USPS Product Cost Workshop
August 16, 2011
Welcome
• There have been criticisms of current costing system
• We have developed a costing systems review team
– Gathering feedback from external customers
– Gathering feedback from internal customers
– Will examine feasibility of suggestions
• Purpose - To develop recommendations to improve
cost reporting so that it will provide management the
right tools to manage the business – both in terms of
making decisions about operational changes and in
making product and pricing decisions.
1
Meeting Overview
Today’s Agenda
• Welcome
• Project Overview
• Overview of Cost Systems
• Expectations and Comments (Large Group Session)
• Break-out Sessions
• Reports from Break out Sessions
• Closing Remarks
2
Introduce Team
•
•
•
•
Project Owner: Tim O’Reilly 
Project Manager: Jennifer
Bradley 
Lean Six Sigma Master
BlackBelt: Izzy Sanchez 
Project Management Support:
Jessica Powell 
Project Team
• Karen Key (Products) 
• Josen Punnoose Products)
• Tom Henry (Operations) 
• Myles Perkes (Cost Systems) 
• Naline Hampton (Cost Studies) 
• Greg Dawson (Pricing) 
• Marc Smith (CRA) 
• Haley Eubanks (International) 
• Barry Burns, (International) 
• Jim Crawford (International)
• Jay Huntington (International)
• Michael Goldman (IT) 
• Robert Raines (IMB) 
• Brian Code (Retail)
3
Project Overview
•
•
•
•
•
Interviewed External Stakeholders – Complete
Interviewing Internal Stakeholders – In Progress
External Focus Group - Today
Feasibility Analysis of Recommendations
Recommendation of next steps – End of October
4
Purpose of Today’s Meeting
• To clarify and define customer needs
• To gain insights into the prioritization of needs of our
customers
• Feedback from this meeting is one set of inputs into
developing recommendations
– External stakeholder interviews and internal stakeholder
interviews are other two
• Lean Six Sigma Tools will be used to dissect Voice of
the Customer
• Lean Six Sigma Tool was used to take input from
interviews and develop break out session topics
5
Dissecting the Voice of the Customer
Customer
Ken Garner & Leo Rayman
Comments
Rates drive our customers
Increase in initiatives in HQ provides incentive to mail area
ROI is accuracy of cost data
accuracy and precision are the most important to us
there is no transparency on data for initiative
Some colleagues have introduced ABC, and it may have some legs
How are costs built?
Cost data is too old and has to be refreshed
Do bottoms up costing more reflective of real cost
We don’t get involved in cost details and what goes into those
Cost data should use best practices from market
Cost must be precise and accurate
cost must reflect cost of doing business
We are interested in better methodologies for costing
Need to study cost of operations
Need to study how costs are being distributed by type of mail
some type of mail cost is not correct
Some type of mail cost is being assigned incorrectly
personnel not assigned to do work can be increase cost by incorrect system assignment
Make sure postage rates are as low as possible
Rates must represent sustainability of USPS health
Show rate adjustments based on data and proper cost process
We have enough transparency of cost
The degree of transparency helps justify adjustments
Ensure we have reasonable explanation for cost items
Cost methodology must be sound
Cost explanations must be easy to understand
Costs should explain worksharing value determination
Need to have accountability for cost allocations and savings allocations
Variation in customer based required simple explanations of cost explanations
Think from the outside-in when it comes to data transparency
Bob Rosser
Cost methodology seems to be used to justify infrastructure
Not using cost system to justify what is needed
You must look for the least amount of costs and least amount of mail touch
IOCS sampling is used to justify how to distribute bloated costs
Vital that USPS looks at whole cost
Who says cost is calculated correctly?
The need to continue "mail moment" justifies cost
Cost methodology doesn’t even use latest technology available, so cost are not right anyway
Way too much fixed cost that shouldn’t be there
Current cost process is not good
Risks by themselves do not motivate mailers
Mailers have to look at total ROI, and value of spend
Mailers will reduce volume as costs go up
Some rates and costs differentials do not make sense
Worksharing type stuff don’t work
"not covering cost", look into why that is, we don’t buy it
What is all-in-there when it comes to cost?
