprofitable customers can purchase some profitable and products and standard
service lines. Likewise, profitable customers can be purchasing some unprofitable
products and standard service lines. Profitability analysis, which may suggest what
actions to take, requires additional thought because of this product/service line and
customer profit duality. One solution is to provide ABC users with multidimensional
analysis tools so they can examine their profit cube from a variety of angles, including
additional dimensions (e.g., by salesperson or by geographic region). Commercial
software products that perform this analysis are called on-line analytical processors or
OLAPs. Fundamentally, however, it is the customers that consume the products and
standard service lines and not the other way around.
Major portions of accounting and business textbooks are devoted to the subject of full
absorption costing. What ABC lends to this discussion is the ability to gather together
and summarize as a memo item (i.e., does not affect the costing) all of the activity costs
that are not caused by products, standard service lines, or customers. This includes
even the activity costs of the support work activities not directly required by their
downstream consuming activities. All activity costs are collectively gathered into the
broad final cost object category called “business or organizational sustaining.”
With traditional costing, which is usually done at the departmental level, many of these
costs are “baked” into the allocation schemes and in effect are inadvertently loaded into
the products, standard service lines, and customers, even though those costs are
actually caused by someone else. Examples of business sustaining cost objects are:
Knowledge Management Solutions, Inc.
January 2001
29
DRAFT
(Not all Hyperlinks are Active)
•
•
•
Senior management asking for reports and budgets,
The business itself, by requiring landscaping and grass-cutting, or
Government agencies, such as the EPA requiring certain testing activities or the
IRS causing tax filing work.
When all these costs are totaled, organizational sustaining costs can amount to a
substantial part of the enterprise's non-sunk costs. For example, when 40 percent of an
enterprise's total non-sunk costs (sunk costs, such as equipment depreciation, are a
different kind of issue) are not caused by products, standard service lines, or customers,
then management eventually has to ask itself, “How long can we afford this?” and “If our
competitors have a smaller percentage of these same costs than us, won't this
eventually provide them an advantage?”
Resource and Activity Driver Data
Organizations should not use resource and activity driver data simply because it exists
and may be more readily available than other data. Driver data should be selected
because they reflect a causal relationship.
Resource drivers are time-based for salary-related expenses. Non-salary-related
resource costs rely on volumetric driver assignments to activities.
The selection of transaction-based (i.e., measurable) activity drivers should be tightly
linked to materiality and skewness. By materiality, it is meant that collectively, they
should account for a substantial amount of total activity costs. By skewness, it is meant
that despite lower magnitude cost, for the products, service lines or customers that the
activity is being traced to, determine whether the activity is significantly concentrated in
just a few of those cost objects.
ABC can generate dramatically more accurate costs compared to traditional cost
allocation methods. There are, however, diminishing returns in incremental accuracy
from adding each additional cost driver, and there are administrative costs as well to
collect the driver data. It is important to always test whether the extra effort in data
collection will result in much better and more accurate information than the company
had before (without the extra data).
Figure 6 illustrates that not all of the activity driver data must directly come from feeder
systems. The vital few do, but the remainder can come from estimates from
knowledgeable functional representatives. Remember that since ABC/M always
normalizes driver quantities to sum to 100%, then the consequence of modest
estimating error is not that severe. Some cost objects are only slightly over-costed while
the remainder are under-costed. And then this error offsets as multiple activity driver
costs combine.
In general, one should not collect new any data that is not already being collected for
some other purposes. Try not to give naysayers an excuse to criticize ABC/M as forcing
Knowledge Management Solutions, Inc.
January 2001
30
DRAFT
(Not all Hyperlinks are Active)
new data to be collected; re-use existing data. In addition, always apply ABC/M’s
important test question, “Is the climb worth the view?”
Common Pitfalls for ABC Initiatives (And How to Avoid Them), Part 3
This article is the third in a five-part series dealing with the most common
mistakes organizations make when undertaking activity-based costing projects.
When ABC implementation projects fall short of an organization’s expectations, it rarely
has anything to do with the ABC methodology or technology that now supports ABC.
Many failures result from misconceptions by the ABC project team. Regardless of the
problem, however, almost all of the potential pitfalls must be considered and addressed
to achieve the minimal threshold level for success where users can continue to reuse
the data and request updates from period to period.
Five areas expose an ABC project to the risk of failure:
1.
2.
3.
4.
5.
ABC model structure design and architecture
Selection and collection of driver data
Implementation and data integration
Getting management and employee support
Applying the data to decision making
These five categories of risk must be addressed in a balanced way. That is, it is more
important to pay some attention to all areas than to overemphasize one or two while
ignoring the others.
This article is part three of a five part series and will address the third category of risk,
implementation and data integration.
Problems Related to Implementation and Data Integration
A good ABC/M cost assignment structure and good cost driver selection is not enough
to assure success. Even if you have built the most elegant, properly-leveled and rightsized ABC/M model, it provides little insurance that there will be widespread acceptance
and use of the ABC/M data. As Robin Cooper, one of the early luminaries in ABC/M,
would frequently say, do not have a “Field of Dreams” ABC/M project. He was referring
to the popular movie about baseball where a voice from the sky bellowed, “If you build
it, they will come.” In the movie the they referred to deceased players of a 1919 baseball
team, the Chicago “Black” Sox, who were famous for their past behavior. With ABC/M,
the they is the targeted end-users of the ABC/M data. With ABC/M, although you may
build a very elegant model, managers and employees may still say “So what?” after
they receive the data!
In short, there is a lot more to being successful than simply properly designing and
constructing a practical ABC/M model. The issues related to implementation and data
Knowledge Management Solutions, Inc.
January 2001
31
DRAFT
(Not all Hyperlinks are Active)
integration deal with the misconceptions that ABC/M project teams frequently have
when they are beginning their implementations. Implementation and data integration
issues address areas such as selecting the right ABC/M pilot or testing place,
expanding the ABC/M model, and linking the ABC/M model to data from feeder systems
that are used.
[1] Does ABC/M Mean You Must Maintain Two Sets of Accounting Books?
A commonly expressed fear is that since an ABC/M system is initially constructed as a
standalone and one-off accounting system parallel to the official financial reporting
system, then this will make for double the work. It is not a serious problem.
Here is why.
Some organizations can get stymied when their environment has a project accounting
system, and/or a standard cost system, and/or a job cost accounting system. Where
does the ABC/M system fit in? Does it replace any or all of the others?
Before jumping into the discussion to determine which cost accounting approach makes
the most sense, it is probably good to step back a bit and ask a more fundamental
question. Aren’t all of these accounting approaches basically transforming expenses
into calculated costs? By definition, expenses are what an organization incurs when
they exchange their money with suppliers, including employees. When these expenses
are traced and assigned into activity costs or further into product-related or service-linerelated costs, then some assumed factors or measures were used to calculate those
costs. The assignment of expenses into costs is formally referred to as absorption
accounting.
It is simple and straightforward to assign direct expenses, such as material that is
dedicated to a specific product. The controversies and difficult assignments are with
indirect expenses. That is where ABC/M has become a popular refinement of traditional
absorption accounting.
ABC/M’s power lies less in its use as an operating control tool (e.g., variance reporting),
but rather more in its use in assessing what things actually cost, without misleading
allocations. (More recently, ABC/M is becoming popular with “predictive planning” such
as activity-based budgeting.) In order to determine what are the true costs of outputs,
products, standard services and customers, one can use an overarching cost
framework to visualize the ultimate goal of an enterprise-wide direct costing.
Figure 3.1 illustrates the four individual cost assignment approaches that were referred
to above.
The selection of which costing approach an organization should use depends on the
following conditions:
Knowledge Management Solutions, Inc.
January 2001
32
DRAFT
(Not all Hyperlinks are Active)
•
If there is some project work occurring, you clearly would use project accounting.
Project accounting (also called job order costing) can be superior to ABC/M
because it assigns the portion of work that is non-recurring and attaches it to the
planned work-step in the project schedule. ABC/M takes over for all of the
remaining recurring and repetitive work. ABC/M relies on the resource and
activity drivers plus the cost assignment network structure to trace resource
expense into costs.
•
If a pre-existing standard cost system is in place for direct costs, such as material
and repetitive labor (e.g., a bank teller, a machine operator), then these costs get
combined with ABC/M’s treatment of all the remaining non-direct costs. (ABC/M
has been referred to as a refined standard cost system. The refinement is that
with ABC/M, the standards or rates are frequently revised for each time period
that the model is refreshed.)
The consequence of not fully recognizing that multiple approaches of costing can and
should be blended and harmonized, as opposed to having to compete amongst each
other, means that end-users and decision-makers are denied a seamless
representation and image of their organization’s cost structure and economic behavior.
This denial puts the organization at risk because poor decisions may result.
[2] Scaleable ABC/M Models
Many first-timers to ABC/M believe that an ABC/M model must be designed very
detailed, deep, and wide so the results can be summarized in case any user eventually
ever needs an answer to any of their remote requests.
Many ABC/M project teams do not even consider calculating costs at any level other
than at the lowest and most detailed. When those teams decide that in the initial stages
they should restrict their calculated costs for only product families and/or service line,
they are not really clear about what will happen to the ABC/M model’s structure as they
expand it and drive to lower levels of detail. The simple answer is nothing will happen.
ABC/M costing is “renormalizable” – that is, it holds and maintains its proportions as it
size inflates. The summary costs do not greatly shift; only the detail gets for more finely
segmented as both the activities and cost objects are disaggregated. There is one
exception however. As the activity costs themselves are disaggregated further and
deeper, some of the work activities are no longer linked with a cause-and-effect
relationship to the products, standard service-lines and customers, which are the
primary reason and organization exists. Increasingly more activities and their costs are
traced to the business-sustaining activities.
For example, consider the work activity, developing and reviewing the annual budget.
This work is not directly caused by products, standard services or customers. The
primary driver is senior management and their policy of controlling spending by using an
annual budget. Senior management is the recipient of budgets and hence the cause of
Knowledge Management Solutions, Inc.
January 2001
33
DRAFT
(Not all Hyperlinks are Active)
that cost. In organizations that do not produce or use budgets, employees spend zero
time and activity costs on budgeting, but the effort to make products and service
customers is unaffected.
The implication here is that as one further disaggregates the activity dictionary, work
activity costs shift away from customers and towards the business-sustaining cost
objects. This means that products, service-lines, and customers are somewhat overcosted with a higher level activity dictionary, and conversely business-sustaining costs
are under-costed. However this shifting rapidly settles out as the ABC/M model is
scaled-up and expanded.
[3] ABC/M is Not a Mudslide of Data Collection
People who are unfamiliar with how ABC/M actually works have a false impression that
it takes forever and a day to collect the ABC/M data, and it is a cumbersome exercise.
Worse yet, they believe all of their imagined mountain of data is absolutely required in
order to have accuracy. They are in for a surprise.
It is counter-intuitive, but precision input data is not a pre-requisite for accurate outputs.
The ABC/M assignment network can tolerate and, in fact, is somewhat resilient to
estimating errors. Because error dampens out through offsets and does not compound,
ABC/M allows for dramatic reductions in the data collection effort, which reduces
administrative cost.
It is true that the level of detail and accuracy depends on what decisions the
organization is going to make with the data. However, many decisions can tolerate a
predetermined acceptable plus-or-minus percentage of error. ABC/M rapid prototyping,
now heralded as a superior method to successfully implement ABC/M, leverages the
properties described here.
The following are three key areas that involve data collection:
1. The financial data is rarely an issue with regard to accuracy because the source
information is usually directly transferred from the organization’s general ledger
accounting system. These are the general ledger’s chart-of-account expense
balances. Occasionally alternative but similar data sources, such as payroll
spending, are used. But these too are usually very exact.
2. The resource cost driver data falls into two broad groups: wages plus fringe
benefit related expenses and non-wage related expenses. Freed from requiring
excessive precision, ABC/M can use a much smaller group of reasonably
informed employees to serve as “functional representatives” to estimate the labor
time-effort (in hours or percentages) for all of the organization’s employees.
Tracing non-wage expenses is considerably easier after the work activities
dictionary has been defined and leveled. This is because non-wage related
expenses tend to follow or accompany the work, such as sales travel goes along
Knowledge Management Solutions, Inc.
January 2001
34
DRAFT
(Not all Hyperlinks are Active)
with the wage related work activity, “visit customer locations.” The same
“functional representatives” can reasonably assign these splits too.
3. The activity driver cost data are not needed for all the drivers, but only the vital
few. Those vital few are characterized by the activity drivers that: (1) comprise a
substantial portion of the spending (the 80/20 rule), (2) have pronounced skewing
towards concentrations of the cost objects (because a small amount could make
a big impact to those particular cost objects), and (3) experience volatility from
period to period in the amount of the driver cost consumed amongst its cost
objects.
For those activity cost drivers not falling into these three criteria, which could be the 80
percent from the 80/20 rule, the ABC/M project team can again select employees who
can estimate the driver magnitudes from their memory without requiring hard data from
feeder systems or manual logs. These employees can estimate the quantities of the
cost drivers, such as the number of customer sales visits and split that quantity amongst
customers, or simply use percentages. With ABC/M, the quantities are always
normalized into percentages for the cost re-assignment. Quantities, actual or estimated,
are useful for calculating unit-of-work output costs, such as the cost per each sales visit,
which can be trended over time to determine if they are headed in a favorable or
unfavorable direction.
Figure 3.2 shows that it is not necessary to collect transaction-based driver data for
every cost assignment in the ABC/M model. In fact, it is quite the other extreme. As long
as a vital few measured drivers are imported from operating systems, based on a
combination of materiality and the diversity-skewness of their cost objects, then the
remaining cost driver assignments can simply be input as estimates by key employees
who are familiar with that portion of the ABC/M model is mirroring the organization.
These estimates can also be in the form of surrogates, which may be easier to measure
than the underlying driver data, but they will still directionally parallel the driver. A key
rule is to stay away from using traditional allocations where there is little or no causeand-effect relationship.
After gaining experience by performing sensitivity analysis with regard to error on the
existing ABC/M model, one can determine which employee-estimated driver quantities
become the best candidates to become the next transaction-fed driver. This approach of
data collection is in stark contrast to those who want to collect any and all driver data
just because it exists rather than because it is relevant or adds appreciably greater
accuracy relative to the higher costs to collect that data.
The implications from knowing the properties of ABC/M models with regard to error and
accuracy can be critical. Equipped with these understandings, one does not have to
initially define and construct a massively large ABC/M system. Since ABC/M models are
scaleable, increases in detail will not introduce any significant distortion. The implication
is that one can quickly build an ABC/M model with activities and product, service-line or
customer families at a more aggregated level. Much of the input data can come more
Knowledge Management Solutions, Inc.
January 2001
35
DRAFT
(Not all Hyperlinks are Active)
from estimates from knowledgeable employees than from databases. The results will be
a good first cut at calculating reasonably accurate costs. Given this ABC/M model, it is
now substantially easier to modify by iterations and adjust the successive ABC/M
models based on the relevance to the uses of the data and the accuracy requirements
of those uses.
ABC/M’s substantially improved accuracy relative to traditional approaches actually
resides more in the cost assignment network than in the cost driver quantities. That is,
the reason the products, standard services and customer costs from ABC/M are so
reasonably accurate has less to do with their input data than with the architecture of the
cost flow paths comprising the network. Why? In the network, the activities are being
directly costed only to those (also more granular) cost objects that are consuming them,
and not to those which are not. The assignment paths themselves are more important
than the data.
[4] Choosing the Location for the ABC/M Rapid Prototype or Pilot
Some ABC/M project managers (or their senior executives) believe they should start
with the highest impact area to pilot ABC/M, and really show management good results.
Some take the opposite extreme and believe they should start in a fairly simple area so
you don’t mess up the ABC/M implementation. The best starting point for an ABC/M
pilot may be somewhere in-between. A variant of the ABC/M pilot question is, “Should
we start with only a few departments or go all the way to a tops-down enterprise-wide
view of our cost structure?”
There are two separate dimensions to consider the ramifications when selecting an
ABC/M pilot:
•
•
Scope: (1) enterprise-wide (tops down, entire organization) vs. (2) a
departmental project
Duration: (1) rapid (completed in days) vs. (2) lengthy (performed over four to six
months)
The experiences and results for each combination are different. Organizations that are
just starting out with ABC/M should begin their first five to ten days differently than with
the lengthy traditional multi-month ABC/M work plan. After experiencing rapid ABC/M
model development (using the ABC/M rapid prototyping method), the ABC/M project
team can then follow a traditional plan if they choose to, even if that requires doing an
ABC/M pilot.
Increasingly, however, more consulting firms who assist in ABC/M implementations
advocate starting an ABC/M roll-out with an approach that is stealth-like and noninvasive to employees, as opposed to a formally announced "banners-and-bugles"
approach. A risk from the latter approach is it can create false expectations by
employees that then create obstacles later. The organization should start by exposing a
relatively small number of employees to ABC/M, but the exposure should be intense –
Knowledge Management Solutions, Inc.
January 2001
36
DRAFT
(Not all Hyperlinks are Active)
they learn by doing. ABC/M rapid prototyping is similar to immersion learning of a
foreign language.
The benefit is that the participants can then advocate ABC/M to their co-workers, while
revising and customizing the ABC/M roll-out work plan based on their learning.
