Oew8y
HD28
.M414
>
ALFRED
A
P.
WORKING PAPER
SLOAN SCHOOL OF MANAGEMENT
BE31AVI0RAL SIHJIATION
WDEL OF SALES
PLANNING AND CXWTROL IN A DMaCOMMUNICATIGNS CX3MPANY.
by
Jdin D.W. Morecroft
WP-1761-86
MASSACHUSETTS
INSTITUTE OF TECHNOLOGY
50 MEMORIAL DRIVE
CAMBRIDGE, MASSACHUSETTS 02139
D-3806
^
A BEHAVIORAL SIHJLATICN MDML OF SALES
PLANNING AND CXKIROL IN A DAIACGmUNICATIONS OOHPANY.
by
Jc*n D.W. Morecroft
WP-1761-86
D-3806
A BEHAVIORAL SIMULATDN MODEL OF SALES
PLANNING AND CONTROL IN A DATACOMMUNICATIONS ODMPANY
by
John D.W. Morecroft
System Dynamics Group
Sloan School of Management
Massachusetts
Institute of
March
1
986
Technology
D-3806
A BEHAVDRAL SIMULATION MODEL OF SALES
PLANNING AND CONTROL IN A DATACOMMUNICATIONS COMPANY
Abstract
The paper describes a behavioral simulation model which represents the
procedures for sales planning and control in
policies and
datacommunications companies. The model allows you to think about the
different 'players' whose actions and choices regulate sales. There are
quite a few of them -- account executives, customers, sales executives,
business planners and compensation planners. You step into their shoes.
You examine their responsibilities, their goals and incentives and the
sources
information that attract their attention -- all with the
intention of understanding the logic behind their choices and actions. Then
you stand back from the detail of the individual players and, with the help
of diagrams and computer simulations, you explore how the players
interact and cooperate, and how the system as a whole balances the sales
goals of the company, the efforts of the salesforce and the needs of
customers.
of
D-3806
INSIDE THE
SALES ORGANIZATION
Imagine
you
are
vice-president
the
sales
of
a
in
datacommunications company, with revenue responsibility
major product
lines.
Included
in
growing
for several
your product portfolio are telephones, key
systems, medium-sized PBXs, personal computers and workstations.
Due
to
advancing technology and increasing competition the products are
frequently changing
~ new models are introduced and
offered with
enhanced
You have
your
at
models are
features.
command a
sixty high-calibre
existing
professional salesforce presently comprising
account executives, each one assisted by a
staff of
three or four people that provides technical and clerical support.
salesforce sells mostly to business customers,
with annual sales revenue
As you
think about the
several questions
products,
lines?
come
between $500
problems
to mind.
of
million
and $10
that
to large firms
billion.
managing the sales organization,
Given the company's wide range
how do you ensure balanced growth
How do you know
medium
The
account executives
of
of the different product
will
allocate their selling
time so that you meet your negotiated sales quota, so the firm meets
its
How
do
sales objectives and so the factories can satisfy customer orders?
you know that account executives are pushing the products customers
most want
to
and growing
buy?
How do
you know that your salesforce
at the right rate?
is
the right size
D-3806
To get another perspective on these questions you
one
the position of
of the
account executives. You know most
They are a talented group
of individuals, extrovert,
company's products and persuasive
Most earn
their talents.
put yourself
try to
at
least
in selling
of
them
knowledgeable
well.
of the
them. They are well-paid
$60,000 per year and the
in
for
'stars'
sometimes as much as $250,000 per year. What motivates them? How do
they decide which products to sell? Money, prestige, and competition with
colleagues are doubtless important influences. But,
think of
it,
to
they are obviously not motivated by the firm's overall sales
goals (how
still
when you come
many account
executives actually
know these
goals),
and
less
by your own negotiated sales quota or the factories' production
commitments! So how can you plan and control the sales
of
your own sales
organization?
To help explore these questions a
simulation model of a sales organization
has been developed which represents sales planning and control
datacommunications companies. The model allows you
different 'players'
quite
a few
of
whose
them
--
to think
sources
of
about the
actions and choices regulate sales. There are
account executives, customers, sales executives,
business planners and compensation planners. You step
You examine
in
their responsibilities, their goals
information
that
attract
their
into their
shoes.
and incentives and the
attention
intention of understanding the logic behind their choices
--
and
all
with
actions.
the
Then
D-3806
you stand back from the
detail of individual players
and, with the help of
diagrams and simulations, you explore how the players
cooperate, and
company, the
how
and
interact
the system as a whole balances the sales goals of the
efforts of the salesforce
and the needs
of
customers.