People want to know allocated labor costs, etc
There are better ways than sampling to get cost right
Use ABC, it was well received
Motivate and incent mailers to get data
Unique piece ID, use smaller data sets and more accurate models
Old, obsolete costing methodology
Anne Daley & Ashley Lyons
Overall, happy with costs system
Pitney Bowes
no need to do overhaul, only needs basic improvements
Needs rule making improvements
You should tweak elements of system from time to time
Some products are small qtty, and may not have enough samples
System is generally good
Sampling can be updated
IMB only captures automation, can help improve but its not the end of ends
Transparency is good enough to understand make recommendations to PRC
It's an open system, we can identify what we need
Competitive products are less transparent, but understand why that is
Domestic transparency is good enough
International transparency is not good, and we have to guess on its reasonableness
Deeper discount opportunity if we had International transparency
We will like to see international data to help save more
Flats and parcels costs can get confused and not capture correctly
Basic cost system works well, transparency meets requirements
Don’t use assumptions when you have empirical data
Models don’t work well due to mail flow variety
Parcel size can drive costs that are no captured
Re-examine attribution, should not be done in isolation
We can live with current cost system
Would not advocate major overhaul needed
Cost attribution allows discrimination and flexibility, but can create issues in revenue side
Larry Buc
No overhaul needed, some pieces can be updated
SLS
IOCS cost by product works well
IOCS doesn't get to pricing model
Don’t use bottoms up costing
Expand IOCS survey methods
Expand IOCS more than anything else
Cost avoidance models "bottoms up" and replace assumptions with real data
Bottoms up is very hard due to differing plant layouts
IOCS will resolve bottoms up defenders
Regulators stop you from changing wrong models
What are regulators more likely to accept vs not in cost models?
PRC will not accept volume variability
PRC will not accept pricing elasticity
Attribution levels can be seen as exploiting monopoly
Bottoms up costing should not replace IOCS
Lesson learned is that mail processing is too complicated
Mods data is not reliable to get unit costs
What would you do with manual ops for bottoms up costing?
How to allocate non-handling in bottoms up costing?
IMB will show how mail flows through but not cost
Mail movement will not be captured by IMB
Bottoms up costing used for studies, but not costing
Bottoms up tends to be under-inclusive
IOCS has understood properties and withstood time and test
Need more sampling on IOCS
Need more data and less assumptions in IOCS
Where IOCS cant work, use bottoms up pricing
Cost avoidance models capture workshare savings accurately
Need to get the data for volume variability
Do not throw the whole thing away
People with criticism just don’t know how you do things, they don’t have a clue
Tom Underkoffler
Everything needs to be measured to find optimal times
Medco
All costs should be based on measurement system
Costs should use time studies for how long things should take
If cost is higher than expected, it should be investigated
We present mail very detailed, others don't and we subsidize them
We don’t know building blocks of cost
We have accuracy of costing concerns
Workshare is too high for benefits received
Not having cost building blocks prevent us from reducing costs
High concern with how data is compiled
Using averages and not knowing the reality of real data
Worry more about treatment of data vs accuracy of data
Use tracking diagnostics to see true costs
Not everyone is paying their fair share
We are taking on new work and still pricing is going up
Costs are not attributal but overhead
Don’t know how cost models are built
Build future products with better data
We want to partner with you, but need idea of what costs really are
Do a quick turnaround in costs models, weeks not years
Need to know your cost components so we can help reduce costs
Need to have highly accurate costs so we can agree to it
Art Scakler
Underlying discounts for cost avoidance are not adequate
National Policy Council
Discount don’t take into account what mailers have to do and spend
Uses Larry Buc
what cost is being counted for avoidance?