[5] ABC/M’s Achilles Heel: The “Leveling” Problem
“Leveling” is a major problem with ABC/M – that is, determining the right level of detail
for work activities, products, standard service lines, channels and customers. Too often,
ABC/M teams improperly build their first pilot too deep and too detailed without realizing
it. They over-engineer it and thus do not get the results they expected. Traditional
information systems development methods, that schedule the first few months to
perform the "requirements definition phase," are not adequate for designing the ABC/M
architecture because there is too much interplay between definitions, levels of detail,
precision inputs and accurate outputs.
As a result, an acceptable and perhaps more appropriate way to accelerate an ABC/M
project team’s learning about ABC/M is to very quickly ( in days, not months ) build a
high level ABC/M model of the entire organization (enterprise-wide) using commercial
ABC/M modeling software, such as ABC Technologies market-leading Oros. You build it
and then "adjust" it afterwards. Because ABC/M models are scaleable, the local ABC/M
project team can then much better revise and modify the originally proposed work plan.
The idea is for the team to make its mistakes up-front and early, when it is easier to
change the ABC/M system, not later when it is more difficult.
The objective of an ABC/M rapid prototyping pilot process is to help employees quickly
gain a working knowledge of ABC/M. ABC/M rapid prototyping provides accelerated
learning by rapidly immersing the ABC/M project team and a few key employees by
using ABC/M. They will answer their own questions, possibly with assistance from an
ABC/M management consultant or internal coach, based on their own real experience.
The ultimate goal of an ABC/M rapid prototyping pilot is beyond demonstrating a “proof
of ABC/M concept.” The exercise produces real results, including hard data, that can be
acted upon. The ABC/M project team can subsequently and promptly follow the ABC/M
rapid prototype model with iterative re-modeling; the ABC/M model’s shape remains
constant, but the model expands to greater depth, granularity and accuracy.
This “practice round” ABC/M rapid prototype pilot requires only about three days. The
ABC/M rapid prototype model’s data will repeatedly be replaced with higher-grade data,
but the model’s shape, geometry and cost drivers will remain reasonably solid and
relatively constant.
The next phase of ABC/M rapid prototyping is repeating the prototype model while
maintaining the same constant scope, amount of money, and headcount included in the
original model. This next round of modeling should expand on the original model in
Knowledge Management Solutions, Inc.
January 2001
37
DRAFT
(Not all Hyperlinks are Active)
terms of scale, detail, granularity and credibility. This next iteration involves collecting
data from additional key employees acting as functional representatives for the entire
operations.
It is conceivable that the initial ABC/M rapid prototype model will attain sufficient
accuracy as to answer immediate questions related to problems and issues of the
organization. To the degree that it falls short, the ABC/M project team will be on their
own capable of refining the model until it is sufficiently accurate.
Some people have a misconception that ABC/M can only work well if the data is fully
integrated, such as with an ERP system. They believe that ABC/M is cumbersome if it is
interfaced from disconnected systems.” As illustrated in Figure 3.3 , one way to view the
information processing space is to consider three layers of technology as follows:
•
•
•
Reporting – The distribution tools (to the desktop) include on-line analytical
processing (OLAP) tools, data warehouses, and data mining.
Analytical – The “solution set” tools include shareholder value analysis (SVA),
performance measure weighted scorecards, activity based cost management
(ABC/M), and process modeling and simulation.
Transaction-Based – The mega-transaction engines of operational, ERP, and
financial ledger systems that are feeders to the middle layer.
ABC/M is located in the middle layer that is formally referred to as “analytical
applications” by information technologists. By definition, analytical applications
transform or draw on summary data from the transaction-intensive operational systems.
Analytical applications, such as ABC/M, are separate and apart from the transactional
systems.
ABC/M’s popularity is partly due to problems with using the General and Fund
Accounting data for decision-making. The Fund Accounting and General Ledger’s chartof-account scheme is now recognized as being structurally deficient for cost analysis
(other than for historical spending control against budget or fund account) because the
account balances do not reflect the variability of costs with activities. ABC/M’s chart of
activities, a listing of work activities worded in verb-adjective-noun grammar, resolves
that deficiency.
The transaction and large volume software vendors, such as the enterprise-wide
planning (ERP) vendors, design and wire their software code for speed. As a result,
their core competency has not been providing the middle summary layer so critical to
reporting and analysis.
The top layer – the on-line analytical processing (OLAP) tools – are mainly viewing
tools. The heavy lifting and data transformation is done with the analytical applications.
The OLAP tools provide the flexibility to display the data in multiple ways. The output of
ABC/M is excellent input to OLAP tools.
Knowledge Management Solutions, Inc.
January 2001
38
DRAFT
(Not all Hyperlinks are Active)
[7] The Level of Manpower to Implement ABC/M
Some people havea misconception that ABC/M is a huge system, deserving a huge
effort, which means a huge project team and budget to go with it. In practice, the
administrative costs of maintaining an ABC/M system can be dramatically lightened by
following the ABC/M modeling principles. Most organizations perceive ABC/M a major
undertaking; and even after the organization is fairly far along with their ABC/M, they
still do not realize how unnecessarily deep and detailed they have gotten. ABC/M does
not require a huge project team or budget.
Parts 4 and 5 of this article are not yet published on the Web.
Ref: www.bettermanagement.com
Knowledge Management Solutions, Inc.
January 2001
39
DRAFT
(Not all Hyperlinks are Active)
Examples and Applications
1.0 Activity-Based Costing--Getting to the Root of Cost Drivers in the
Warehouse
from Distribution Center Management, October 1997
By using Activity-Based Costing, distribution managers can identify unprofitable
warehouse practices, eliminate them, and implement process improvements in their
facilities.
ABC may seem like a fairly new concept, but it's not, according to Don Rice, professor
emeritus, Industrial Distribution Management at Texas A&M. Unlike typical costaccounting methods used by most companies, ABC identifies all the cost drivers, or the
activities involved in carrying a product and selling it to a customer. These activities can
include: generating a purchase order, creating special kitted orders, etc.
"ABC is very work-centric, whereas a company's general ledger is very transactioncentric," further explains Gary Cokins, Director of Industry Relations for ABC
Technologies Inc. (ABC Technologies, based in Beaverton, OR, is a provider of activitybased information systems.) Cokins notes that "although both have their place, general
ledger data is too raw to be considered business intelligence for decision support. But
ABC does more than just translating the ledger account expenses into their work
activities. It links the activities into networks to cost out items for analysis, trade-offs and
decisions."
Rice observes that while companies are often willing to provide their customers with
many services, they remain unaware of the actual costs involved. For example,
accepting returned items can impact an entire organization's bottomline, from
warehousing to billing. Yet, it may go unfactored into the pricing of a product for a
customer who returns product frequently.
ABC can address such issues as: which products and product lines are creating profit;
which parts of the company's operations are effectively contributing to the customer's
perception of value; how indirect labor and overhead costs relate to product costs; and if
customized products are paying off.
Cokins adds that businesses, like the Coca Cola Company and Allied Signal
Corporation, "have been performing ABC for many years." These companies are
"advanced, mature ABC users," interested in two goals:
a. institutionalizing ABC companywide into a permanent, repeatable and reliable
production system; and
b. establishing the ABC output data to serve as an enabler to their ongoing
improvement programs, like Total Quality Management, change management,
cycle time compression, core competency, business process reengineering,
product rationalization, target costing, and channel/customer profitability.
Knowledge Management Solutions, Inc.
January 2001
40
DRAFT
(Not all Hyperlinks are Active)
The Framework of ABC
Basically, the framework of ABC – or looking at what things actually cost – can be
broken down into three areas.
1. The resource drivers, which includes the resources and the resource cost
assignment.
2. The management of activities, which means reviewing the actual individual
activities by themselves.
3. Looking at the activity cost drivers, which includes assessing individual activity
cost assignments and the cost objectives (or outputs).
"Activity cost drivers will have their own higher order drivers," says Cokins. "That sparks
root-cause analysis." Since the activity cost drivers for ABC costing are output-based,
the quantity of drivers can change over time, he warns. "ABC can either historically
report the trend in terms of per-unit cost of work rates of the activity outputs and
ultimately of the products. Or, it can provide the per-unit rates for use in predictive
planning and what-if scenarios, which are popular uses of ABC data."
The Cost Connection
How does a company, and a warehouse facility, prepare for looking at the cost factors
in an ABC assessment? Paul Larsen, ISP, product costing, Michael Foods Inc., a food
processing and distribution company based in Minneapolis, MN recommends these
steps:
•
•
•
•
•
•
•
assess the facility;
establish relevant cost centers;
identify activities, resources and drivers;
lay out the schematic/model;
collect data and rules;
design the model to fit the new cost centers; then
validate the model and the information.
How do you know if your ABC program is on target? Larsen, who has experience
developing these programs, says you need to ask yourself:
1.
2.
3.
4.
5.
Does it support the company's strategy and vision?
Does it really identify customer costs?
Can we use it for benchmarking externally?
Does it support our decision-making process for what-if scenarios?
Does it include performance measurements?
Knowledge Management Solutions, Inc.
January 2001
41
DRAFT
(Not all Hyperlinks are Active)
© 1997 Alexander Communications Group, Inc. Reprinted with permission of the
publisher located at 215 Park Avenue South – Suite 1301, New York, NY 10003.
Telephone (212) 228-0246.
Knowledge Management Solutions, Inc.
January 2001
42
DRAFT
(Not all Hyperlinks are Active)
2.0 Beyond Theory: The Practice Of Defining Activities
by Alan J. Stratton, Stratton Associates
In theory, there is no difference between theory and practice.
In practice, there is.
The central concept in activity-based cost/management is that of an activity. Yet, there's
very little published guidance on what constitutes an activity. According to the ABC
Glossary published by CAM-I (Consortium for Advanced Manufacturing International),
an activity is:
1. Work performed within an organization;
2. The aggregations of actions performed within an organization that are useful for
activity-based costing purposes.
This definition is so general that it is of little assistance to modelers. Consequently,
modelers have difficulty with activity definition issues that come up in the modeling
process. For example, it is easy for modelers to confuse activities with processes. Some
modelers puzzle over people activities versus equipment and facility activities. Speakers
at trade shows and conventions further confuse novices by talking about "top-down" or
"bottom-up" activity definition and about the number of activities in their models without
communicating the strategy governing their model requirements. Modelers also
frequently define separate activities that are, in reality, part of a more useful activity.
This article examines these issues and offers practical guidance on defining appropriate
activities for your ABC system.
Process and Activity
Recently, many organizations have progressed toward managing by processes or
making significant improvements with process re-engineering. While process orientation
is to be encouraged, many people have confused processes with activities in their ABC
models. Understanding the difference between activities and processes is key to
relating them to each other in useful models.
Processes consist of several activities directed to common outcomes or objectives.
Performance measures such as cost, quality, and cycle time are valuable in measuring
processes. Several different functional organizations may perform the activities that
make up individual processes. Sometimes, an ABC model calculates a cost per unit of
output for the process. This unit cost is an average cost of each output. It is useful for
many performance measures but is inadequate for cost object costing.
An activity is more detailed than a process. Each activity is made up of a set of more
detailed tasks. An activity is the lowest unit that is costed in a typical ABC model. Each
Knowledge Management Solutions, Inc.
January 2001
43
DRAFT
(Not all Hyperlinks are Active)
activity in a process may have its own activity driver to measure how cost objects
consume it. Potentially having its own activity driver enables the model to recognize
consumption differences from one activity to another in a process rather than averaging
consumption as in the process unit cost. Tasks are the detail steps involved in
performing an activity. Their value is mostly in understanding what makes up an activity.
Incorporating tasks into the ABC model results in a very large model but does not yield
significantly better information.
Process Costing
In models where costed processes are an objective, modelers are tempted to assign
resources directly to processes. Later, when they publish their results to department
managers, however, the managers can't relate to the results nor can they take action.
Why? Managers are accustomed to seeing their own departments' costs. With their
costs co-mingled with other departments in the processes, they don't know what portion
of the cost they own. In the end, they neither take ownership nor put the information to
its intended use ( see Figure 1). A better model design (see Figure 2) recognizes the
present functional organizational realities and difficulties. These realities and difficulties
are addressed in a model that:
1. Creates centers for each department
2. Identifies activities within each department center
3. Maps department activities to processes with attributes
Now, the model has the best of both worlds: the ultimate process view and the present
departmental activity view. Meanwhile, department managers have the opportunity to
manage their own activities as well as participate in cross-functional process
improvements. This model design also preserves departmental detail for good cost
object costing.
People Versus Equipment and Facility Activities
According to the first sentence of the CAM-I definition, an activity represents "work
performed within an organization." Who or what within the organization performs this
work? An organization is a system of employees, equipment, facilities, vendors,
subcontractors, and so on. Very often, an ABC model focuses on the work that
employees do because these activities are the most easily identified. Then all resources
from equipment, facilities, vendors, and subcontractors are assigned to these people
activities. Where these other resources are significant to the organization, this modeling
method force-fits the activities and may distort the results.
Equipment
It has been said that before machines became machines, they were people. People
once performed almost all work now performed by machines. Where equipment is a
significant organizational resource, the ABC analysis should ask what activities
Knowledge Management Solutions, Inc.
January 2001
44
DRAFT
(Not all Hyperlinks are Active)
machines perform in an activity center. In the analysis, it may turn out that people are
incidental to the equipment. In this extreme case, the ABC model would not have
people activities for this activity center. For example, in a recent ABC analysis of a
chemical process, the operators' prime responsibilities were to monitor the equipment
status and occasionally adjust the machines via a computer keyboard. Most of the
operators' time was spent watching the computer monitor. In this case the ABC model
assigned the people resource to the equipment activities rather than to a people activity.
On the other hand, if the machines are incidental to the work the people are doing or the
machines are tools people use to do people work, no separate activity may be
necessary. For example, many office functions now require a personal computer.
Employees use computers to prepare reports, analyses, letters, and so on. Generally,
when the employees take a break, the computers also take one. In this case, the
machine is simply a tool the employees use to do their work. No unique equipment
activities would make sense in this activity center.
Facilities
Most often facilities are areas housing people or equipment. In this case, the facility is a
resource to the people or equipment activities. However, what about a warehouse?
Does a warehouse perform significant activities other than the people or equipment
activities? Aside from putting material into the warehouse, taking it out, and recording
movement, the other most significant activity is sheltering materials or products. People
and equipment resources would be only incidental to this sheltering activity.
Vendors-Subcontractors
Careful attention should be given to vendor and sub contractor resources. These
resources represent significant activities that only differ in that the people performing the
activity are not employed by the organization itself. Logistics functions may be
subcontracted; Auditing is performed by outside auditors; Information Management may
be outsourced. Where these are significant, appropriate activities should be identified.
In the final analysis, activities should be defined with a broad view that includes
equipment, facility, subcontractor, and vendor activities in addition to people activities.
Top-down or Bottom-up
Which is the best approach to defining activities? The top-down approach looks broadly
at the organization's processes then decomposes them to some level of detail. On the
other hand, the bottom- up approach looks first at the frontline operations then maps
these activities to the business processes. Each approach has advantages and
disadvantages (see Figure 3) . This makes the final choice dependent on the model
objectives. In many cases, however, the approaches can be effectively merged to "have
the cake and eat it too."
Top-down
Knowledge Management Solutions, Inc.
January 2001
45
DRAFT
(Not all Hyperlinks are Active)
When activities are defined top-down, the model has fewer activities, and the activities
are more broadly defined. Successful top-down modeling approaches start with an
inventory of business processes and decompose them into sub processes and activities
to achieve the appropriate level of detail. This approach is more appropriate if the
primary purpose of the model is strategic information for upper management. Upper
management can quickly and easily understand the limited set of activities, and there is
not excessive detail. Middle and lower management levels, however, typically have
more difficulty relating to the broad activities and find the information difficult to use for
process improvement. If this approach is selected, an excellent starting point for
adaptation is the Process Framework developed by the American Productivity & Quality
Center and Arthur Andersen & Co., which is downloadable from www.apqc.org.
Bottom-up
When activities are defined bottom-up, the model has more activities, and the activities
are more narrowly defined. Successful bottom-up modeling begins with interviewing
managers and groups of employees to determine what work they do, as the basis for
activity definition. This approach is more appropriate if the primary purpose of the model
is operational information for process improvement by operating personnel. They can
easily understand the activities because this is exactly what they do, and they helped
define the activities. Because of the activity detail, more information is readily available
for process improvement efforts. After activity definition, the activities should be mapped
to the inventory of business processes to obtain process information. Typically there will
be more activities and more detail than upper management generally finds useful for
strategic direction. If unique activity drivers are selected for all activities, more time and
expense may be required to trace activity costs to cost objects.
Eating the Cake
When model objectives require both strategic and operational information, a
combination of these approaches is feasible and appropriate. In this case the model
should:
1. Define the activities with operating personnel.
2. Map the detail activities to business (see Figure 4) .
3. Report only process summary information to upper management.
4. Report detail activity information to operations management.
5. Create additional summary activities that combine activities having similar activity
drivers (see Figure 5) .
These summary or macro activities are used to consolidate activities, and only the
summary activities are traced to cost objects. This reduces the number of activity
drivers and model size without impairing model accuracy. In addition, cost object bills of
activities are more compact and understandable. While not replacing the detail activities
these summary activities can eliminate excessive model detail.
Knowledge Management Solutions, Inc.