Account Executives' Time Allocation
Let's begin with the
account executives. To keep things simple imagine
they have to choose between selling just two product lines
system
like
a PBX
selling for
--
a large
perhaps $100,000 and a small system
like
a
personal business computer/workstation selling for $20,000. (We'll base
the whole model on just two product lines.
It's
simpler than reality of
course, but even so can teach us a great deal). Naturally they're interested
in
the 'payoff.
what
for
What
is
their
compensation
for selling
a large system, and
a small system?
To make a choice they have
These schemes come
in
to
know
many
the terms of the compensation scheme.
varieties.
Some companies use a
commission (account executives are paid a
straight
fixed proportion of the selling
price),
some use a combination
of
base 'guaranteed' salary and commission.
In this
case a points scheme
is
used. Account executives are awarded
25,000 points
workstation.
dollars at
for the sale of
They keep a
a rate
of
a
PBX and
tally of
points
5,000 points for the sale
and convert
The terms
of the
a
their points to real
say $1000 per 5000 points, which
base, guaranteed pay.
of
is
added
to their
compensation scheme can, and do.
D-3806
Change from month-to month. So account executives must keep up-to-date
and watch
for
announcements
of
changes
the points to dollars conversion rate.
scheme can be
the points value of products, or
in
The
details of
quite complex, but the basic principle (from the
executives' perspective)
is
quite simple
and
scheme. The more points they earn, the bigger
Account executives know
and only 5,000
for
they'll
is
similar to
their sales
bonus.
PBX
each workstation. But points alone don't
number
account
a commission
earn 20,000 points for each
time allocation. They also take account of the
sell
any given points
of
they
sell
dictate their
hours
it
takes to
a PBX versus a workstation. There are several reasons why a PBX sale
more time consuming. The PBX
complex than the workstation, so
to explain
its
it
is
takes account executives a long time
features to a customer.
to win his
company's approval
chances
of
winning an order
for
a
much more expensive and more
It
may
take the customer a long time
large expenditure
may be
1
summarizes the
allocation.
Finally, the
low because, given the expense,
customers look closely at competitor's products
Figure
on a PBX.
too.
factors that influence account executives' time
They keep abreast
of the points system.
From
their field
experience they estimate the time per sale of large systems and small
systems. Knowing points and time per sale they can judge, roughly, the
payoff, in terms of points per hour, to selling either
small system. For example, suppose
it
a large system or a
takes, on average,
60 hours
to sell
a
D-3806
Time Allocated to Sale
of Large Systems
Time Allocated to Sale
of Small Systems
Total Sales
Time Available
Estimated Time
Per Sale of
Large Systems
Estimated Time
Per Sale
of Small Systems
Points for
Points for
Large Systems
Small Systems
A-3784
Figure
1
:
Account Executives' Time Allocation
D-3806
PBX
8
(the total time spent with
delivery
and
choose not
--
installation
to buy)
a customer from
contact to
final
including time lost with those customers
who
and 15 hours
a workstation. Given the terms
to sell
the compensation scheme, the payoff for selling
points per hour
and the payoff
points per hour. Therefore,
devote more of
in this
their time to selling
They have a
PBXs
is
of
25,000/60 or 416
for selling workstations is
5000/15 or 333
case, account executives
want
will
to
PBXs.
But account executives don't switch
exclusively.
first
in
a single day
variety of contracts already
to selling
PBX's
underway which they
are obliged to complete. Their time allocation changes only gradually,
depending
on
new
the
selling
opportunities
commitments and the current payoff
process
is
dynamic. Conditions change from day
The compensation scheme may be
change as the
vary.
New
for large
relative prices
products
may be
modified.
and delivery
introduced
available,
and small systems. The
to
day and week
The time
to sell
intervals of
whose
product
line
which,
gradually shift
selling time
in their
more and more
week.
each product
will
line
one can only
and quite predictable, once the points and time per sale are
will
to
a system
guess. Despite these complexities, salesforce time allocation
Account executives
existing
is
logical
specified.
of their time to the
judgement, gives the biggest payoff.
Compensation Plannino
Compensation planners design the compensation scheme. They are the
D-3806
interface
between sales objectives and
tlie
salesforce.
they look at performance against objective
and decide how
large systems)
account executives
objective. So,
if
to sell
line
it
more
it's
case
in
shows,
for small
to adjust product points to
products
will
and
encourage
the quantity called for by the
increase the points value of the product,
attractive to sell.
exceed the sales
decrease
figure 2
sales of a particular product line are below objective,
compensation planners
making
(in this
As
On
the other hand,
if
sales of a product
objective, planners will maintain or possibly
points value.