Need better measurement of 1st class price elasticity
First class mailers are more sensitive to price increases
Mailers will aggressively move mail out of system due to budget
All prices we have are down, but USPS prices are up
For every $1 spent, $0.65 goes to postage
USPS sees mail volume decline due to their increase in prices
First class elasticity is way off in forecasted numbers
USPS must internalize that things have changed for good
Give us accurate measurement on pre-sort costs
Discount for IMB is a joke
No ROI in UMB until 2020
Pre-sort not getting savings like mail sizes, etc
Savings requirements don’t cover cost to mailers
USPS changes create huge costs for us, then mail rate increases
Much more issues with accuracy of information
Problem is misjudgment with elasticity
Cost avoidance is a problem
Price elasticity should be urgent and easy to fix
Need to get a grip on compliance cost
No more changes until financial crisis over
Cost to mailers is enormous
Mailers have to leave, reduce volume to afford service
It is easy to get mail out of the system, electronic media can ramp up capacity
Figure out what is cost to mailers first, what is the impact
Budget guys will refuse to put money in mailings, and go on-line
APWU contract hurts the USPS
Labor contracts don’t take into account mailer views
Jim O'Brien -Time Inc.
Can't get to the bottom of periodicals cost raise
Halstein Sadberg
Periodicals has been increasing at higher rate compared to others
Higher costs pass through as higher rates
Current mail flow model needs to be updated since it is old and not real
Cost of containers is no longer valid as old BMCs not friendly to pallets
CRA cost don’t reflect marginal costs
When volume went down, costs did not
USPS can not reduce costs with lower volumes
Costs underwater are not real at all
Advantages in technology should have lowered costs, but doesn’t
Need to pursue accuracy of data
Cost figures for moving 5 digit pallets is study from 1980
Costs are no longer accurate since network no longer in use
Periodicals costs larger than postal wages increase
Improvement by mailers should have lower costs, but don’t
When USPS applies costs to rates, rates do not reflects cost
Marginal costs are not accurate
When USPS reduced staff, low wage people go first, so average wage is up
Cost recovery is less than 100% with lower volumes and high automation
Flat sorters are not maximized, they have over capacity
Map out workflows and put cost behind flow to track true costs
Update data as well as models
What happened to pallets? They became too expensive for mailers
Operations wanted more 5 digit pallets to handle
Additional cost of pallets made them not attractive to mailers
Mailers have to react to rate increases and save costs
IOCS is off the table from discussions with USPS management
Need to create grid with different containers, pieces, for cost activities
Cost base rate can help match cost to rate
AFSM 1000 specs for machinability is based on phased out machines
IOCS reflects what goes on in processing
Problem with IOCS is missing activities when people not handling mail
IOCS needs to reflect time in different stations
Mail handling time is less and less, but can't capture all with IOCS
Mail cost pools are not reliable
Mail is overloaded with other old attributed costs
Automation refugees are union agreement for excess people - manual bull pen
IOCS off due to lack of RIF, moving people around
Update costs to date
Update costs models and use Dave Williams new SOP
Use rate grid to have cost match rates
HQ needs to make more central decisions to standardize and reduce costs
Joy Franckowiak
IOCS is increasingly obsolete in Flats operations
Bill Olsen
IOCS carrier time in office should be less since FSS is in
Jeremiah Morgan
Cost models are not up to speed with IOCS
John Haldy
If IOCS is cut back, why not use new system anyway
Valpak
Cost system not kept up with automation, small sampling
A lot of full cost attribution is added, creating controversy
Use IOCS and ensure all systems are additive
Non IOCS costs must be captured from ground up
Workshare discounts are based on old ways and is not data driven
Discount rates should be the difference between before and after
What does it cost to work mail that is not workshared
Do bottoms up costing more reflective of real cost
Do baseline of most pre-sorted mail
Don’t use cost incurred, use cost calculated
Account system does not capture cost not incurred
Workshare discount is a crazy idea
Rate difference between presorted and others should not be more than $0.10
Why does the USPS want to charge less than cost of processing mail?
You can't do top down discounting
In future, street time will increase a lot more, city carrier time will be huge
USPS cost of excess capacity is nebulous
Costs for excess capacity are made up on the fly
The law requires you know what the costs are
On competitive products you still need to know costs correctly
Cost base pricing doesn’t matter, do market base pricing
Stop doing studies, use data
Plenty of mailers tell you cost don’t matter, they are stupid
Is the overall level of attribution on the table?