January 2001
46
DRAFT
(Not all Hyperlinks are Active)
Problem Activities
Modelers also find some situations where significant time is spent difficult to model
productively. For example, what activities does a supervisor do? Are attending a
meeting or answering the telephone or sending email considered activities?
Supervisors
Most supervisors spend the majority of their time:
•
•
•
•
•
Scheduling
Job coaching
Disciplining and counseling employees
Overseeing work </LI.
Contributing as individuals to work
Individual contributions may constitute activities in their own right. Successful models
trace time involved in the tasks of scheduling, job coaching, and work oversight to the
activities being performed by the employees supervised. Since these are considered
tasks under the normal activities and easily traced, no additional activities are required.
The one unique activity is that of disciplining and counseling. If it is a significant
element, it should be separately defined and an appropriate activity driver identified.
Meetings, Telephone, Email
Time spent in meetings, answering the telephone and sending email is, one hopes,
intended to accomplish some objective either for the sender or the recipient. A tool is
not an activity by itself; it is only used to support other activities. It is very difficult to
identify an appropriate activity driver if a tool is identified as an activity. If, however, the
time spent attending meetings or otherwise communicating is traced to the activities for
which the meeting was held or communication occurred, the time and cost follows this
real activity. The only people for whom a meeting is a valid activity are full time meeting
facilitators.
Guidelines for Defining Activities
When defining activities for an ABC model, successful modelers will:
1. Distinguish processes from activities while preserving departmental activity views.
2. Define special activities, when significant, for equipment, facilities, vendors, and
subcontractors in addition to people activities.
3. Consider the purposes and uses of the model information in determining whether to
define activities from the top-down or from the bottom-up or both.
4. Not define activities for time spent using a tool to perform another activity.
Knowledge Management Solutions, Inc.
January 2001
47
DRAFT
(Not all Hyperlinks are Active)
In the end, modelers ought to remember the Willie Sutton rule: Focus on the high cost
activities. The rule is named after bank robber Willie Sutton, who was once asked, "Why
did you rob banks?" He is reputed to have replied, "Because that's where the money is."
Don't let your model get bogged down in unnecessary detail: it isn't where the money is.
Alan Stratton CPA, CMA, is an independent consultant specializing in activity-based
management implementation and software. He has been a significant contributor to the
field of ABC/M for many years. He has led ABC/M implementations around the world
and in many cultures. Alan is a Certified ABC Affiliate and works closely with ABC
Technologies' Application Services. He is co-author of An ABC Manager's Primer, the
leading introduction to ABC/ABM and co-author of Capacity Measurement &
Improvement, addressing capacity management issues. Comments and suggestions
are welcome. Please address them to the author at stratton_aj@compuserve.com.
Knowledge Management Solutions, Inc.
January 2001
48
DRAFT
(Not all Hyperlinks are Active)
3.0 The fifteen defining characteristics of an ABC/M system
(and without them, there is no way that a vendor can claim
they do ABC/M)
ABC/M Design And Architecture
•
•
•
•
Stand-alone Yet Integrated Application
ABC/M systems have sufficient functionality as to be most effective as a
stand-alone system, apart from the data feeding subsystems. ABC/M
systems work effectively as a "data mining" tool serving as an analytical
application. (refer to IDC's position paper on ABC/M by Henry Morris.)
Deficient ABC tools are imbedded in the traditional general ledger
accounting system. In that location, ABC costs get commingled with
traditional non-ABC/M ledger accounts and ultimately become dismissed
by managers as another mysterious blackbox restricted to operating and
use by cost accountants.
Minimum Two-Stage Cost Assignment
ABC/M systems reflect how cost objects use work activities and how
resources supply those activities with costs. This requires two cost
assignments.
Deficient ABC tools allocate the expenses from the traditional accounting
data and usually stop there. In some cases they may allow computing the
unit cost of an activity's output, but cannot further relate that cost to each of
the many unique cost objects consuming the activity; ABC/M does.
Multi-stage Cost Assignment Network
ABC/M systems re-assign costs through a cost assignment network. A
two-stage assignment is the minimum. But most ABC/M designs are
multi-stage because of: (1) activity-to-activity assignments, and (2) final
cost object-to-final cost object assignments (e.g., products into
customers).
ABC/M segments the diversity and variation of resources consumed by
diverse outputs, products, service lines, channels, customers, and cost
receivers. Activity costs may be used directly by these cost objects or
alternatively by other activities.
Deficient ABC/M tools are restricted by clumsy columns-to-rows cost
allocations. They are incapable of linking a source cost to two or more
stages of destination costs. Those using spreadsheets, for example, to
compute ABC/M often complain how quickly they "hit the wall."
Flexible Modeling
ABC/M systems allow for real-time re-configuring of the cost assignment
pathway architecture, substitution of alternative driver measures, and
immediate drill-down and analytical investigation of the data.
Deficient ABC/M tools require lengthy intervals between when design
changes are requested and when the programmers (often unfamiliar with
Knowledge Management Solutions, Inc.
January 2001
49
DRAFT
(Not all Hyperlinks are Active)
•
the intent) can re-configure the parameters that will compute the new
results.
Consolidation
ABC/M systems allow for multiple children-to-parent roll-up
consolidations--with rules to monitor model design consistency. This
allows for local managers to use ABC/M for managing, while their output
data can be combined for enterprise-view purposes.
Deficient ABC/M tools are stand-alone without capability of combining
multiple low level ABC/M models into higher level summarized one.
Using The Data From The ABC/M System
•
•
•
Contributed And Unitized Cost Elements
ABC/M systems can display for each cost object the entire list of cost
elements contributing to the cost object's total cost. In addition, the
ABC/M system can accommodate volume or quantity measures to
display the unitized cost (e.g., cost per processed invoice) for both the
cost object and each of its constituents' cost contributions (frequently
from the work activities).
Deficient ABC/M systems often do not even provide for any cost reassignments beyond the activity costs--therefore they have no cost
objects and cannot provide any visibility to the cost elements, total or
unitized, that comprise a cost object.
Reverse Cost Flowing/What-if Scenario Analysis
An ABC/M system can reverse the flow of costs to reflect incremental
changes in the quantity, mix, frequency, and intensity of drivers. These
changes can be for planned, budgeted, or hypothetical outlooks, and are
convenient for users to evoke in the software. The ABC/M system will
convert these changes into the impact on activity and resource costs.
This capability is essential for what-if scenario analysis and cost
estimating.
Deficient ABC/M tools usually can not even calculate the driver rates,
which in turn are a pre-requisite to even perform what-if scenarios. For
those lesser deficient ABC/M tools, they do not accommodate convenient
user-commanded changes in drivers to forecast future costs.
Balanced Scorecard & Strategy Alignment Integration
An ABC/M system must easily send its output data to become the input
for managerial use (ABC/M systems produce data and the "M" in ABC/M
results from using that data). Two popular uses for ABC/M data are
performance measurements and aligning the cost structure to the
alignment with the organization's mission and strategy (how much costs
support which strategy and the "degree of fit.")
Deficient ABC/M tools have no automated linkage and integration to
performance measurement and strategy definition applications.
Knowledge Management Solutions, Inc.
January 2001
50
DRAFT
(Not all Hyperlinks are Active)
•
•
•
•
•
•
Multi-Period Reporting
ABC/M systems recognize that trend analysis across time periods
becomes the favored way for users to analyze results to draw
conclusions. As ABC/M is computed at more frequent intervals, the
data serves more as "actual costing" measurement.
Deficient ABC/M tools, often imbedded in the general ledger, do not
provide quick and convenient trend analysis.
Attributes (For Scoring Or Tagging)
ABC/M systems allow users to conveniently create an unlimited number
of unique tags or scores, each one with its own user-defined scale (e.g.
high, medium or low value-adding) and applied to the ABC/M derived
costs. Then, two or more attributes can be concurrently viewed with any
ABC/M cost (see dimensionality).
Advanced ABC/M systems can trace the proportionate attributed costs
throughout the network into cost objects.
Deficient ABC/M tools have no facility to tag or score ABC/M costs.
Cost Assignment Network Analysis
ABC/M systems allow the user to conveniently and rapidly view the cost
assignment pathways from any location in the assignment network. All inward
(many-to-one) cost contributions and their source costs can be viewed; similarly
all of a cost's outward reassignments (one-to-many distribution) can be viewed.
Users can conveniently traverse down any assignment path and capture the
contribution/distribution view of costs.
Deficient ABC/M tools usually have no re-assignments. If a process-based
planning tool claims ABC/M functionality, it is likely mix-blind and has no cost reassignment capabilities.
Process View Costing
In addition to an ABC/M systems' popular capability to trace and reassign costs
based on cause-and-effect relationships, it can also chain-link the same activity
costs for a sequential cost build-up across time through business processes.
Deficient ABC/M tools may in fact only be tracing activity costs through business
processes. But they are process-dedicated and incapable of reassigning costs to
reflect and measure the diversity and variation of the mix of cost objects. They
are mix-blind.
Capacity Measurements
ABC/M systems can accommodate the input measures of available capacities at
very local detailed levels. Using those potential constraints, the ABC/M system
can historically report the location and magnitude of unused capacity costs.
Advanced ABC/M systems, using reverse cost flowing future projections, can
perform rough cut capacity planning (see APICS' definition) and alert users to
capacity constraints. Deficient ABC tools are capacity insensitive. They assume
infinite capacity in the historical view.
Yield Measurement
ABC/M systems allow for measurement and separation of off-spec and scrap
produced at any activity. It integrates these measures with other operating
systems.
Knowledge Management Solutions, Inc.
January 2001
51
DRAFT
(Not all Hyperlinks are Active)
•
Deficient ABC/M tools ignore measuring yield and relegate those measures to
alternative systems.
Cost Dimensionality Viewing
Costs re-assignments are linkages. By definition, there must be a contribution
and a distribution view of any cost. When other attributes or categories (e.g.,
sales region, product family) are combined with the cost assignment network,
there is an opportunity to perform multi-dimensional analysis. ABC/M systems
leverage on-line analytical programs (OLAP) to view costs.
Deficient ABC/M tools usually do not have cost objects that are the more popular
costs for multi-dimensional viewing.
If your ABC/M software cannot handle these critical functions, then you should take a
serious look at ABC/M software that can. Many software vendors put a "check in the
box" for ABC functionality. Put them to the test: their definition of ABC functionality must
meet the above characteristics, or you are being misled.
Knowledge Management Solutions, Inc.
January 2001
52
DRAFT
(Not all Hyperlinks are Active)
4.0 Government Uses of Activity-Based Costing
by John M. Vann, John Vann Associates Management Consulting
In the federal government, as in industry, there is a real hunger for accurate, easy-tounderstand financial data. Activity-Based Costing (ABC) satisfies that hunger for
government managers at all levels who have not been as involved in financial
management as they should, or would like to be.
Their lack of involvement is partly due to the complexity of government fund accounting
and its own insular language, which deals in such terms as obligation authority and
requires accounting for different kinds, or “colors” of dollars. It is also due to old financial
systems and old thinking that satisfy the need for accurate accounting of every
Congressional appropriation, but leave federal managers bereft of information for
decision making. For example, ask nearly any senior manager—comptroller or
engineer—on a typical Army installation for the full cost of electricity for an entire fiscal
year, and you are not likely to find a person who knows. The supporting financial system
was not built to provide that information. Until recently, managers have also not asked
for it.
In still other situations, government managers have simply ceded their responsibility for
financial management to the comptroller, who has frequently been quite willing to
accept that responsibility—along with its matching power to decide how money is spent.
That is changing due to the Chief Financial Officer’s Act of 1990, the Government
Performance and Results Act of 1993, the requirement to consider outsourcing of
government functions, and general cutbacks in funding due to downsizing of the federal
government and the emphasis on efficiency due to the National Performance Review.
Together, these changes require auditable financial statements, strategic planning,
performance measurement, knowledge of full costs of products/services and the
associated activities. The government environment is tailor-made for Activity-Based
Costing and its rapidly growing corollary, Activity-Based Management. Some of the
many uses for ABC and ABC-related information are shown below, with examples
based on Defense Department experience pioneering ABC work, primarily in the Public
Works arena, since 1992. (See Figure 1).
Product/service costs
Within the Department of Defense (DoD), as in industry, the most frequently cited
potential use for ABC information is to help installations determine the true costs of their
goods and services. The Defense Finance and Accounting Service (DFAS) does not
provide those costs. In the case of the Army, its own management structure does not
even identify outputs as the focus of the DFAS accounting system. The consequence is
that at a typical Army installation, the Directors have no supporting financial information
that provides them the true costs of basic services they provide such as water,
electricity, or building maintenance. If they want such data, they must devote scarce
maintenance and repair funds to an extensive manual effort to gather basic financial
data.
Knowledge Management Solutions, Inc.
January 2001
53
DRAFT
(Not all Hyperlinks are Active)
One such example is the provision of Public Works support to a military installation. At
one Army installation, the costs for the support totaled nearly $85 million dollars when
totaled from all accounting data provided at the installation. However, another $6 million
dollars, or about seven percent of the total cost, was not even shown as a cost in their
installation accounting records. (The funds were accounted for elsewhere and not lost to
DoD, but the installation had no visibility of costs.) How could an installation manage
when its accounting information is so incomplete? ABC helped provide the answer.
Military labor and depreciation, excluded in the standard accounting systems, were
included in ABC information.
In the case of military labor, both the Director of Resource Management and the
Directorate of Public Works were asked, based on accounting information each had
available, how much military labor was being used to perform public works functions.
The highest estimate they provided was about 5 man-years. After they used the ABC
technique, however, they discovered over 100 man-years of military labor, valued at
over $5 million dollars, performing tasks ranging from grass cutting to environmental
spill response. The increase in use of military manpower had all occurred because of
funding cutbacks in the blue collar workforce as part of the defense cutbacks. Senior
leaders knew there had been severe impacts due to funding reductions that many
believed had eaten into the “bone.” Until they had ABC data, however, they had no way
to articulate the results of the cuts. ABC gave them the tool they badly needed.
Reimbursements
DoD guidance concerning reimbursements has been and continues to be chaotic from
the installation perspective. The problems have been amply documented and forwarded
for resolution, but continue to be shelved within the accounting community due to their
complexity and the lack of understanding at higher staff levels of the installation level
impact of the reimbursement “swamp.” ABC can help solve some of the reimbursement
problems, particularly in determination of appropriate rates. The example of military
labor costs described above presents an example of how this can occur. Because
military labor costs are paid from a centralized account at Department of the Army level
and not shown in installation accounting systems, their “invisible” costs have in many
cases been ignored at installations. In reimbursement documents, this translates to
funding being lost. With ABC, these costs are included and can result in both increased
funding and a better understanding of costs.
Activity costs
One of the major benefits decision makers find from ABC is knowing the costs of
activities performed by their people. As obvious as this sounds, it is apparent following
introduction of ABC to every organization, that such information has simply not been
available or accessible until the advent of ABC. Value added vs. non-value added
activities can be identified, costed, and earmarked for change as appropriate using ABC
information. The opportunities to review the effort and desirability of continuing to
perform selected activities in any organization stand out readily. For example, in one
directorate studied in 1996, the activity “Take Phone Messages for Others” was
Knowledge Management Solutions, Inc.
January 2001
54
DRAFT
(Not all Hyperlinks are Active)
evaluated. Skeptics said that the activity was defined at too low a level of detail, until
they saw the annualized cost of $261,000. Once known, the Director took immediate
action to procure answering machines for the entire organization, at a cost of about
$30,000. The cost of implementing ABC at the organization was less than a quarter the
cost of the time saved by that single activity.
Business Process Reengineering
The costs of entire business processes can be determined and used as the bases for
changing, or reengineering, throughout an entire installation. As funding cutbacks force
reduction in the scope of services provided at an installation, it is increasingly important
to know what processes, not simply organizations, are costing the most. Armed with the
“horizontal” perspective in addition to the traditional stovepipe perspective of costs, one
Army installation recently reorganized from the traditional directorate concept to a
concept oriented around a reduced number of “business centers” that take advantage of
more efficient organizations, reduced overhead, and leaner support structures. ABC
data helped decision makers identify the most effective way to structure the business
centers (See Figure 2).
A Directorate of Contracting office that recently applied Activity-Based Costing for the
first time discovered that the activity of applying the “streamlined” purchasing
procedures promulgated by DoD was actually costing more money than it was saving
for small purchases (under $5,000). Under the new rules, installation personnel were
reviewing a significantly increased number of procurement actions, resulting in a cost of
$141,000 annually. ABC was used to identify the unusually high cost of the activity, and
subsequent management action to reduce unnecessary workload resulted in a saving of
$57,000.
Planning and activity-based budgeting
DoD’s management system is called the Planning, Programming, and Budgeting
System (PPBS). Created in 1962 in an early effort to bring business-like solutions to
Pentagon thinking, it is now outdated, and its multiple problems are well documented.
While it does provide detailed information about Congressional appropriations
associated with broad military missions, it does not provide the kind of resource
justification information that is easily understood and usable at all levels. It also does not
focus at lower levels on the outputs that result from congressional appropriations. How
can ABC data help? There are multiple ways. Two key contributions stand out
immediately, however.
First, ABC offers a way to begin to show managers the impact of budget cuts on
activities and missions performed BEFORE the cuts are made. The existing system
focuses on accounting for organizations that are funded and on the general functions
they perform. For example, for an installation Director of Logistics, the Army accounting
system accounts for the cost of the Office of the Director—as if it were an output. What
does that office do? What would happen if that office absorbed a cut? What would it do
to the full cost of the outputs of the DOL? At present, there is no linkage, and a
subjective evaluation at best is all that could be offered.