Foster associate reports
Subsidizing of products creates polarization of products
Price cap prevents loading of more cost into 1st class letters
Reduce rates to increase volumes of 1st class mail
what are truly fixed costs?
Consolidate more plants, then use IP engineering to get to the real costs.
Cheryl O'Day - Discover
Understand that sampling methodologies are dictated
Bob Brinkman
Cost avoidance methodology is backwards
Nicholas Wolf
Buy into ABC are more accurate costing method
Cost methodology is subtractive and should be additive
We pay an inflated price for our efficient mail
Studies and models are not substitutes for real data
Data proxies are old
Costing system lacks credibility because it lacks data
Use real data and IMB for real data
Data today is decades old
Large discrepancy of cost method between real and rated cost
Bottom up pricing will reveal discrepancies
The cost difference between flats and letters is too large
Underlying cost data must be real to justify costs
Need to know the real costs
Costs developed with proxies and layers becomes useless
Models are not current
Pull direct data from full service IMB, sample the rest
Information about each piece at each rate should be available to mailers
No useful data is available
Need real data, not useless pdf reports
Need to know of our 17 million pieces mailed this month, how many are 3 or 5 digits, etc
We will need higher transparency in dominant space to create more opportunities
Mailers don’t have data to justify 3% price increase
We have to add cost to third parties to explain your rate increases
The indirect cost of mailing is high in trying to figure out your data
Indirect cost of using mail may help decide if we use USPS or not
Use ABC and IMB for costing data
Priority will be to create bottoms up costing
Second priority will be to create useful reporting
Don’t rely on proxies, special studies and complicated models
Avoid analysis of convenience vs what is needed
Provide analysis in a format we can use computers to use (not pdf)
Mury Salls - DST Mailing Svcs
Cost pools too generic for commercial business volumes
Cost pools haven't changed in decades and don’t apply to Commercial
Cost pools only apply to 1st class mail
Commercial costing includes retail costs, even though we don’t use counters
Collection costs don't apply to commercial mail since we deliver to plant
Other mail we have is cross docked and not worked in plant
Our mail doesn’t go through originating plant, still we pay for it
Collection and processing of commercial mail is not represented in cost pools
55% attributal cost is horrible, need to get much better
Not familiar with what is not attributed costs
USPS has not shared actual cost it takes to process 1 oz. or 2-3 oz. pieces
Commercial 1st class needs move out date requirements and other activities not in workshare
all the extra cost for 1st class mail is not in cost avoidance
mail volume variability, what is that?
Cost of accepting mail has high impact on volume variability
we have trucks dedicated to go into UPS planes
cost avoidance for volumes are not captured
data is available for cost avoidance, but not used
why are you using estimates when data is available
Don’t mix time and motion studies with different type of mail - doesn’t work
Need to start with data, but data is old today
IMB and full service IMB should make data capture easy
Use information to create real costing for each mail piece, use your data
isolate cost pools that are really attributable
with IMB, carrier time to deliver can be found and costed
Use IMB to drive real cost data
Use DPS % to estimate in-office delivery costs
USPS assumes 0% savings on workshare collections
Benchmarks used are obsolete
Cost avoidance is a bad way to do pricing
Carol Kliewever - HarlandClarke Why are parcels not covering their costs?
Patrick Gallivan
ACR report doesn’t tell story of lack of coverage
Joe Rawlins
There is no explanation as to why no coverage
No feedback as to what we can do as mailer to help cover cost
We can be more efficient is we knew where and why we don’t cover costs
Our parcels are easy to process but get lumped with inefficient parcels
No info as to how overhead is allocated- is a mystery
Overhead is easily manipulated to do anything
We are surprised when ACR comes out, confusing and doesn’t add up
When rates goes up, we need to go deeper into workstream to make up the cost added
No transparency on overhead, so we don’t know how to react
Mailers feel USPS stuck it to us, high rate increases
We are incented to take action, but when we do we get penalized on the other side
If we induct mail close to destinating location, plant volume will go down and overhead goes up
Rate goes up to make up for loss volume that the USPS generated in the first place
Overhead goes up to make up for bloated plant costs
Efficiency is empty USPS trucks is excessive
We don’t get recovery for additional work to make USPS better by cross docking our mail
No one has price model with all the work we do to reduce costs and rates
Can USPS trucks go full to sectionals and rent empty truck space?