Knowledge Management Solutions, Inc.
January 2001
55
DRAFT
(Not all Hyperlinks are Active)
With ABC, the contributions to the outputs of the Directorate (repaired tactical
equipment, supplies provided, transportation provided, etc.) of the Director’s office, and
all other “overhead” offices, as well as the activities they perform in helping produce the
Directorate’s outputs, are known. With current software, “what-if” scenarios can be run
to determine the cost implications of changes in activities, organizations, or resource
inputs. Secondly, the kind of information provided by ABC can contribute to a major
improvement in budgeting. The example of military labor described earlier offers a good
example of how this can happen. Military labor costs are paid from a centralized
account retained at Department of the Army level. Installations do not budget or account
for military labor, and the cost of military personnel is not shown in installation financial
systems, even though it is a legitimate cost to the Army and should be charged to some
customers. Since it is not paid for by installations, it has traditionally been considered
“free” labor. However, if military labor is used to replace civilian labor, which is budgeted
for and accounted for at installation level, the workload remains the same (or may grow
as military personnel attempt to replace a skilled civilian workforce) while the civilian
labor cost to accomplish that same work-load goes down. This shift results in everincreasing understatement of base support costs. With ABC, the full military labor costs
are considered part of base support, giving Directors at an installation a way to
articulate the impact of cuts. Other examples abound of ways ABC, through its complete
audit trail, easily understandable terms, and full cost methodology, can contribute to
improved planning and budgeting in DoD.
Benchmarking
Benchmarking is simple comparison, whether it be with the “best practices” of a similar
service provider or the cost of the service he provides. ABC data, if prepared with
sufficient accuracy, can provide managers with useful information upon which to base
such comparisons. It can be used locally to compare costs of utilities, equipment
maintenance, child care, or a host of other services installations provide. It can provide
a “first cut” to managers concerned with the needs to perform studies concerning
outsourcing (A-76 studies). And it can help Directors estimate in broad terms how their
organizations perform compared to Army averages.
However, care should be exercised at higher levels to avoid “apples vs. oranges”
comparisons. If one installation produces its own water and another is forced to buy
water from a local water authority for twice the price due to scarcity of water, what
sense does it make for an analyst at higher headquarters to perform the non-value
added activity of comparing the costs of water at the two sites (See Figure 3).
Performance measurement
With the advent of the 1993 Government Performance and Results Act (GPRA), all
government managers are under a mandate to evaluate their work based upon
performance. While the concept has long been used in the uniformed military and in
industry, it is relatively new in government. Performance has traditionally been
measured more on increasing budgets rather than on specific measures of output. ABC
information tied to performance is tailor made to the intent of GPRA.
Knowledge Management Solutions, Inc.
January 2001
56
DRAFT
(Not all Hyperlinks are Active)
With its focus on full costs, ABC can simply divide the full cost by the appropriate output
measure to obtain a performance measure for the output of a Directorate. For example,
the unit cost of electricity could be expressed as dollars per megawatt/hour.
ABC also can add, however, the aspect of unit cost of performance of activities.
Managers want to know about the relative efficiency of their workforce. For example, a
manager curious about the cost of overhead administration associated with each
contract can find that information using ABC. He could also determine the costs of the
specific activities associated with the contract administration. For example, pre-award
activities could be separated from post-award activities, enabling him to evaluate the
true cost and effort associated with different types of procurement vehicles and
providing substance to his intuitive knowledge.
Conclusion
As shown in the examples described above, DoD is representative of the United States
government’s efforts to use ABC to help become more efficient, cope with the realities
of reduced resources, and make sure that taxpayers are all getting the most from their
tax dollars. ABC is performing an increasingly important role in helping shift the culture
of the federal government toward increased awareness of costs and more careful
management of costs. Through these outcomes, ABC is helping the federal government
to retain the public trust.
Portions of this article appeared previously in Armed Forces Comptroller, Winter 1997
and have been reprinted with permission. Copyright © 1997. All rights reserved.
John Vann is an independent consultant who specializes in Defense-related issues,
focusing on defense management practices. Following graduation from the United
States Military Academy, he had a 26-year career in the Army and held a variety of
senior positions in the Department of Defense, including Director of the Program
Evaluation Directorate, Defense Inspector General, an organization created in 1990 to
review big picture issues for the Secretary of Defense. Prior to that, he was Chief of the
Program and Budget Division in the Army Deputy Chief of Staff for Logistics; Deputy
Director, Budget Integration, Office of the Assistant Secretary of the Army for Financial
Management; and Associate Director, Program Integration, Office of the Assistant
Secretary of Defense for Manpower, Reserve Affairs, and Logistics, where he
specialized in analysis of defense-wide issues. He has served in command, staff, and
advisory positions from installation level, as Assistant Chief of Staff for Resource
Management, to the Office of the Secretary of Defense level. Mr. Vann introduced the
Activity-Based Costing concept to the Army community at multiple locations. He has
spoken extensively concerning Activity-Based Costing, including giving presentations at
the International Productivity and Quality Symposium and International Research
Institute. He taught both financial management and its relationship to ABC at the Army
Installation Management Course.
Knowledge Management Solutions, Inc.
January 2001
57
DRAFT
(Not all Hyperlinks are Active)
5.0 ABC at NASA'S Lewis Research Center
by Jeff Walters, Pat Rattan, Steve Smith, and Dr. Douglas Webster, American
Management Systems
Lewis Research Center (LeRC), in Cleveland, Ohio, is one of four primary research
facilities of the National Aeronautics and Space Administration (NASA). LeRC conducts
research and development in both aerospace and aeronautics, specializing in microgravity science, aero-propulsion, and space systems.
To support these research missions, LeRC provides a number of "institutional services"
that are used throughout the Center across many research programs. Examples of
these services include Scientific Computing, Fabrication, Graphics and Reproduction,
and Imaging. These services are funded as a group by Congress via a "program
support budget" which is augmented as necessary by "taxing" direct research projects.
In more prosperous days at LeRC, the "program support budget" was an adequate
solution because the practice of withholding funds from or "taxing" direct research
projects did not significantly limit the Center's ability to perform its mission. However, the
program support approach provided little visibility in the sources and uses of funds. As
budgets diminished, research managers became more aware of, and accountable for,
the use of all research dollars. They began to question the "taxation" methodology of
augmenting the institutional services budget, creating a demand for a different means of
accounting for the cost of institutional services.
In 1992, a Center team offered a potential solution: establish an "Internal Economy"
within the center. The Internal Economy concept espouses a market-based charge-back
system in which institutional service providers "sell" products and services to internal
customers (direct-funded research programs and projects).
The Quality Council of LeRC quickly affirmed the Internal Economy concept and
established a team to investigate an Internal Economy at LeRC. In late 1994 the
Internal Economy Process Action Team (IEPAT) completed an exhaustive analysis of
the opportunities and risks related to implementing a market-based charge-back system
and consequently was granted authority to proceed with the design and implementation
of the Internal Economy at LeRC.
Design and Implementation
American Management Systems, Inc. (AMS), an Information and Management Systems
consulting company headquartered in Fairfax, Virginia, was asked to help LeRC apply
activity-based costing as a tool to implement the Internal Economy concept. AMS was
Knowledge Management Solutions, Inc.
January 2001
58
DRAFT
(Not all Hyperlinks are Active)
tasked to support the process action team's development of a project implementation
plan and develop costing methods as well as baseline product and service costs for
three internal providers.
Implementation Planning
AMS facilitated a planning workshop of the IEPAT and the key stakeholders of the
internal economy process. The workshop served as a vehicle for gaining the critical
executive support and resource commitments required to enable success of the project.
Workshop attendees together developed a schedule and plan for the remaining design
and implementation tasks. The plan was based on four critical tasks:
•
•
•
•
Purchase Transaction System
Policies, Procedures, and Mandates
Product Costing
Education
Each task was assigned a lead LeRC or contractor representative who was responsible
for the execution and completion of all associated activities. The AMS team was
assigned responsibility for the product costing portion of the plan and for the
introduction of ABC as the mechanism to accomplish that task.
Service Provider Product Costing
Accomplishing the Costing phase of the initiative required AMS to lead three initial
service providers in:
•
•
•
•
•
Product Identification and Definition
Process Modeling, Activity Definition, and Activity Driver Determination
Resource Identification, Resource Decomposition, and Resource Driver
Determination
ABC Model Development
Baseline Costing
The Imaging Technology Center (ITC), the Scientific Computing Line of Business of the
Computer Services Division, and the Metrology Services Branch were chosen as the
three initial service providers to transition to the Internal Economy. Their selection was
based on a combination of breadth of impact, potential savings, data availability, and
probability of success.
Product Identification and Definition
This phase of the project consisted of developing a "catalog" of products and services
offered. Each entry in the catalog consisted of a product name, description, and unit of
purchase. The level of effort required to develop the catalogs varied by organization,
based on the type of products offered—standard or non-standard. Standard products
(or services) require the performance of the same set of activities in the same
proportion each time a unit of output is produced. Non-standard output may require
different activities in different proportions for each unit of output.
Knowledge Management Solutions, Inc.
January 2001
59
DRAFT
(Not all Hyperlinks are Active)
The Metrology Services group, the "television repairmen" of the Center, could not define
standard products for inclusion in a catalog. Each job order defined a different set of
requirements and therefore a different set of activities. Metrology Services chose to
define only two product lines, Instrument Pool Rentals and Repair Work Orders and
then develop a job costing method to support each. In other words, by knowing the cost
of all their activities, they could "construct" the cost of a particular service depending on
the activities performed to provide the service.
Similarly, the Computing Services Division (CSD) began the analysis with a "job-shop"
outlook but clearly recognized that some of its products or services, such as networking
and data storage, could be defined, cataloged, and sold in repeatable units. To facilitate
the identification of CSD products and services, the project team led a structured
workshop with the customers and staff of CSD. Independently, each group identified the
products and services from its perspective. The groups then developed a common set
of products and services identifiable to all parties.
The Imaging Technology Center (ITC) presented a different set of circumstances.
Having operated in a pseudo-charge-back system prior to the Internal Economy
initiative, the ITC came to the effort with a product catalog already designed. The
catalog only needed to be updated to reflect recent organizational and service changes.
Process Modeling, Activity Definition, and Activity Driver Determination
A process model was required for each organization in order to determine which
activities supported which products. The team developed the activity model using
structured interviews with LeRC representatives from each organizational sub-element
that supported one or more aspects of the entire process. Workshops were used to
validate the integration of the smaller models into the larger organizational model.
Once activities were identified and defined, activity drivers were selected based on
accuracy of measurement and degree of difficulty of data collection. For many activities
two sets of drivers were identified—one that could be used based on current data
available and another more accurate driver that should be used if measurement and
data collection techniques could be established.
Resource Identification, Decomposition, and Resource Driver Determination
The objective of resource modeling was to identify the resources consumed by the
organizations in performing their activities. In LeRC's case, the combined resources
used by civil servants and contractors to perform work were captured in four different
general resource categories:
Civil Service Costs
Labor costs, including payroll and benefits, were decomposed into hourly rates based
on pay grades. "Number of labor hours" was used as the driver to trace these costs into
activities. Travel and training costs were broken down by "trip" and "class" respectively
and traced directly to the activity which each supported.
Knowledge Management Solutions, Inc.
January 2001
60
DRAFT
(Not all Hyperlinks are Active)
Contractors
LeRC support contractors were already under contract to provide cost information by
task to the government for auditing. Since the tasks defined in the contracts were not
always equivalent to the activities of the ABC process model, the contract tasks were
converted to the ABC-defined activities, and the costs were assigned accordingly.
Government Owned Equipment
LeRC maintains an extensive database with data for each piece of owned equipment.
An amortization schedule was created for each item, and those that still had current
year costs were retained in the model. The costs were then assigned to the activities
that applied the equipment.
Government Purchases
Using government purchase order information, a schema to identify 'expenseable'
purchases from 'depreciable' items was developed, and the former were retained as
period costs. These purchases were then allocated to the activities which consumed the
resources.
ABC Model Development
With resources, activities, products, and drivers defined, a reusable model was needed.
Relationship matrices were used to integrate the segments of the model. The matrices
allow process participants to identify the relationships between these elements without
worrying about the availability of data to populate the model relationships.
With each element of the ABC models now defined for each organization, three models
were built using EasyABC Plus. The majority of the data was input to the model
structure using the software's import capabilities. The resulting ABC models
incorporated:
•
•
•
•
•
•
•
complete resource listings
complete activity hierarchy, with definitions for each activity at each level
complete cost object (product and service) catalog, with definitions for each
resource-to-activity links
activity-to-product links
resource and activity driver assignment and frequencies
activity and product output units and volumes
Baseline Costing
The calculation of costs for products and activities, both on an annual basis and on a
per unit basis, was a relatively simple next step. Reports for each of the three service
providers were generated directly from EasyABC Plus. The reports displayed every
aspect of the model, including the cost breakdown of each product and activity. LeRC
now had baseline costs with which to develop prices for use in an internal economy.
While it is understood that more goes into pricing decisions than just costs,
understanding the true costs is critical to survival in an internal market economy.
Knowledge Management Solutions, Inc.
January 2001
61
DRAFT
(Not all Hyperlinks are Active)
Activity-Based Costing is one of the most accurate methodologies available to help gain
that understanding.
Jeff Walters is a Senior Business Process Re-engineering (BPR) Analyst with American
Management Systems. In addition to his ABC work for NASA LeRC, Jeff has supported
the Langley Research Center with ABC, BPR, and Performance Measurement
expertise. Jeff received his BS in Aeronautical Engineering from Pennsylvania State
University and his MBA from Rensselaer Polytechnique Institute.
Pat Rattan is a Senior Cost Management Analyst with American Management Systems
(AMS). Mr. Rattan received his BBA in Finance from Texas A&M University and his
MBA from William & Mary. Mr. Rattan has expertise in the banking and insurance
industries and served in a variety of financial management positions with the U.S. Navy.
He is currently helping an AMS client in the not-for-profit sector develop and implement
an ABC infrastructure.
Steve Smith, a Principal at AMS is a Senior BPR Analyst in AMS' Industrial Consulting
and System Group where he leads a performance measurement business practice. Mr.
Smith is currently supporting an ABC project with the U.S. Army and is a frequent
speaker and workshop facilitator in the area of performance measurement. Mr. Smith
received his BA from Allegheny College and his MPA from Penn State University.
Dr. Douglas Webster, co-author of Activity-Based Costing & Performance, and ABC
Technologies Affiliate, is a nationally-recognized expert in ABC. Dr. Webster, a Principal
at AMS, has supported NASA's LeRC and Langley sites, and is currently assisting the
U.S. Army in developing a strategic plan for a multi-site implementation of ABC. Dr.
Webster is a frequent invited speaker at national conferences and workshops on ABC.
Dr. Webster received his doctorate degree from U.S. International University and his
MS in Systems Management from the University of Southern California.
Knowledge Management Solutions, Inc.
January 2001
62
DRAFT
(Not all Hyperlinks are Active)
6.0 If ABC is the Answer, What is the Question?
by Gary Cokins, Director of Industry Relations
ABC Technologies, Inc.
"Traditional managerial accounting is at best useless, and at worst dysfunctional and
misleading."
- Professor John K. Shank, Dartmouth College
Keynote Address, Institute of Management Accountants' 75th Anniversary Conference,
June, 1994.
Why do managers shake their heads with disbelief when they think about their
company's cost accounting system? Imagine if you were a roving reporter and asked
managers throughout your organization, "How happy are you with the existing financial
and accounting data to support our decisions to improve our competitiveness?" Thumbs
up or down? Many give it thumbs down.
How can traditional accounting that has been around for so many years all of a sudden
become considered so bad? The answer is the existing data is not necessarily bad as
much as it is incomplete and unprocessed. Figure 1 provides the first hint of a problem.
The left side shows the classic monthly responsibility statement report that managers
receive. If you ask managers how much insight they get of the more controllable costs
they incur, they will answer, "Not much!" This is because the salary and fringe benefit
costs make up the most sizable portion of controllable costs, and all that the manager
sees are those costs reported as lump-sums. But when you translate those "chart-ofaccount" expenses into the work activities that consume the financial general ledger's
expenses, a manager's insights from viewing the activity costs begin to increase. The
right side is the Activity Based Costing (ABC ) view.
To be further critical of the left side "chart of accounts" view, notice how inadequate that
data is to report business process costs which run cross-functionally, penetrating
through the vertical boundaries of the organization chart. For example with a
manufacturer, what is the true total cost for processing engineering change notices
(ECNs) that travel through so many hands? For a service organization, what is the true
cost of opening a new customer account? Many organizations have flattened and delayered such that employees from different cost centers frequently perform similar
activities and multi-task in two or more core business processes. Only by re-assembling
and aligning the work activity costs across the business processes, like "process ECNs"
or "open new customer accounts" can the end-to-end costs be seen, measured and
eventually managed.
In effect, with traditional cost systems, managers are denied visibility of the costs along
the end-to-end business processes, particularly the stocking, distribution, marketing and
selling costs which the traditional accounting "expenses to the month's period" and does
not proportionately trace the costs to the unique products, containers, services,
Knowledge Management Solutions, Inc.