We cant help pricing committees, we don’t know what's in there
If parcels cover 72% of cost, what is missing 28%?
when you get to understand the model, it changes
there is a black hole so we cant understand models
Pricing group in USPS makes up models, cant explain them
USPS can spin anything on the models, no methodology
Allocation is not based on bottoms up, no engineering help to get it right
USPS sticks costs to others if they miss revenue
USPS will have nothing to give us in Sept for cost coverage
USPS thinks they are the only game in town
USPS service providers will need to raise prices, and you will pass that issue to us
Why additional $3 in parcel select returns? No explanation given
Forcing mailers to go to sectionals is due to labor contract issue
USPS does not use cost decisions based on efficiency
Work with mailers, tell us why we are coming up short
Make sure all understand overhead costs and allocations
consistently apply overhead costs
USPS plays with rate classes and punishes certain subclasses
USPS is not sharing information now days - hard to obtain
With competitive products things will get worse, no cap
Delivery of a product can cost more than the product itself
USPS is arrogantly saying "we are the only game in town"
When products go up 20%, there is no reasonable way to explain that
Pricing is intentionally inflated
need to have a fair, consistent, transparent process
Rita Cohen
Alarmed at unit cost attributable to magazines
Sandy Glick
Cost increases faster than inflation
Magazine.org
Magazines are not being measured correctly for costing
Unit cost should go down with workshare, but it doesn't
High increase in automation should reduce costs
We take on more work to reduce cost and USPS raises rates
IOCS designed to capture what was being performed
IOCS data is old and needs refreshing
too many tallies of non-mail handling due to automation, people just watch mail
Bad assumptions for subclasses in IOCS
Operations have changed, but cost systems have not changed with them
Costs system should reflect how mail should be handled, not how is handled
Wrong assumptions in cost system adds costs to us, higher rates
We tried various workshare options, and still rates go up, why?
Manual handling- should be more machine use for mail
if mail comes in late, USPS processes mail without thinking about costs
Facilities know how they do in services scores, but no idea on cost of products
plants do not look at costs, only hours in their budget
Something is missing in costs, but don’t know what
Critical entry time for periodicals is not adhered to
SOP's for mail handling are not used in the field
Excess capacity shows up as elevated costs
Field achieve their budgets, but products costs are not being looked at
Manual extra unnecessary handling adds more cost
Mixed mail handling at offices can add extra costs
USPS does pretty good job on reports
Once a year data is problematic
Accounting numbers are not consistent yet
field does not know how they are doing product by product
A system design in 1970 wont work today
Because of system not useful at detail level, it confounds the costs
Models have correction factors that don’t relate to anything
Institutional costs can be allocated differently if you know real costs
Bottoms up system will take too long to create
Tweak current system with improvements and make it valid
Review IOCS and other systems and improve only what is needed
no model will catch all costs no matter how good model is
Gene DelPolito
Costing system is ancient, same since inception
Jessica Lowrance
system not flexible for today's environment
Postcom
system is stuck in anachronistic way
Class of mail impairs proper workshare costing
cost pools for worksharing not sophisticated
Workshare typically not captured in cost
Workshare not reflected in pricing based on current state
Business mail cost should not be lumped with single piece
costing should reflect reality of bulk mail
Benchmarking used does not represent reality
consider a product for each discreet method mails comes in
need to capture mail cost based on real flow
Price future based on modern marketplace
USPS needs to know how to do commercial pricing
Bottom up costing should be enabled
Top down costing does not capture real $$
Measure true cost end to end and stop the benchmarking
If you do ABC, Workshare goes out the window as cost built into product
Don’t come up with so many products
Huge investment in IMB, stupid not to use it
Pricing is constrained by marketplace
support market driven pricing / don’t under cost and loose
need to know real product cost so we can help with selling it
old world thinking, time to break that
costs are masked and reconstituted, hard to see
Old models are outdated due to major changes in hardware
IOCS have been criticized, use MODS data
Hamilton Davidson
Rates have gone up significantly
Robert Mitchell
Perception that flats are not covering their costs
American Catalog Association
What can we do to reduce costs?