January 2001
63
DRAFT
(Not all Hyperlinks are Active)
channels or customers that cause those costs to occur.
In sum, the left side of Figure 1 is the "accounting police" command-and-control tool.
Have you overspent your budgeted target? If you have, who says that budgeted target
amount was fair when it was initially imposed? When managers receive the left side
responsibility center report, they are either happy or sad but rarely any smarter. Today's
competitive world will be dominated by "learning organizations," not one straightjacketed with spending restrictions while overstaffing and non-value adding work occurs
elsewhere in their organization. The right side of Figure 1 defines the same costs but in
a "verb-adjective-noun" grammar convention. This gives ABC its flexibility. It is cost
information for employees and managers, not just accountants. People can relate much
better to this language, and they can favorably impact the work that the activity name
describes.
Cost accounting is outside many individual's comfort zones. ABC makes cost
understandable and logical.
ABC as a Translator not Replacement
Figure 2 describes that ABC is not a replacement for the traditional general ledger
accounting. Rather, it is a translator or overlay that lies between the cost account
accumulators in the general ledger and the end-users, like managers, who apply cost
data in decision making. ABC translates costs into a language that people can
understand and into elements of costs, namely the work activities, which can be more
flexibly linked or assigned to business processes or cost objects based on demanddriven consumption patterns, not simplistic cost allocations. The reason ABC is
becoming popular is because the general ledger is now recognized as being structurally
deficient for delivering good business information for decision support --- the general
ledger is a sound mechanism for collecting and accumulating transaction-intensive
costs, but not for converting those costs into useful managerial information.
In contrast to traditional accounting, ABC focuses on the work activities associated with
operating the business. ABC is very work-centric, whereas the general ledger is very
transaction-centric. Both have their place, but the general ledger's data is too raw to be
considered business intelligence for decision support. ABC solves the general ledger's
flaw of unprocessed cost data. But ABC does much more. Just translating the ledger
account expenses into their work activities is an incomplete story of ABC. The total ABC
picture comes from linking the activities into networks to cost out items for analysis,
trade-offs and decisions.
How does Activity-Based Costing compute better accuracies?
ABC was developed as a practical solution for problems associated with traditional cost
management systems, which we now realize are incomplete. However, traditional
indirect expense and overhead cost allocation practices of traditional systems bring
more damage than good to organizations. Indirect costs are usually too aggregated, use
excessively broad-brushed average rates to allocate the costs. And worse yet, theyt rely
on a sales volume-based factor or basis, like direct labor hours or department
Knowledge Management Solutions, Inc.
January 2001
64
DRAFT
(Not all Hyperlinks are Active)
expenses, which is rarely reflective of the true and specific cause-and-effect relationship
between the indirect overhead expense and the part, product, service, channel or
customer (i.e. the cost object) that is actually consuming the cost. Many managers are
tired of the "allocation foodfights."
The result of inaccurate cost allocations, because allocating is a zero-sum error game,
is that some cost objects get over-costed while the remainder are under-costed. In
practice, one discovers the under-costed products are substantially under-costed
because they may be low-volume with small lot sizes, require more technical attention,
consume more handling, and need extra inspection. In effect, we have allowed the
accounting profession to construct a costing scheme that distorts reality and violates
variable costing as a manager understands it. The ultimate problem is companies are
actually losing money on certain products, orders, services and customers when their
accounting system states they are profitable. And since the price quotation practices
usually rely on the same flawed cost data, quoting unprofitable orders to unprofitable
customers is perpetuated with the illusion that they are profitable!
ABC corrects for these flaws by identifying the work activities that are responsible for
costs. ABC provides a cost flow assignment network which allows the work activity
costs to be continuously re-assigned or passed on only if the products, services or
customers, or in some cases other work activities, use the activity. Figure 3 shows a
popular diagram called the CAM-I* Cross.
The CAM-I Cross reveals that work activities, which are located in the center of the
Cross, are integral to reporting both the costs of processes and the costs of cost
objects. The vertical axis reflects costs as they are sensitive to demands from all forms
of product and customer diversity and variety. The work activities consume the
resources, and the products and customer services consume the work activities. It is a
cost consumption chain. When each cost is traced based on its unique quantity or
proportion of its cost driver, all of the costs are eventually re-aggregated into the final
cost objects. This method provides much more accurate product and customer costs
than the traditional "peanut-butter spreading" cost allocation method. The activity cost
drivers behave like a "pump and valve" in this cost re-assignment network. Cost drivers
are critical to ABC because not only do they segment and flow the costs to reflect the
diversity of the products and customers, but they govern the accuracies as well.
In practice, ABC systems will trace work activity costs to two or more other intermediate
work activities that consume the work upstream from the ultimate products and
customer services which initially trigger the demands on work. The re-assignment
network of cost-segmented consumption is responsible for the majority of ABC's
superior costing accuracy. ABC can tolerate reasonable cost driver estimates as proxies
for actual transaction detail drivers because the error does not compound--it dampens
out on its way to the final cost objects. It is somewhat counter-intuitive, but with ABC,
precision inputs is not synonymous with accurate outputs. This property significantly
lightens the load for data collection. And this is also why the ABC advocates' mantra is,
"It is better to be approximately correct than precisely inaccurate!" In the end, the level
Knowledge Management Solutions, Inc.
January 2001
65
DRAFT
(Not all Hyperlinks are Active)
of accuracy and detail depends on what decisions are made with the data. And usually
the accuracy requirements are not unreasonably harsh.
In the CAM-I Cross' vertical cost assignment view, activity cost drivers will have their
own higher order drivers. That sparks root cause analysis. But usually the activity cost
drivers used for ABC costing are output-based. Therefore, as the quantity of the drivers
rise or fall over time, ABC can either historically report the trend in terms of per-unit cost
of work rates of the activity outputs and ultimately of the products. Alternatively, ABC
can provide the per-unit rates for use in predictive planning and what-if scenarios, which
are popular uses of the ABC data. Cost estimating with ABC is very natural because the
activity costs react and behave linearly with changes from their cost drivers. Too often
with traditional costing, the cost rates do not directly vary with changes in volume which
results in mis-estimates and ultimately errors and lost profits.
The horizontal view of the CAM-I Cross is the business process view. A business
process is defined as two or more activities or a network of activities with a common
purpose. Across the process, the activity costs are sequential and additive and satisfy
the requirements for popular flow charting and process modeling techniques and
software. Business process-based thinking, tipping the organization chart ninetydegrees, is now dominating managerial thinking; and ABC provides the cost elements
for process costing that are not available from the general ledger.
What gave rise to ABC?
In the early 1980's many companies began to realize the adverse consequences of
allowing their traditional accounting systems to generate inaccurate costing information.
An organization's cost structure had been substantially changing. For most businesses,
overhead and indirect costs were increasing and mainly displacing the direct labor
costs, primarily due to increased complexity caused by: (1) proliferation of product and
service offerings, (2) more types of channels and customers, and (3) new and complex
technologies.
With greater overhead costs relying on a basis for cost allocations tied to unrelated
volumes, the traditional costing method had become invalid relative to how the products
and services consumed costs. Therefore the unfavorable impact of the costing errors
were becoming much more intense than in the past. Many managers understood
intuitively that their accounting system was distorting the product costs, so they
sometimes made informal adjustments to compensate. However, with so much
complexity and broad product and channel diversity, it was near impossible for
managers to predict the magnitude and impact of the adjustments they should. Those
conditions, in fact more heightened, exist today.
Figure 4 reveals how ABC firsts
"segments" resource costs to reflect diversity, and then ABC re-aggregates costs
through its cost assignment network into the demand-causing objects, namely suppliers,
products and customers.
In addition to the distortion of true costs from traditional systems, the rise in ABC has
resulted from external factors. The level of competition that most firms face has
Knowledge Management Solutions, Inc.
January 2001
66
DRAFT
(Not all Hyperlinks are Active)
increased dramatically. In the past, organizations were reasonably profitable. They
could make mistakes and their profitability would mask the impact of their wrong or poor
decisions. They could carry unprofitable products and customers because the winners
would more than offset the losers. They could still survive with misleading cost
allocations and without visibility of their costs of business processes.
Today the margin for error is slimmer. Businesses can't make as many mistakes as they
could in the past and remain competitive. Price quotations, capital investment decisions,
product mix, technology choices and make vs. buy decisions all require a sharper
pencil. More competitors are better understanding the cause-and-effect connections
which drive costs and fine-tuning their prices. The resulting price squeeze is making life
for businesses much more difficult relative to the past. Knowing what your real product
costs and the costs-to-serve channels and customers is becoming key to survival. With
ABC visibility, businesses can identify where to remove waste, low-value adding costs
and unused capacity plus understand what drives their costs.
And today, a business' road is no longer long and straight, but it is windy with bends and
hills which don't give much visibility or certainty to the future. Organizations needs to be
agile and continuously transform their cost structure and work activities. This is difficult
to do when their organization doesn't understand their own cost structure and
economics.
A final reason giving rise to ABC is the cost of measurement has fallen as information
processing has improved. Not too long ago, it was cost prohibitive to accumulate,
process and analyze the data necessary to run an ABC system. Today, such activity
measurement systems are not only affordable, but much of the data already exists in
some form within the organization.
ABC in Advanced Mature Users
Businesses like the Coca Cola Company and Allied Signal Corporation have been
performing ABC for many years. They are advanced and mature ABC users who are
interested in two goals: (1) to institutionalize ABC company-wide into a permanent,
repeatable and reliable production system and (2) to establish the ABC output data to
serve as an enabler to their ongoing improvement programs, like TQM, change
management, cycle-time compression, core competency, business process reengineering, product rationalization, target costing, and channel/customer profitability .
More recently, new issues for the advanced and mature ABC users are emerging and
include: integrating the ABC output data with their decision-support systems, such as
their cost estimating, predictive planning, and activity-based budgeting (ABB) systems;
learning the skills and rules for re-sizing, re-shaping, re-leveling and otherwise readjusting the model's structure in response to solving new business problems with the
ABC data; and collecting and automatically importing data into the ABC system. With
experienced ABC users, it is evident that ABC eventually becomes part of their
information technologies.
Knowledge Management Solutions, Inc.
January 2001
67
DRAFT
(Not all Hyperlinks are Active)
More specifically, the output data of an ABC system is frequently the input to another
system, such as a customer order quotation system. ABC data also complements other
productivity or logistics management tools such as software simulators, process
modelers, business process flow charters, schedulers, executive information systems
(EIS), and on-line analytical programs (OLAP). In the next three years, there will be a
convergence of tools as these now somewhat separate software products become part
of the analyst's tool suite.
Advanced mature users are also masters with employing ABC "attributes" which are
scored and graded against the activities. ABC attributes allow managers to differentiate
activities from one another. A popular attribute involves scoring activities along their
"high vs. low value-adding" scale. Multiple activities can be simultaneously tagged with
these grades, and of course the dollars trail along at the activity level. As an option,
activities can be summarized into the processes. This way organizations can see, for
example, if they are very, very good at things they have judged to be very unimportant -- and they spend a lot of money doing it! Although some attributes are subjectively
scored or graded by managers and employees, they introduce emotionally compelling
business issues. I have often said that, "ABC adds the air-conditioning to the ABC
data."
The Behavioral Side of ABC
The ABC data is simply a means to ends. That is why ABC itself should not be labeled
as an improvement program because then it becomes a candidate for "fad of the
month." ABC data simply makes visible the economics of the organization and cost
consumptions which are occurring with or without ABC present. ABC in one sense
operates as "an imaging system" similar to radar, sonar or CAT scans. Costs measure
effects moreso than the root cause, and the ABC systems provides the enterprise-wide
image of all the effects plus the causal relationships causing the cost effects.
ABC project managers have been slow on the up-take to recognize the behavioral
change management aspects of the ABC data. ABC is a socio-technical tool, and the
emphasis should be on the social side. Many managers and ABC project teams see
ABC as simply a better measuring scheme or cost allocator. However, it's real value is
introducing undebatable fact-based data which can be used by employees to build
business cases, to quickly recognize business problems or opportunities, and to test
hypothesis. ABC has many of the characteristics of an organizational methodology.
Many managers are frustrated by the difficulties in bringing about change within their
organizations. Behavioral change management is receiving wider attention, and ABC
data is playing an important role with change.
A Business is Multidimensional
ABC contends that many important cost categories vary not with short-term changes in
output, but with changes in the design, mix and range of a company's products, services
and customers. Once product costs are identified users begin to see the value of
understanding the activities and their associated costs. The primary use of ABC shifts
from that of an accounting tool to that of a management decision support system for
Knowledge Management Solutions, Inc.
January 2001
68
DRAFT
(Not all Hyperlinks are Active)
operational streamlining - which can be called Business Intelligence. Information
technology gathers and manages this ABC information, combining not just cost but nonfinancial information and performance measures. This enterprise-wide technology is
termed an activity-based information system. As more managers become aware of the
activity and information that is available, additional applications for ABC will open-up,
including unused capacity management. ABC provides the lens that focuses an
organizations' efforts.
ABC and the Future
An overarching issue with ABC involves its perception as just another way to spin
financial data rather than its use as mission-critical managerial information. The
Information Age we are entering can be mind-boggling. In our future, as technology
advances, so will the demand to access massive amounts of relevant information. The
companies that survive will be those that can answer these questions:
"How do we access all this data?"
"What do we do with it?"
"How do we shape the data and put it in a form with which we can work?"
"What will happen when we apply technologies developed during the Information Age
for the Information Age?"
Clearly, as information technology evolves, organizations will increase their
effectiveness. Further, as markets change, companies and organizations will run into
global competitors that increasingly look to information and information technology for
competitive advantage. ABC is involved in this broad arena of "outsmartmanship."
ABC puts the "management" back into management reporting. It will be fun watching
organizations move from their learning stages into mastery of building and using ABC
systems.
*Consortium for Advanced Manufacturing - International, Dallas, Texas
Knowledge Management Solutions, Inc.
January 2001
69
DRAFT
(Not all Hyperlinks are Active)
Case Studies – Applications
1.0 Banc One Assesses Private Label Credit Services
by Jonathan R. Schwind, Arthur Andersen LLP, Cincinnati
One Private Label Credit Services (Bank One) is one of the nation's leading providers of
private label credit card programs offering comprehensive services to a wide range of
industry segments nationwide. As a subsidiary of Banc One Diversified Services
Corporation, Bank One has the support and backing of Banc One Corporation, the
eighth largest bank holding company in the nation with assets in excess of $140 billion.
Currently, Banc One Private Label Credit Services (PLCS) processes and manages
over 7 million card accounts for over 40 partners with combined balances in excess of
$2.4 billion, and services over 48,000 retail locations nationwide. Among private label
card services providers, Bank One is fourth in size and is the only one of the top ten
bank holding companies that also ranks in the top five in the private label credit line of
business.
Private label refers to the type of credit card that is issued at a retailer. Typically this
card can only be used at the retailer whose name appears on the card. Examples of this
type of product are department and computer store cards. They are not Visa or
MasterCards, but alternative means of offering credit to customers while strengthening
the retailer's brand recognition. One of the advantages of a private label card is the
customization of the credit offering such as deferred payments, no interest for 6 months,
or reward programs. PLCS supports the retailer by servicing all aspects of the product,
from loan origination to collection.
The ABM Implementation
The private label industry is highly competitive with low barriers to entry. One of the key
factors in a retailer's decision to select a private label provider is price. Price is not only
significant in the acquisition of clients but also in the retention of clients. With increased
competition, PLCS finds it critical to understand what activities are draining the most on
internal resources, as well as to understand the profitability of their 40 plus existing
clients. According to Kay Glover of Banc One, "A deeper look into the operational
expense revealed that the resource consumption was not proportional by client."
Earlier attempts to "cost out" internal activities were performed; however, the
methodology used was inconsistent, and the confidence level of the results was low. It
became clear that PLCS needed a proven methodology to achieve their objectives.
Therefore, Arthur Andersen implemented a full ABC study. Upon completion, the study
was to provide detailed unit costs of all activities performed at PLCS, as well as an
equitable distribution of costs to each client's profit and loss report.
Arthur Andersen began the ABC project with a pilot in two operational departments at
PLCS. A diverse steering committee comprised of COO, CFO, Controller, and VP of
Expense Management was created to drive the pilot's success. This ensured that all
Knowledge Management Solutions, Inc.
January 2001
70
DRAFT
(Not all Hyperlinks are Active)
major areas in the company were represented and that there would be successful
communication of the project's significance. It was determined by the steering
committee that it would be optimal to test the ABC methodology on just two areas
before taking ABC company-wide. The pilot was scheduled and successfully completed
within four months. The project was spearheaded within the finance division with heavy
involvement from the operations area. A full-time PLCS expense analyst was solely
dedicated to the initiative, while part-time PLCS resources were used on an as-needed
basis.
With the help of Arthur Andersen, PLCS chose software from ABC Technologies and
began to move forward with the study. Some of the driver and performance measure
data was provided by the departments involved in the pilot. Additional data was
collected by manually tabulating historical reports and running custom queries off the
transaction processing system (Triumph). PLCS wanted a non-proprietary ABC
software package and chose Oros software because it worked well with their existing
system needs. At the end of the pilot, PLCS received a list of unit costs, including the
activities that made up the costs and an allocation of expenses to the existing client
base. Such activities included: taking/transferring customer calls, waiting on calls,
processing paperwork, adjusting/monitoring accounts, assisting other reps, and
meetings.