No one has defined what bottom up costing is
people has different opinions of what bottom up costing is
IMB gives you piece counts, but no idea about time
Notion of marginal cost will help understand the beginning of cost
Marginal costing approach is not arbitrary
Rates should reflect the effect of the costs they create
Don't forget marginality if you make changes to system
Assumptions used to create costs are no longer valid
Carrier assumptions for costing are wrong based on today's world
Why does letters rates go up a little and flats a lot
the commission neglects what we told them
costs are pretend costs and not real
non profit rates are extremely low
costs at year end should be lower based on all improvements during year
You are irresponsible managers to make decisions on pretend costs
non sampling error is good compared to sampling error
A "should cost" model should be developed
Identify best practices for each operation
Time to review assumptions in cost models
need to understand variable costs
cost system should be as simple as possible, and transparent
there are no standard processes in USPS
Cost should be visible so people can see it
old data into new models if bad practice
if you can't process catalogs efficiently, don’t do the business
ABC doesn’t use marginality at all
ABC is like spending dollars chasing dimes
Explain Factor Price index
ABC doesn’t do variable costs
do a cost benefit analysis for all projects
embrace variability into costing
Anita Pursley
Buckets in cost pools are too broad to be useful
RR Donnelly
What is the real cost of processing catalogs? Attack those costs
skepticism about costs, since there is no rationale
Need to explain what the components of costs are
All improvements we make don’t impact pricing
need to know the difference in pricing from ACD
If we knew the costs, we could help USPS be more efficient
Topic
Bottom-Up Costing
Workshare
IOCS
Cost Accuracy
From over 850
items captured
directly during the 1:1
interviews
Marginal Cost
Cost Attribution
Summary need
Advocate Bottom-Up Costing
Advocate mapping out and determine cost of each activity
Move to bottom-up, start small, use IMB for one product for a full year
Don’t use Bottom-Up (overall)
Don’t waste $ chasing dimes
Don’t replace valid system
Bottom-up not well defined
Bottom-up not get what we need – marginal verticality
Bottom-up won’t work can’t capture all costs – still need adjustment factors
Models do not include cost or value of compliance with prep rules
The concept of cost avoidance is a problem
The BMM benchmark is a problem
Not all cost differences are included in models
Worksharing is more complex than the models track
Models include costs that should be attributed to retail products
The models are impressive
Use a 2-part CRA adjustment factor in FCM
Improvements made by mailers don’t flow through
Input data needs to be updated
Flows need to be updated
IOCS not reliable
IOCS developed for manual operations now out dated. Automation – too many non-handling and indirect tallies
Overall works well
IOCS seems to justify high costs
Historically OIG questioned IOCS
Problems with small sample size
Even if use IMB, need sampling for manual operations
Cost pools too broad
Should consider expanding IOCS to include workshare
Use IOCS and ensure all systems are additive
Important to have accurate costs. Needed to support all types of efforts, it’s the law, especially so for competitive
products.
Concern on inaccuracy. “costs seen as too high or greater than they should be.
Need to know real costs.
Defect in the way costs are calculated
Alarmed at unit costs for magazines, rise in costs.
Flat, parcel costs done wrong
Costing lacks credibility because it lacks data.
MODS data not reliable
Operations have changed, but cost systems dated.
CRA doesn’t reflect marginal costs
Excess capacity cost made up on fly
Costs/models out of date
Costs supporting pallet rates are outdated
Old mail flow models
Data is decades old
Use new SOP for Periodicals models.
(unlike other methods) is not arbitrary
don’t lose focus on marginal costing (needed in delivery costs as well) to support pricing and management decisions
Over attribution occurs due to inclusion of non-handling, mixed mail tallies and overhead costs.
Need to study mail processing volume variability to get this correct – lamented outcome of past rate case decisions
by PRC.
On other hand, some said there was under attribution – surprised that USPS can only attribute 50 percent or 55
percent of its costs. Attribution should be much higher, only SG & A should be institutional.
Need to use short-run or short-term costing. While long-run based costing is useful, also need short-run to
consider short-term contracts and consideration of discounts more than just seasonal price adjustments. Need to
be able to determine when additional mail volume offers a positive contribution or covers the truly variable costs.
Volume Variability
No Overhaul
Actual Data
General Operations
Understand costs
Reporting
IMB
Transparency
Carrier Costs
Using cost to manage
Should cost
Average cost
Costs didn’t go down as much as volume. Much inflexibility on reducing staffing and infrastructure. Excess capacity
shows up as elevated costs. Need to study excess capacity.
Assumptions out of whack with reality – excess capacity arose due to big volume decline – overstating costs.
No need to overhaul whole system
Fix only what doesn’t work
Do basic improvements
Don’t start over
Use real measured data and not samples
Replace assumptions with facts
Update co sts to current workflows
Ensure sampling is accurate for small volume products
Replace assumptions with data for workshare
Pallets not being scanned, not used due to price increase
Manual handling should be limited and sorters maximized
Labor contract adds costs
Identify best practices and use for pricing
Consolidate more plants, use IP engineering for true cost capture
What makes up cost?
Need building blocks of cost
Need to understand cost coverage
Why parcels not covering unit cost
Need to understand reason for cost increase
If we know costs we could help you
Simple explanation of costs needed UC
Explain why rates go up after workshare implementation
Format of the data isn’t presented in a useable manner
Frequency (>1 year)
ACR doesn’t tell full story- lack of coverage
Would like more granularity (e.g. by customer)
USPS does an overall good job
No ROI on IMB until 2020
USPS should use IMB for cost data
IMB is not enough information, no time dimension, mail not on machines, doesn’t tell you where labor is
Domestic side is transparent
International is not transparent enough
We can’t help pricing committees if we don’t know what’s in there
Cost system should be precise and accurate
Cost system doesn’t seem transparent or understood by industry
Too many non-handling tallies to used to allocate costs
With automation more work moving from carrier in-office to street time
Facilities don’t know their costs by products
Processing decisions made irrespective of cost
Engage mailers - show mailers (mailflow) and ask which steps actually add value
Can’t reduce/manage what you don’t measure
Don’t always see USPS make good cost decisions based on efficiencies
Develop and use optimal cost for models
Costs should be based on work needed, not actual expenditures
Mailers don’t want to be charged for inefficient operations
Cost pools set up to do generic FCM
Should know weight based cost (2-3 oz letters)
Higher volume more efficient mail subsidizing lower volume mail
Commercial mail subsidizing non-profit
Offsite Focus
Group
Topic 1
Topic 2
Affinity diagram
session completed to
103 distinct Customer
Needs in 18
Categorical topics
Topic 3
Topic 4
6
Project Overview – What we have heard
Need to completely rethink Cost system
Need for Bottom Up Costing
IOCS is obsolete
Need to get rid of cost avoidance
models
Using the IMB scans as much as
possible
Can’t manage what you
can’t measure
Don’t overhaul Entire System
Bottom up cost model is not panacea
IOCS works, consider expanding
Overall, cost avoidance
models do good job
Not clear IMB tells you a lot about what
is going on
More than ever, accuracy and
precision of costing data is
important today
7
Break Out Session Topics
• Understanding Bottom Up Costing
– Specifically define bottom up costing.
– What are the risks/costs/benefits of using?
• Data Source Discussion (IMB, IOCS, etc)
– Discuss if and when IOCS sampling should be replaced by another data
source (ex. IMB).
– What are the risks/costs/benefits?