The results were intriguing, and the identification of activities, with the corresponding
expense, exposed process flows that could be improved or eliminated. In addition, they
provided a tool for analyzing value added vs. non-value added activities. Prior to the
pilot, it was conceivable that the customer service reps were spending 90% of their time
on the phone. However, the activity analysis showed that they were performing several
other tasks while driving the time spent on the phone well below 90%. Under the ABC
methodology, PLCS identified which clients were drawing more heavily on their
resources. Because of these results, it was evident that PLCS needed to follow through
with the company-wide study.
Next, PLCS and Arthur Andersen implemented ABC into the Collections department,
which employs the largest number of people at PLCS (over 300 full-time equivalents).
This department posed a challenge because it was comprised of several different and
sometimes independent functional areas such as bankruptcy, recovery, skip trace,
agency, and front-end and back-end collections, all of which rolled up into the same
cost center. Each area in essence was its own cost study.
Upon completion of the Collections department, PLCS continued to implement the ABC
methodology throughout the organization such as credit processing, merchant
operations, shipping and receiving, fraud detection, marketing, finance, accounting,
sales, senior administration, and risk management. After this point, Arthur Andersen
facilitated phases of the model development and periodic quality assurance checkups.
The project's duration amounted to nine months of total work, which included defining
the project approach, conducting the interviews, building the model, validating the
findings, and generating the appropriate reports.
Knowledge Management Solutions, Inc.
January 2001
71
DRAFT
(Not all Hyperlinks are Active)
The Results
Due to the thorough interview process and the flexibility of the software, PLCS identified
comprehensive reporting on the costs of all their internal and external activities. Part of
the reason that the reporting was so robust was because PLCS categorized each type
of expense.
Examples of expense categories were internal, external, pass-thru, and processing. The
end result was a fully loaded unit cost, which could be broken down by any expense
category. The costs were also tagged with fixed and variable labels. This is very useful
when analyzing the impact of additional volume or lost clients. An example of this is
when the customer service department would go to create the budget for the next year.
If ABC says the totally loaded cost for a phone call is $3.00, then he could multiply
$3.00 by the number of forecasted phone calls to get the total budget amount for the
whole department. The $3.00 call includes fixed cost, which will not increase
proportionately with the call volume. Without "tagging" the expenses, it would be much
more difficult to assess volume-related impacts on expenses. Some of the specific costs
or performance measures that PLCS wanted to achieve were a cost per customer
phone call, faxed credit application, 30 day delinquent account, transaction,
correspondence letter, and a dispute, to name only a few.
The objective for such a list of performance measures was to create a menu of costs.
When it came time to re-price existing clients or make bids on potential clients, PLCS
felt confident in the customized expense numbers for each different scenario, as
opposed to using a generic approach. The menu of costs can also be used to create
value and savings for the client. For example, suppose a client has two vehicles with
which they can submit credit applications, while being charged $3.00 per application
regardless of type. PLCS informs the client (after ABC) that application type A costs
$1.00 and application type B costs $4.00. This gives the client choices and an
opportunity to increase profits by shifting volume to type A, which will ultimately solidify
the "client/provider relationship." A win-win situation is created simply from better
understanding the business.
Lessons Learned
By far, the most valuable lesson learned was that it is critical that senior management
support the initiative from the start. Getting the information from the employees is one of
the key inputs into the model. The accuracy of the data that they provide early in the
project equates to better cost information in the end.
Similarly, when driving cost to the client level, it is equally important to use allocations
that have integrity. At times the optimal driver cannot be obtained, and rather than
guessing, it is better to drop a level and use the next best driver. For example, PLCS did
not have the number of correspondence letters answered by the client. However, nearly
every customer who writes a letter also places a phone call. In this scenario, it was
feasible to use the number of phone calls by client as the next best driver for the
correspondence expense.
Knowledge Management Solutions, Inc.
January 2001
72
DRAFT
(Not all Hyperlinks are Active)
The review and validation stage can also be challenging. Everyone wants the expenses
to be someone else's. Keep in mind that the project started with 100% of expenses and
has to end with 100%. But, the expense has to go somewhere, and trying to force the
expense to one unit or product jeopardizes the accuracy of the entire project. This
scenario can be seen when allocating overhead. Overhead is a necessary and real
expense, which should be allocated to all activities. Heavily loading select activities in
efforts to make the cost of other activities look better is not recommended.
Several of the ABC results were not what PLCS had expected, which led them to think
that the costs were suspect. This is a common mistake. If the integrity was evident in
the interview sessions and in the input data, then chances are that the end results aren't
too far off. Surprise costs are evidence of a good opportunity to improve or modify the
process. PLCS is in the process of implementing activity-based management (ABM)
which will update the model when the organization changes show the immediate impact
on the costs and client allocations. ABM will also enable PLCS to continually go through
the process improvement steps as the results are analyzed. PLCS facilitate the need for
ABC as they follow the right path of using the results to make change and decisions.
Knowledge Management Solutions, Inc.
January 2001
73
DRAFT
(Not all Hyperlinks are Active)
2.0
Mobil Corporation
ABC Gathers Speed at Mobil by Tom Kang, Mobile Corporation
Imagine a tiny snowball slowly rolling down a hill, becoming bigger and bigger, gaining
more and more momentum, until it is large, visible, and omnipresent. That would be
somewhat like the development of ABC as a concept at Mobil. In the beginning, ABC
was a small initiative in one affiliate of Mobil's downstream divisions (Marketing and
Refining). Today, the ABC scope inside Mobil extends over sixty affiliates worldwide
and is expanding to the other business units.
Current scope of ABC at Mobil
Lubricants are one of the highest profit margin areas of Mobil's downstream business.
This product category includes items that are readily recognizable to retail consumers,
such as Mobil 1 passenger vehicle lubricant, and thou-sands of other industrial and
commercial grade lubricants. Uncharacteristically complex, the lubricant business is an
exception to the conventional pattern of the few products/high volume oil industry. Mobil
is an integrated supplier of lubricants, meaning that it possesses the technology and
infrastructure for producing lubricants from base components (such as crude oil and
chemical additives) and delivering finished products to end-customers. There are over
30,000 lubricant stock keeping units (SKUs) within the Mobil Network with multiple raw
material supply, manufacturing, and distribution centers.
Mobil operates lubricant businesses (called affiliates) in more than sixty countries with
forty manufacturing locations (called Lube Oil Blend Plants, or LOBPs). Each country
operates as an individual affiliate. These individual entities are loosely collected into
regional groupings, which in turn sum to the global business. The principle business
metrics are at the local affiliate level; profit and cost optimization occurs at this level.
This business is within the scope of ABC at Mobil today.
The first ABC project at Mobil and its results ABC's start at Mobil can be traced to the
US Lubricants Business in 1994, when its General Manager George Madden,
recognizing that the US business had thousands of SKUs and multiple manufacturing
facilities, suspected that the high degree of complexity was adding cost to his business.
He wanted to know the cost of the complexity, that is, the return for maintaining such a
complex business.
Consulting firm A.D. Little was hired to help analyze the situation. Focusing at first on
analyzing manufacturing complexity (the product dimension), the joint Mobil and A.D.
Little team quickly realized that the current cost accounting system had no way of
reflecting the cost of complexity, because it treated all products equally. Stated simply,
manufacturing costs of $80MM were volumetrically applied to the 250MM gallons of
lubricants which were manufactured. This method could not address cost to produce
each product individually. At this point, A.D. Little suggested using ABC methodology to
better reflect the real costs of producing the multiple product and package combinations
Knowledge Management Solutions, Inc.
January 2001
74
DRAFT
(Not all Hyperlinks are Active)
within Mobil's Lubricant Network.
ABC proved to be the perfect tool for identifying the cost of complexity in Mobil's
Lubricant Business. By using activity drivers instead of volume drivers, ABC could
expose fundamental differences in the costs of producing thousands of product and
package combinations.
In the past, when all costs were volumetrically applied to products, marketers added
products with a high degree of manufacturing complexity to the product line by
evaluating their economics on an incremental cost basis, believing that any new
volume, regardless of production complexities, would decrease the unit cost. By
applying costs using activity drivers instead, ABC exposed the weakness of the old
school of thought.
The results were startling. They created a stir within the organization and were the start
of a culture change. In surprisingly short order, the incremental volume/cost theory gave
way to activity-based full cost theory. The view of manufacturing costs shifted from one
pole to another within a matter of months!
Management liked what they saw and wanted to incorporate ABC into everyday
business decision making processes. At this point, management took a risk. Without
fully understanding the impact this new information would have on the decision makers
in the business, they decided to incorporate ABC into the single most important tool
used to manage the business then and now: the profitability reporting tool (called the
Lube Profitability Report, or LPR). LPR contained the data used to evaluate all
customer/product profitability. Every sales person had access to it. The impact of ABC
was to change the view of profitability of all products within LPR. LPR no longer
reflected or rewarded indiscriminate volume growth. Enhanced with ABC, LPR now
rewards profitable growth or divestment!
Management's risk paid off. In the years since adopting ABC into LPR, every business
indicator has become more positive—profits, ROCE, manufacturing expense reduction.
The only business indicator with a negative trend is volume, which decreased slightly
during this period. A few years ago, this would have been anathema. Our principle
metric had been volume growth for such a long time that some still have trouble
comprehending the change, and some still manage the business primarily to optimize
this metric. But an irrevocable culture change is underway. Profit focus and business
simplicity are the themes, and the business results are their tangible manifestations. It
would be incorrect to assume that ABC was the only factor contributing to the positive
business results, however. A multitude of other initiatives along with ABC contributed to
the results significantly, for example the closing of multiple manufacturing facilities with
excess capacity. The impact of ABC must not, however, be underestimated either. In
many instances, ABC served as a catalyst for decisions that would not have been made
in prior years, including divesting of large volume/unprofitable businesses, reducing
product line complexity, focusing on target segments, and so on. This was the genesis
of ABC at Mobil.
Knowledge Management Solutions, Inc.
January 2001
75
DRAFT
(Not all Hyperlinks are Active)
Mobil's globalization of ABC
After the end of Mobil's 1996 fiscal year, Chairman Lou Noto announced a corporatewide strategic goal to become Number 1 (most profitable) in lubricants worldwide. To
that end, the World Wide Lubricant Strategy Study (WWLSS) was commissioned with
the goal of achieving a 33% increase in after tax profits by the year 2000. The WWLSS
recommended three broad strategies: growth, cost reduction, and a stronger
competitive position. These strategies, while not in themselves revolutionary, set the
tone for Mobil's Lubricant Business to begin developing tactical initiatives that would
support these strategies. While growth and cost reduction were strategies that were
very familiar to Mobil employees, the third strategy strengthening competitive position
was not. To achieve this, they needed to leverage Mobil's global nature by identifying
and sharing internal best practices used in the business around the world and creating a
more efficient global manufacturing and marketing organization.
ABC was one of the key best practices identified by the WWLSS, based on the track
record set by the US affiliate, which had already implemented ABC and increased
profits through improved business decision making and a change in culture. WWLSS
suggested this best practice be shared throughout the Mobil Lubricant Network with the
intent to produce beneficial results similar to those achieved in the US business.
Subsequently, a team was chartered to develop a global ABC model template and
determine the best method to implement ABC throughout the multiple affiliates. Late in
1996, a team comprised of members from multiple global regions met in Brussels to
develop a global model template and develop an implementation plan with
costs/benefits and a timeline. The team also evaluated software. They selected Oros,
for several reasons: Mobil's US experience with Oros; the recommendation of A.D. Little
Consulting; and the features and performance characteristics of Oros. Two initial pilot
sites were chosen: Mobil Korea and Mobil Brazil.
The ABC pilot implementations
Korea and Brazil were chosen because of the size of their businesses in terms of
volume/revenue and complexity. Both affiliates were medium sized with moderate
business dimensional complexity (that is, a manageable number of market segments,
customers, and products). The global ABC model template was divided into two parts:
Sales & Marketing and Manufacturing. The two parts would be brought together into
one model to enable dimensional analysis.
For the initial pilots, the decision was made to implement the manufacturing part only in
Korea and Brazil. At that time, ABC was still mainly perceived as a product costing tool,
and the implementation team concluded that driver quantities would be easier to obtain
and validate (or defend) in manufacturing. We hoped to leverage the success of
manufacturing ABC to promote the Sales & Marketing portion. In retrospect, this was a
good decision.
These pilots were conducted concurrently in late February 1997. Both were one month
long and involved fifteen hour days, conquering language barriers, flying tens of
Knowledge Management Solutions, Inc.
January 2001
76
DRAFT
(Not all Hyperlinks are Active)
thousand of miles between Sao Paulo and Seoul, and both were hugely successful.
Korea proved to be the most dramatic example of ABC's success. The affiliate was
wrestling with a decision to divest a large portion of their business because of its
apparent unprofitability. The traditional cost accounting that applied all manufacturing
costs to products volumetrically suggested that a Marine Lubricant comprising nearly
20% of the total affiliate volume was losing money. In lubricant manufacturing, however,
there is tremendous variability in the effort, time, and capital required to manufacture
equivalent volumes of various products. Activity based analysis revealed that the Marine
Lubricant was among the simplest and least expensive to manufacture on a unit basis.
The existing cost accounting overstated production costs for Marine Lubricants by
300%! This was a real eye-opener as it corroborated the intuition that Marine business
was profitable in Korea. The affiliate had grown in volume in the last few years mostly in
the Marine product segment—and production unit costs had decreased, and profits had
increased nearly 100% over that period. ABC analysis helped explain these results.
What would have happened to cost/profit trends if the affiliate had grown in the most
complex, expensive to manufacture products and segments? A similar relationship
between manufacturing complexity and costs was seen in the other pilot as well. Figure
1 reveals the distortion which traditional cost accounting introduced in the costs to
produce various products.
As the tables show, the high volume, simple to produce products were subsidizing the
low volume, complex products. Costs and profits were incorrectly reflected, thus
distorting information and leading to incorrect decision making. When confronted with
this new data, management asked the obvious next question regarding Sales &
Marketing costs. Since all costs, excluding raw material costs, were volumetrically
applied, it was reasonable to theorize that other costs were similarly distorted. What
would the results be if costs related to customers and market segments were applied
using activity based methodology?
To answer this question, the initial manufacturing ABC pilots were extended to include
Sales & Marketing costs. The results were equally revealing. Two new dimensions
customers and market segments were added to the existing product dimension. This
allowed a multidimensional view of the business with the dimensional hierarchy
following the order of market segments, customers, and products. This analysis opened
up a Pandora's box at Mobil. Traditionally a marketing controlled organization,
marketing costs were not clearly identified and were assumed as fixed. By analyzing
activities and their contribution to the market segment and customer dimensions,
marketing expenses now became visible and identifiable. With visibility came
accountability, and this was not necessarily welcome. As can be imagined, this created
some resistance to ABC; some of which still exists today. Moreover, having activity
information regarding Sales & Marketing at Mobil is still relatively new, and we are
learning about the best ways to use the information. In the interim, we use the cost
information to evaluate the profitability of multiple profit dimensions. To do this, we
developed a common global profit value chain using Oros to establish a common
Knowledge Management Solutions, Inc.
January 2001
77
DRAFT
(Not all Hyperlinks are Active)
business language and develop a baseline for business discussions. (These replace the
numbers reconciliation discussions which were prevalent before the adoption of the
common profit value chain). Figure 2 shows the global profit value chain.
Our current ambition is to help the business to understand costs and profits using the
value chain and ABC methodology. When we begin using not only cost but also activity
information, we will have migrated to activity-based management, the phase beyond
activity-based costing. This advanced point is one which we have yet to reach.
Current implementation progress
The scope of Mobil's ABC initiative includes forty manufacturing locations and sixty
Sales & Marketing affiliates (23 major affiliates and 37 minor affilitates). The project
began in February 1997.
By June 1998, ABC had been introduced to 32 manufacturing locations and twenty-two
Sales & Marketing affiliates.
By year end 1998 (less than two years after beginning the project), the plan is to have
introduced ABC to all forty manufacturing locations and implemented Sales & Marketing
ABC at the 23 major affiliates.
By year end 1999, ABC will be implemented at the remaining smaller marketing
affiliates in a cluster approach. During this time frame, the focus will shift from
implementation to communication and ongoing use of ABC.
To get to the next level, tremendous challenges must be faced in the manufacturing and
marketing affiliates that have begun implementing ABC. A critical period in an ABC
implementation occurs during the lull after the initial excitement of the tool's introduction.
Once ABC is introduced, an all too human tendency is to put it aside, because it is an
unfamiliar and not yet natural methodology. If this is allowed, ABC becomes a onetime
exercise and most of the initial implementation work is wasted. During this vulnerable
period, continual focus must be maintained. Critical in this is communicating the steps
(and the resources) required to complete the phases of ABC implementation required to
make a real and continuing difference which can impact an organization's P&L (See
Figure 3).
ABC is not a one-time, static exercise. It is an evolutionary process which requires
effort, time, and patience. Many organizations don't get past Phase 1 of ABC
implementation, primarily because they lack continued focus and resource commitment.
Only organizations with long term focus, patience, and senior management commitment
are likely to make it to the latter phases of ABC.