• Workshare Model Discussion
– What should be measured / included in the cost avoidance estimates
– What are the risks/cost/benefits
• Cost Data Discussion
– Is there cost information that USPS does not either currently produce or
provide, that you believe you need?
– What are the risks/costs/benefits?
8
Break-Out Session Overview
• Four Rooms – Room are assigned
• Within each room, will further break down into four
groups.
• Each room has the same four topics
• Participants will rotate through topics; Moderator will
signal when to change
• Discussion led by participants
• Participants vote on most important points – vote done
by stickers
• Report out to main group
9
Costing System Overview
What do we mean, we say costing system?
- The costs used to develop our products
- Both the methodology and data used to develop these
costs
- Includes how the costs are reported – both frequency
and level of detail
10
Costing System Overview
Cost and Revenue Analysis Report (CRA)
• Produces costs by product
• Methodology overview
– Obtain total accrued costs
– Estimates attributable costs
– Distributes Attributable Costs to Products using Distribution Keys
• In-Office Sampling System (IOCS), Transportation Accounting
System (TRACs), City Carrier Costing System (CCS)
• Produced Annually
• Reports Data on a National Level
• Does not have a line item for every level of workshare
11
Costing System Overview
Workshare Cost Models
• Provide additional granularity beyond CRA
• Uses modeling to estimate specific cost savings associated
with specific workshare activities
• Most models tie back to the CRA with adjustment factors
• Most focus on cost avoidance estimates, not bottom up cost
12
Costing System Overview
Costing for Contract Pricing
• Costing begins with national average unit cost for most
similar product
• Adjustments made for partner-specific characteristics
when possible
– Costs for parts of postal network bypassed are removed from
national average unit cost
– Costs for additional activities not ordinarily part of national
average unit cost are added in
– Partner’s mail mix (shape, weight, cube, haul, degree of
dropship) incorporated into the unit cost considerations
13
Cost System Overview
• The Current Cost System has undergone various types of
reviews and audits.
– PRC Annual Compliance Determination
– Ernst & Young (for revenue and volume)
– OIG audits regularly, as required by PAEA
– GAO audits, as requested
– Reviewed by customers and competitors
• Never failed audit
• PRC has found compliance
14
Cost System Overview - Level of Accuracy
• Statistical Significance of the Sampling System
First-Class Mail and Standard: ± 1%
Periodicals and Priority Mail: ± 2%
Package Services and Express Mail: ± 3%
Parcel Select and Parcel Return: ± 4%
International: ± 5%
• What Does this Mean ?
Product
Cost per Piece
95% Upper Bound
95% Lower Bound
First-Class Mail
$0.211
$0.210
$0.213
Priority Mail
$5.246
$5.102
$5.391
Express
$11.647
$11.256
$12.038
International
$2.798
$2.579
$3.017
15
Cost System Overview
•
Current System….
–
–
–
–
•
Is trusted and approved by PRC
Ties back to general ledger and operation costs
Reflects what actually happened, not “ideal cost”
Adapts to product and operation changes
Limitations
–
–
–
–
–
–
–
–
Limited to national focus
Only provides annual reporting
Lack of product data at finer detail levels
Less precise measurement of low-volume products (special studies
required)
Some dated special studies
Cause of product cost changes are not readily apparent
PRC regulates costing methodology changes
No estimate of short-run marginal cost
16
Expectations/Feedback
• Follow-up
– Report out highlights to MTAC on Thursday, August 18th
– We will send a summary of feedback to you by end of next
week
– We will share recommendations and justification with you in
early November
• Your Expectations / Feedback
17
Rules of Conduct
•
•
•
•
Feel free to voice your opinion, but let others speak
Please do rotate through each of the topics.
Please keep to your assigned break out room.
If you need help, seek out the moderator.
18
Break Out Sessions
• Break Out Room Assignments are on Name Tag
• Questions?
19
Next Steps
• Report out highlights to MTAC on Thursday, August 18th
• We will send a summary of feedback to you by end of
next week
• We will share recommendations and justification with
you in early November
20