At Mobil, there is a normal distribution of affiliates at different phases of ABC
implementation. Most affiliates have migrated to Phase 2; some are still mired in Phase
1 with very little progress; and a few have progressed to Phase 3 and are actively using
ABC to make everyday business decisions. The principle differences between affiliates
Knowledge Management Solutions, Inc.
January 2001
78
DRAFT
(Not all Hyperlinks are Active)
at Phase 1 and Phase 3 are clear—local management buy-in and resource
commitment.
Expected benefits
The expected benefits of ABC can be defined from both local and regional/global
perspectives.
Local benefits
•
•
•
•
Identification of (non) profitable market segments/customers/ products.
Identification of process costs (enable outsource/capital investment/
divest/pricing decisions).
Reproducible activity based business scenario building using Oros.
Inter-department communication/ cooperation leading to better (congruent)
strategy building "actions yield predicted reactions."
Regional/global benefits
•
•
•
•
Meaningful benchmarking/best practice sharing between affiliates now "applesto-apples comparisons" due to adoption of common ABC methodology and
common profitability evaluation.
Global manufacturing network optimization meaningful product sourcing from the
40 LOBPs with potential capacity reduction.
Better inter-affiliate communication/cooperation.
Better regional/global strategy building
Like most corporate initiatives, the desired ultimate outcome of ABC is improved profits.
But ABC is only information which can be used to make decisions which can potentially
improve the bottom line. It is not a panacea nor a source of magical answers that
translate to profits. Without good people who understand both ABC and the business to
make good decisions, ABC is only an intellectually diverting accounting exercise. When
used correctly, though, it is a powerful tool that highlights business improvement
opportunities. ABC has the potential to sharpen and focus the strategic direction of an
organization by providing insights into its dimensions of profitability.
Activity-based management (ABM) flows from a deep level of an organization's internal
understanding of activity based concepts. Through analysis of activity drivers, ABM
provides insight into why an organization's dimensions of profitability are profitable (or
unprofitable). ABM provides decision makers with a lever to change and improve a
business by managing the level of activities. Solid understanding of processes, costs
associated with them, and returns from these processes is a precursor to making better
business decisions, with positive impact on the bottom line. Without activity analysis, an
organization only sees in itself what is captured in the ledger and no deeper, and hence
is not able to fully comprehend its strengths and weaknesses or the reasons for such.
Knowledge Management Solutions, Inc.
January 2001
79
DRAFT
(Not all Hyperlinks are Active)
ABM is a long-term goal at Mobil. We believe that ABC will logically lead to ABM
because the changes in business dimension profitability will lead to the reasons why.
Using ABC an organization can build a baseline of its business by getting its core
processes under control. Once that line in the sand is drawn, improvement opportunities
can be seized and exploited. More over, by sustaining ABC, subsequent improvement
initiatives can be measured.
The quality of the results from an ABC initiative depend a great deal on the decision
makers. As with any tool, the skill of the hands on the tool determine whether the results
are ordinary or elegant. In the hands of a novice, a Stradivarius is only a violin, and the
music it makes is not much better than any other violin. A Stradivarius in the hands of a
master, however, produces sounds of unmatched quality. The ABC potential is great.
To actualize ABC's potential lies with its users. We firmly believe ABC will help us reach
our target of a 33% increase in after tax profits by the year 2000.
Lessons learned
Here are three important lessons learned from this global initiative:
•
•
•
Manage expectations
Evaluate local systems information prior to implementation
Involve people with the correct skill sets to implement and sustain ABC
Managing expectations is critical to keep the ABC momentum going after initial
implemetation. ABC takes real effort to establish and sustain within an organization. We
believe that the cycle of ABC introduction to its evolution to ABM is no less than a twoyear proposition, and this involves concerted effort within these two years. This is not to
say that information from ABC can't be used before the cycle is complete. On the
contrary, there will be many uses of ABC information in the interim. A static point in time
ABC analysis of the dimension of profitability of an organization is very useful and can
have major strategic implications. Many organizations use ABC to redirect their current
strategic focus. But to fully leverage ABC/M and institutionalize it into the business
culture of an organization, the ABC/M cycle must be allowed to run its full course. For
this to happen, there must be resources within the organization focused on
implementing, validating, streamlining, automating, and communicating ABC
information. Depending upon the size and culture of an organization, the resources
required vary. Within Mobil, managing expectations has been a major challenge. One
widespread but unrealistic belief is that ABC is a "plug-in-and-go" solution. Continual
communication is slowly eroding this belief, and resources are aligning to support ABC.
A recurring question asked when resources are needed and justification is required is
"What has ABC done for me so far?" When the answer is that ABC is a process that will
yield the desired benefits only if the proper resources are assigned to it, the
Knowledge Management Solutions, Inc.
January 2001
80
DRAFT
(Not all Hyperlinks are Active)
conversation has the potential to become a circular argument very quickly. To break out
of this closed loop, there must be a small "leap of faith." Examples of small victories or
successes of ABC both within and outside the organization can help. These examples
can also help allay management's impatience to get results. This statement may not be
culturally or organization behaviorally correct, but top management commitment and a
top down driving of ABC initiative seems to be the most effective method to implement
ABC and ensure its sustainability. This has certainly been the case at Mobil.
Another critical barrier to ABC implementation has been data integrity and availability
from information systems. ABC is not and should not be a solution to fix data integrity
from poor information systems. ABC is only as good as its inputs. Many affiliates within
Mobil have looked to ABC to somehow correct data integrity problems existing within
their present information systems. This expectation is incorrect on two accounts. First,
ABC is not capable of doing this on a sustainable basis: this requires manpower and
intensive data manipulation. Second, this doesn't resolve the real problem of data
integrity and availability from existing information systems. In every ABC implementation
at Mobil, our motto has been "We'll build the best ABC model with the data available
from existing information systems." The pilot implementations taught us that our task
was not to re-engineer data resident in systems such as JDE and SAP. This in itself is a
substantial task. We realized that we would never meet our objectives and goals if we
attempted to do this. In instances which data integrity, availability, or accessibility was a
problem, we identified this to the local management and built ABC solutions capable of
using the data when it became integral or available. The identification of data problems
became an unanticipated outcome of the ABC implementation process. The Mobil ABC
model requires fundamental data from information systems data which must be
available independent of ABC. This data is generally scattered and its integrity is not
often validated. ABC brings many of these key business data elements into a business
profitability model. To balance a profitability model, the data elements must match in
assumptions and timing and be accurate. This was a persistent problem throughout the
Mobil network. An unanticipated outcome of global ABC implementation was the
identification of the quality and usefulness of information systems in each affiliate.
At Mobil, we've learned that ABC is not easy. There are many elements such as
management support, good data, and ample resources that are prerequisites to
successful implementations. Even if all of the prerequisites are in place, ABC is
susceptible to failure if the local implementation team does not have the technical and
communication skills required to build an ABC model and effectively communicate the
results. Building a reproducible model in Oros requires substantial database
management skills, keen knowledge of the business, and good working knowledge of
the software. It is critical that local personnel be trained for these skills. Moreover, the
group with ownership of ABC must have visibility as well as credibility within the
organization. ABC must be communicated to the business. If this is not done, it
becomes an isolated tool used only by a select few with access.
The more people with access to ABC information, the better the buy-in and the deeper
the ultimate impact ABC can have on an organization. This doesn't mean that everyone
Knowledge Management Solutions, Inc.
January 2001
81
DRAFT
(Not all Hyperlinks are Active)
should have access to the ABC model. Within the business, however, people should
have access to information which they can impact. An important part of ABC is
information collected from field personnel. So that they may provide the best input, they
must be trained to understand the impact of their input. The best way to do this is to
show them the output. The output will be challenged, but this is a healthy dialogue that
can build a channel of communication that will ultimately help in developing a more
accurate ABC model with better quality input.
Next steps and conclusion
The next steps are to continue implementations to new sites, support existing
implementations and help them progress to latter stages of ABC, support the global
lubricant manufacturing network optimization study, develop ABC best practice
networks/databases, continue to train users, share the ABC experience with other
business units within Mobil, and tie ABC to a performance measurement system.
One of the exciting next steps we are exploring is using ABC information in our
performance measurement system, the balanced scorecard. Within Mobil, the balanced
scorecard is "the Bible" and the basis uponwhich variable pay (bonuses) is
administered.
By using ABC driver information as performance measures and a behavior modification
tool, we hope to link ABC to strategy. This will have the effect of allowing businesses to
visibly manage the levers (activity drivers) which translate to positive business
outcomes—increased competitiveness and profit. If properly incorporated, ABC will be
used at the operational mid-management level, and also at the strategic senior
management level. If we can do this, ABC's place within Mobil will be comfortably
assured. If senior management begins to view the strategic initiatives from an activity
based perspective and begins to ask for activity-based information on a regular basis,
ABC will transform from being a tool which is viewed as new or innovative, into one that
is the norm; a tool which is an essential component to business decision making. When
this occurs, the ABC/M life-cycle will have run its course and will be ready to evolve to
more sophisticated uses. Some of us at Mobil are working to ensure that we get to this
inevitable point before our competitors. We are confident that we will succeed.
Tom Kang is the Global Planning/ABC-ABM Support Manager for Mobil Corp. Mr.
Kang's current responsibilities include managing the global implementation of ABC to
more than 40 affiliates, while developing and applying the common profit value chain.
Mr. Kang holds degrees in both Chemistry and Business Administration.
http://www.bettermanagement.com/abcauthority/library/case_studies/oil_gas/#tomkang
Knowledge Management Solutions, Inc.
January 2001
82
DRAFT
(Not all Hyperlinks are Active)
3.0 U.S. Airways – Implementation Lands $4.3M in Process
Improvement Savings
by Joe Donnelly and Dave Buchanan
US Airways is one of the world’s leading air carriers and one of its largest. It is also
one of the largest U.S. airlines and holds the dominant market position in the eastern
U.S. With about 42,000 employees, the company operates more than 2,000 mainline
flights daily, serving more than 100 airports in the U.S., Canada, Mexico, the Caribbean
and Gulf of Mexico, Germany, Italy, France, and Spain.
The US Airways system also includes the US Airways Shuttle and US Airways Express,
and the system as a whole operates more than 4,500 daily departures to more than 200
airports. The US Airways Powerplant Department, or aircraft engine shop, at the
Pittsburgh International Airport, maintains and overhauls 350 jet engines powering 152
of US Airways’ 380 aircraft. When an engine arrives at the US Airways Powerplant
Department for maintenance, it can have as many as 11 separate modules, which can
be repaired or overhauled. In-house repair and overhaul capabilities focus on the seven
model types of the Pratt & Whitney JT8 engine family. The engine shop operates 24
hours a day, five days a week.
There are about 500 employees in the engine shop. The work force is divided into 24
separate, self-directed work teams. The majority of the powerplant employees are
represented by the International Association of Machinists (IAM). US Airways and the
IAM implemented a High Performance Work Organization initiative whereby they
establish a collaborative work environment with common goals. Each work team
develops its own charter and selects a team leader. Team members actively manage
their work areas and identify opportunities for improvement. Understanding the
Business Needs, US Airways needed detailed cost information with particular focus on
engine overhaul costs. The lack of detailed operational and financial information did not
allow management to fully understand the costs associated with producing or
overhauling an engine and, to a lesser extent, the costs of each of the modules making
up an engine. This led management to ask other key questions which would enable
them to determine the business solution needed to provide this information. Some of the
key questions included:
•
•
•
•
•
•
What are US Airways’ critical business issues;
What industry trends are driving these issues;
What is the strategy to meet the needs of the changing business environment;
What operational and financial information would provide decision support to
meet these business needs;
What level of detail is needed to manage; and,
What frequency of reporting is needed?
Knowledge Management Solutions, Inc.
January 2001
83
DRAFT
(Not all Hyperlinks are Active)
The answer to each of these questions would provide insight to the action needed to
deliver the necessary information. US Airways determined that more detailed and
insightful information was needed to manage the business, drive improvement, manage
costs, and support third-party pricing. The business solution would have to support each
of these needs.
Business Solution
Activity-Based Cost Management (ABCM) was the business solution chosen for many
reasons. ABCM could not only help determine the true cost of engine maintenance, but
could also provide operational and financial information to be used by the self-directed
work teams to identify opportunities for improvement. ABCM would also be the key
enabler for US Airways to better understand the following:
•
•
•
•
Repair in stock vs. buy new—inventory decision making;
True cost of operations to support third party pricing;
Operational metrics for improvement and bench-marking; and,
Impact of improvements made by self-directed work teams.
An ABC team was formed at US Airways to implement ABCM in the powerplant
department. The team was composed of two full-time US Airways employees from the
engine shop, a US Airways financial staff member, and two Arthur Andersen members.
Project Approach
The project was divided into four main phases. The phases consisted of introduction to
ABC, data collection and information gathering, model building, and data analysis and
reporting.
Introduction to ABC Phase
To ensure a common starting point, several meetings were held with all employees to
explain the ABC project, why it was being undertaken, and the insights it would provide.
These initial meetings served to achieve employee buy-in and support. Subsequent
meetings were held with team leaders and representative members of each of the 24
work teams who were most familiar with the engine shop processes and who could
describe the objectives and the benefits of the initiative.
Data Collection and Information Gathering
One of the initial steps in the data collection phase was to understand the level of cost
object detail to be costed. By understanding the level of detail to be costed, the team
could gain insight into the types of activities and the level of detail to be collected and
measured. Once the products to be costed were understood, the team began to gather
specific cost pool information. The team determined it was necessary to collect and
Knowledge Management Solutions, Inc.
January 2001
84
DRAFT
(Not all Hyperlinks are Active)
include the cost of rent, utilities, and benefits, which were paid by the corporate
headquarters, in order to determine the true cost of the powerplant department. Costs
were also collected for other areas known as the periphery groups which include
Warranty, Stores, Training, and Purchasing. These are separate groups within US
Airways who operate their own budgets and perform services not only for the
powerplant department, but for other areas as well. These costs were also important to
include in the analysis in order to understand the true cost of the powerplant
department.
Next, activities were collected from each of the work teams, including core and non-core
activities, such as tear-down, welding, waiting for tooling, and rework. The diversity of
the work force and layout of work areas into teams made collecting the necessary
operational and financial information challenging for the team. The final activity
dictionary included 410 activities across the engine shop, including 47 non-value added
activities. Representatives from each of the 24 self-directed work teams were then
interviewed to determine how their effort was distributed across their team specific
activities. The activities were captured on team-specific spreadsheets and populated by
the representatives based on percentage of effort. In addition, the spreadsheet design
captured any cost diversity between the different engine types, which illustrated if
certain engine model types required more effort for a certain activity. To assist the
employees in completing the effort spreadsheets, a time conversion table was
developed to convert the actual time spent on activities into percentages. The periphery
groups who performed services for the powerplant department were also included in
this process in order to determine the number of employees dedicated to powerplant
tasks and the associated effort spent on those tasks. The team then attributed each of
the activities as either core, support, contractual, or irrelevant activities to understand
how the power-plant was investing its time and effort.
Across all four phases of the project, project management controls were used to assist
in defining the scope of the tasks and progress being made in completing them. These
controls were essential considering activity information was needed for 500 people
working three shifts, with 12 separate classifications of employees such as mechanics,
welders, machinists, and inspectors—most with different labor rates.
Obtaining the final cost and headcount per process was complicated as teams of people
were dedicated to many different processes. A headcount control sheet allowed the cost
management team to identify the true headcount associated with each of the processes.
Another project control included a progress chart. This chart detailed each task the
team needed to be accomplished for each of the 24 work teams. As the tasks were
completed, the color-coded chart was updated. Using visual controls, any member
could quickly see what remained to be accomplished as well as the scope of the
remaining tasks.
Model Building Using Oros
Building the Model using the Enterprise Pack software, including OROS ABCPlus, EIS,
and the Links Engine, the US Airways ABC team constructed its entire model with
Knowledge Management Solutions, Inc.
January 2001
85
DRAFT
(Not all Hyperlinks are Active)
import files. The first module to be constructed was the cost object module based on the
level of detail determined in the data collection phase. The cost object hierarchy would
accommodate the variability in levels of service for each engine type. Each engine has
different maintenance needs or service levels; the model allowed US Airways to
determine the true costs for each type of engine by each service level. The activity
module was built next. The activity dictionary, developed earlier in the data collection
phase, was easily imported into the activity module from a spreadsheet format. Finally,
the resource module was built. The resource hierarchy was also imported from a
spreadsheet format based on the level of detail determined earlier in the project. Driver
data was imported after each module structure or hierarchy was in place in the model.
The cost management team built intuitive abbreviated names and initials into its model
reference numbers to help them diagnose any problems they might have during model
building. In doing so, the team could immediately determine where specific account
dollar amounts were derived. The use of these intuitive reference numbers proved to be
invaluable during error and warning diagnosis by enabling the team to quickly pinpoint
the problem areas. Once the model was running, it provided true cost information
previously unavailable within the power-plant department.
Data Analysis and Reporting
After the model was built and the ABC data was available, the data needed to be
analyzed and reported. The team first conducted reasonableness checks to ensure the
model was assigning costs appropriately. Next, the team had to assimilate the data into
a format that could be reported and easily understood by the engine shop employees.
Report books were created for each of the 24 self-directed work teams with their
individual results as well as a view of the entire engine shop. These reports were well
received by the teams and validated their contributions during the data collection
phases. Team leaders as well as all team members now had access to operational and
financial information that would enable them to drive improvement, measure results,
and cultivate process ownership.
Project Results
The ABC model output provided US Airways with operational and financial data to
support strategic and operational decision making. The ABC information identified
process improvement savings opportunities for US Airways totaling $4.3 million per year
and 63 full-time equivalents. The model output provided numerous operational and
financial metrics which were not previously available. The self-directed work teams
could now see the activities’ cost by labor classification and by shift. In addition, a rank
order analysis for the activity costs was conducted. This view provided insight into the
most expensive non-value added activities occurring in each work team. This
information was immediately available to begin analysis for process improvement
opportunities. For example, one work team found they spent about 80% of their effort or
$85,000 per year reinspecting due to the use of an older piece of inspection equipment.
The ABC information justified the purchase of the new piece of inspection equipment,
the cost of which was less than the $85,000 spent on reinspection each year.
The team, now equipped with operational and financial information, needed to capitalize
Knowledge Management Solutions, Inc.
January 2001
86
DRAFT
(Not all Hyperlinks are Active)
on the identified opportunities in response to the business issues set forth at the
beginning of the project. The ABC model output could now be used to support the
business through:
•
•
•
•
Support of repair vs. buy inventory decisions;
Understanding the true cost of operations to support pricing;
Supporting benchmarking; and,
Providing a baseline for measuring improvements by the self-directed work
teams.
Although irrelevant activities were identified, root cause analysis would have to be
undertaken by the team to identify the causes for the effort. Upon identifying the true
cause, the company and teams will be able to eliminate or reduce the impact of nonvalue added activities. The ABC team now publishes a unit cost guide three times a
year. This guide details per unit costs for each of the seven engine types and their
modules.
Next Steps
Powerplant management made the decision to update the cost model on a trimester
basis. Therefore, every four months, headcount for each of the teams and processes
will be updated in the model. Plans also call for annual activity and effort updates via the
effort grids, which will provide progress reports on process improvements made by the
self directed work teams in the engine shop. When the engine shop undertook this ABC
project, there were a number of shortcomings in the ability to provide necessary
operational and financial data for the cost model. For example, while the company knew
how many engines it produced, there was not accurate information on how many
modules were produced. As a result, the team set up procedures to capture accurate
production by engine and by module. The ABC team now publishes a monthly engine
and module production summary for the powerplant department.
The team also found weakness in tracking for overtime costs. Many of these costs were
grouped together, which did not allow a view of what team or discipline actually incurred
the costs. The cost management team brought greater detail to these costs by using an
overtime report, also published monthly.
The need for accurate and timely data to support the cost model will drive the team to
continually validate their data frequency, source, format, and capture methods.
Communication of results is an ongoing effort. The team will need to communicate ABC
results to the people who can and will implement change.
Joe Donnelly is a Senior Manager with Arthur Andersen’s Advanced Cost Management
Team. With 15 years experience, Joe leads implementation projects for clients in
transportation, financial services, manufacturing, engineering, and philanthropic
Knowledge Management Solutions, Inc.
January 2001
87
DRAFT
(Not all Hyperlinks are Active)
organizations. Joe served as Chairman of the Capacity Management Interest Group for
the Consortium for Advanced Manufacturing-International (CAM-I), and is a member of
the ABC Technologies Affiliate Program.
Dave Buchanan serves as co-leader for US Airways cost management implementation.
With more than 12 years of experience in the airline maintenance field, his background
includes serving as a licensed aviation maintenance technician. The authors would like
to thank Eric Delegarza of Arthur Andersen’s Business Consulting practice in Atlanta for
his assistance with this article.
Knowledge Management Solutions, Inc.
January 2001
88
DRAFT
(Not all Hyperlinks are Active)
4.0 United Technologies’ Activity Based Accounting Is A
Catalyst For Success
by Robert Adams and Ray Carter, Otis Elevator
Effective activity management requires examining activities to determine their efficiency,
effectiveness, and how well they are leveraged. Activity-based accounting (ABA)
facilitates this examination on a continuous basis. In addition to activity-based costing
(ABC), Easy ABC provides a workable framework for devising effective activity-based
accounting schemes.
Emergence of Activity-Based Accounting
ABA is another step in pushing the envelope to advance activity-based management
(ABM). Once an organization has built a model, experienced some of the benefits, and
is comfortable with the methodology, it makes sense to begin publishing and tracking
the operation’s results in this manner to better engage the entire organization in ABM.
Thus, ABA is simply defined as the organizing, reporting, and tracking of information in
an activity format as well as developing product costs—standard and actual—and using
resource and activity drivers to assign overhead.
Overhead Classifications
One concept of ABM is that all overhead consumption is not the same. In addition,
some overhead or supporting manufacturing activities are more desirable than others.
Further, each company will have its own set of unique issues that drives its cost
structure and, accordingly, must customize overhead reporting to its needs. For
example, some organizations have found it useful to highlight value-added versus
nonvalue-added activities to eliminate waste.
Two subsidiaries of United Technologies are using various ABM concepts. Otis Elevator
is using the core management process improvers, problem fixers approach promoted by
Blaxill & Hout. Carrier Corporation has adopted the unit, batch, product/process
sustaining, structural format promoted by Robin Cooper and Bob Kaplan.
Cooper and Kaplan popularized this methodology for grouping overhead into the ABM
hierarchy categories of unit, batch, product sustaining, and structural. These
classifications were somewhat modified to meet the particular business needs at
Carrier, the world’s largest manufacturer of air conditioning and heating products.
However, in principle, Carrier’s methodology maintains the same concepts promoted by
Cooper and Kaplan. The distinction between these groups is based on their activity
drivers and their variability basis. Carrier uses this approach to improve operational and
strategic decisions by highlighting how cost changes with column and complexity.
Complexity costs—batch and product/process sustaining—are associated with the
structure, congestion, and flexibility of the plant. These costs rise with increased variety,
especially in a traditional manufacturing setting. Carrier defines each category in the
following way:
Knowledge Management Solutions, Inc.
January 2001
89
DRAFT
(Not all Hyperlinks are Active)
•
•
•
•
Unit. This activity or cost occurs every time a unit is produced. An example is the
utility cost for production equipment. This level of activity usually relates directly
to production volume.
Batch. This is an activity performed for each batch produced or acquired.
Examples include the movement of raw material between the stock room and a
production line or setting up a machine for a run.
Product Sustaining. This is an activity performed to maintain product designs,
processes, models, and parts. Examples include expediting parts, maintaining
the bill of materials, or product change orders. Sustaining activities are required
for supporting a key manufacturing capability or process.
Structural. These are activities performed to enable production to occur.
Examples are managing or cleaning the building. These activities are
fundamental to supporting the business entity at the most basic level. Once
Carrier customized Cooper and Kaplan’s ABM hierarchy categories to fit the
organization, we began our ABM implementation.
Chart of Activities
Extensive interviewing was performed to determine what key activities needed to be
reported at the plant manufacturing level. The results of these discussions became a
standardized set of activities and accounts that became the framework of the plant’s
activity management efforts. This "chart of activities" is now the preferred reporting tool
at carrier for any part of the organization—from the manufacturing cell to a complete
division. Essentially, this chart of activities is analogous to the traditional chart of
accounts that is common in traditional financial reporting systems. But unlike traditional
financial reporting systems, the categories were customized to fit the organization’s
needs.
At a minimum, most plants’ activities are reviewed to determine the appropriate
classification into one of the recognized activities. As a result, the activity accounting
report from an individual cell naturally rolls-up into the activity account report at the plant
level and ultimately into a complete division’s group of plants. Thus, senior management
can see the critical activity consumption at any desired level of the organization.
The initial chart of activities at Carrier is based on the previously mentioned Cooper and
Kaplan framework of unit, batch, product sustaining, and structural. These are the broad
categories for activity reporting. They’re also the most relevant for reporting factory
results at the plant of division level. A series of discussions and senior management
reviews resulted in the identification of 27 individual activities and 8 accounts.
Improving Factory Accounting
By using the standardized set of activities, it’s now possible to report the consumption of
expense at the factory in a more precise and relevant manner. Rather than report
departmental or plant expenses based on general ledger expenses, we’re now able to
report using the chart of activities and the general ledger account/activity mapping
Knowledge Management Solutions, Inc.
January 2001
90
DRAFT
(Not all Hyperlinks are Active)
determined from the prior ABM analysis. The activity mapping is simply the detailed
composition of assigned costs to an activity from the general ledger accounts. In terms
of Easy ABC, it’s simply the assignment from the resource module (general ledger
accounts) to the activity module. Once this mapping is established, period results are
readily converted into the improved factory accounting format.
Now management quickly understands in "activity terms" what happens during a
reporting period. Material handling, expediting, shipping, sustaining engineering, and
other activities are easily traced to the lowest level of the organization with the same
consistency in reporting and classification.
Prior to adopting this system, the traditional system often resulted in a garbled answer
to the most basic questions. This is particularly true as inquiries were passed through
several layers of the organization. Many times, nuances in the general ledger systems
at different plants within the same division made reporting and comparison of activities
at the unit, batch, product sustaining, and structural level very difficult.
Continuous Improvement, TQM, and ABA
Prior to the ABM implementation, Carrier used the traditional manufacturing accounting
methods, even though there was a long standing criticism that these methods lacked
relevance in a modern manufacturing environment. For example, absorption-based
accounting at Carrier encouraged overproduction, overemphasis on direct labor, and
one-sided sourcing decisions. With the use of ABA framework, carrier gained a better
understanding of the manufacturing process through a better linkage between
manufacturing processes, relevant nonfinancial performance measures, and overall
plant profitability. And with this understanding, the path toward continuous improvement
was put in place.
One of the more valid criticisms of continuous improvement and total quality
management (TQM) programs is that they’re often initiated without linkage to financial
performance. A misguided quest for improving the process and encouraging intuitively
sound manufacturing practices can lead to disastrous financial results. Even in-depth
and complete analyses do not reinforce desired behavior. This is especially true when
these analyses are supported with traditional product costing and expense accounting.
As a result, the use of traditional manufacturing accounting systems often only
perpetuates the existing sub-world-class manufacturing performance.
The revised factory accounting methodology, ABA, allows for the same degree of
control of overall plant expenses, yet reinforces the intuitively correct behavior initiated
with the ongoing continuous improvement and TQM programs. Overall department or
plant costs are controlled by monitoring the total cost of a given activity or a group of
activities. The traditional expense account detail is also maintained for the particular
activity or activities. And as a result, management is able to quickly determine
improvement opportunities for a given process.
In addition, the ABA methodology provides the information necessary to set relevant
Knowledge Management Solutions, Inc.
January 2001
91
DRAFT
(Not all Hyperlinks are Active)
non-financial measures and targets to monitor the improvement of the process. And
because the activity cost is inherently linked to expense accounts, it’s possible to
determine the financial impact resulting from the process improvement. Based on this
more detailed understanding of the process and its linkage to plant expenses, Carrier
put specific budgetary control limits in place.
Important to the continuous improvement process at Carrier is an ongoing
understanding of both the quality and quantity of manufacturing activities. Kaizen efforts
must be supported with financial analyses that accurately reflect the true cost of
manufacturing processes. The new reporting format helps focus attention on the
underlying activities.
With this increased understanding, cost recourses are improved by performing fewer
activities and/or performing them more efficiently and effectively. This focus is sure to
lead to bonafide cost improvements. Note the following example:
Purchased Parts Receiving = Number of Receipts/Cost per Receipt
Kaizen efforts are focused on reducing total receiving resources. By improving product
designs to use fewer and common parts, the number of receipts is reduced. Further,
having suppliers deliver materials presorted in ready-to-use quantities and ready-to-use
condition greatly improves the efficiency of the receiving process. The existing ABM
model can be readily modified to reflect these changes in use of resources. In each
reporting period, the purchased parts receiving activity cost is dissected to review the
progress of implementing these cost improvements. Management now has immediate
process feedback as to the success of the Kaizen activity. Moreover, this simple formula
provides a starting point for refining the Kaizen activity.
This approach is also valuable for evaluating process improvements on a pro-forma
basis. The EasyABC ABM model is readily modified to show any number of hypothetical
changes in the cost of activities, activity drivers, and driver quantities. Consequently,
plant layout, cell location, and capital purchases can all be analyzed in terms of the
expected ABM cost.
Improving Accounting Standards
Horngren and Foster define standard costs as "carefully predetermined costs that are
usually expressed on a per unit basis; they are target costs, costs that should be
attained. Standard costs help to build budgets, gauge performance, obtain product
costs, and save bookkeeping costs. Standard costs are the building blocks of a flexible
budgeting and feedback system."
As standard costs drive a number of decisions throughout the organization, the change
to activity-based costs will force the organization to address some of the prior
dysfunctional decisions. For example, marketing will have to reevaluate many product
line decisions when faced with the new margin reports. At the factory level, the strategic
make vs. buy analysis will be drastically altered when using the activity-based standard
Knowledge Management Solutions, Inc.
January 2001
92
DRAFT
(Not all Hyperlinks are Active)
costs. The supply management process will change because previously nonburden
purchased components will show significantly higher activity-based costs. As a result,
Carrier’s factories are already expanding the traditional "supplier to customer"
relationship to reduce the overall activity based costs of purchased components.
Changing the standard costs is an important step in changing the factory reporting. It
provides reinforcement for activity management and provides a clear linkage between
product costs and manufacturing activities. Consequently, the new factory reporting
provides the proper visibility to effective process management and the results will be
adequately reflected in the new activity-based standard costs.
Using Easy ABC to Set Up an ABA Framework Activity accounting schemes can be set
up in EasyABC using one of two methods. The choice of either method is dependent on
the size of the model, required response time for reporting, and personal preferences.
At Carrier, we have successfully developed models using both methods. However,
Method 1 described in the next section, is more suited for the detailed ABA framework
discussed earlier.
Method 1: Using Attributes
During the data collection process, activities are assigned with an activity-type attribute
and an activity classification attribute. For example, the purchased parts receiving crib
has an activity appropriately titled "purchased parts receiving." Using the framework
developed at Carrier, this activity gets the batch attribute (most receiving activities are
classified as batch-type activities under the Kaplan overhead analysis framework).
Then, the activity is classified as either moving material or receiving material, depending
on the specific content of the activity.
Using this technique, an organization can easily create any variety of ABA frameworks.
At Otis, we tagged with attributes all manufacturing activities as either core
management, problem fixers, or process improves. Still others have adopted a primary
and secondary activity approach. Regardless of the scheme, attributes provide a
powerful tool for activity accounting.
Method 2: Using Activity Centers & Accounts
An alternate method for developing an ABA framework is to build the accounting
structure directly into the structure of the activity module. At Carrier, unit, batch, product
sustaining, and structural activities are grouped into separate centers. Activity names
are used to identify the type of activity vs. describing the particular nature of the activity.
For example, the purchased part receiving activity described previously is simply named
"receiving material" or "moving material." Successful ABM models with detailed activity
accounting information have been developed by simply using the chart of activities and
appropriately naming activities during the interviewing process. Computationally, this
approach is much easier for the software to handle. However, the loss of the descriptive
naming does make supporting ongoing process improvement more cumbersome.
ABA Essential for Positive Change
Knowledge Management Solutions, Inc.
January 2001
93
DRAFT
(Not all Hyperlinks are Active)
EasyABC provides powerful features for developing ABA frameworks. At united
Technologies Carrier, and recently at United technologies Otis, ABA is essential in
driving effective activity management. If the objective is cost control, process
improvement, or improved performance measurement, ABA is a catalyst for success.
Robert L. Adams has over ten years of divers manufacturing and financial consulting
experience. He has worked in various capacities at General Motors, Ford Motor
Company, Coopers & Lybrand, and with United Technologies. Presently, he is the
Program Manager for the activity-based management effort at United Technologies Otis
Elevator Company—New Equipment Business. Prior to joining Otis, Bob was corporate
coordinator for a number of ABM projects at United Technologies Carrier manufacturing
facilities. In addition, he was the program manager for the implementation of the revised
factory accounting methodology at the Carrier NAO Plants. He holds engineering
degrees from Massachusetts Institute of Technology (MIT) and Purdue University, and
an MBA from the University of Michigan, Ann Arbor. Bob is a Certified Public
Accountant, a Certified Management Accountant, and a CFA candidate.
Ray Carter is the Director, Cost Improvement for Carrier’s worldwide operations. He has
held financial-related positions at Blue Cross-Blue Shield of Minnesota, Honeywell, and
United Technologies Carrier. For the past ten years, Ray has focused on using
advanced cost management techniques and effective performance measurements to
help businesses return to acceptable profit levels. Ray has worked on activity-based
management models in a variety of environments, including high-volume repetitive
production, job shop, and city government. Prior to his current assignment, he directed
the cost management activities for Carrier’s North American Operations where he used
activity-based management concepts to stimulate actions to reduce complexity-related
cost and revise the manufacturing accounting methodology. Ray holds a BA in
Accounting from Howard University and an MBA from the University of Rochester
Simons School of Business.
Knowledge Management Solutions, Inc.
January 2001
94
DRAFT
(Not all Hyperlinks are Active)
5.0
The Change is Forever: Activity-Based Costing and
Management in the United States Marine Corps
Knowledge Management Solutions, Inc.
January 2001
95
DRAFT
(Not all Hyperlinks are Active)
References
www.bettermanagement.com
Knowledge Management Solutions, Inc.
January 2001
96