In
practice
it
is
much
even
easier to increase points
than reduce them, because account executives strongly resist any attempt
to
lower their compensation.
Sales Objective
The sales
objective
that consolidates
comes from a complex business planning procedure
information and judgement from market analysts,
account executives,
product
managers,
product
schedulers
manufacturing planners. The business plan starts from
total industry
From
volume
intervals)
an estimate
classes of datacommunications equipment.
price changes,
competitor actions, expected delivery
market analysts compute the company's expected share
industry sales.
of
data and various business assumptions (new product
historical
introductions,
for all
and
By applying the share
to industry
volume, the business plan
generates the company's sales forecast by product
planning horizon.
of
line
over a two year
D-3806
10
Points for
Points for
Large Systems
Performance Against
Objective for
Large Systems
Small Systems
Performance Against
Objective for
Small Systems
A-3785
Figure 2: Compensation Planning
D-3806
11
Although a great deal
important input
the recent history of customer orders, as
Planners compute a base estimate of
figure 3.
volume
is
of information enters the planning process,
of
demand
using
a most
shown
in
last year's
customer orders. Then, during the sales commitment process,
executives increase the base estimate by a 'stretch margin' to arrive
at
the corporate sales objective. For example, suppose the salesforce sold
1000 PBXs
last year.
If
the stretch margin
corporate sales objective for the
margin
is
coming year
a simple but powerful way
for
sales objectives for sales managers,
10 percent, then the
is
is
1100 PBXs. The
executives to set challenging
in
an environment
of
uncertainty about customer
demand. The
to strive for higher sales
volume, by holding them accountable
objective that
is
stretch
objective requires sales
great
managers
for
an
greater than last year's (or last quarter's) sales.
Performance against sales objective (which compensation planners use
adjusting points)
is
in
obtained by comparing customer orders to date, with
the sales objective pro-rated to the
same
date.
D-3806
12
Performance Against
Sales Objective
Stretch Margin
Custonner Orders
A-3786
Figure 3: Business Sales Objective
D-3806
13
Customer Ordering
It's
common
to think that
customer orders are completely determined by
and
factors such as price/performance, availability
Certainly
quality.
these factors are important, but they're not the whole story. Put youself
PBX
the position of a business customer buying a
about your knowledge
of the
or
in
a workstation. Think
product and your information sources as you
begin the purchase.
A
principal influence
for office
systems
will
a demonstration and
sales
is
talking with
is
availability
aware
and
of the
quality
figure 4.
in
Few customers
product and
its
is
long,
is
say 12 months), or
important.
customers.
is
is
when
Each customer has
if
low (or
price
is
,
put another way,
high,
low,
if
in
the
the
if
account executives
own
general,
will
the product
will
his
if
delivery interval
account executives
long time to find willing customers. Conversely,
available, or the price
it
product do factors such as price/performance,
become
product
having
makes them
features. Only
perception of these factors and sensitivity to them. But
availability of the
first
account executives. Very often
the customers' interest and
initiates
of the existence of the
customer
as shown
spontaneously place an order without
account executive who
aware
effort,
soon
is
take a
readily
find willing
D-3806
14
Customer Orders for
(or Small) Systems
Large
Availability
Quality
-Time to Make
a Sale
Sales
Effort to
(or Small)
Large
Systems
Price/Performance
A-3787
Figure 4: Customer Ordering
15
D-3806
Force
A
Hiring. Layoff
and Budgeting
description of sales planning and control would be incomplete
ignored force hiring and
their salesforce,
which people are responsible
motivates their decisions?
shown
How do companies
layoff.
In
if
it
regulate the size of
for
hiring
broad terms, force planning
is
and what
driven by the
Given the salesforce budget and knowledge
of
average salesforce compensation, planners can estimate the number
of
budget, as
in
figure 5.
account executives the budget can support
--
the authorized salesforce.
If
the authorized force exceeds the current force then hiring takes place and
the salesforce grows. Conversely,
if
the current force exceeds the
authorized force then layoffs take place and the salesforce contracts,
(alternatively the
attrition
and
company may
layoff to
The important
characterize budgeting
policy
point
in
is
or
some combination
of
is
a complex process which the model only
to
capture the inertia and myopia that
R&D
New budgets
fVlajor
budget items
expense, sales and service expense)
change dramatically from year
budget.
attrition,
most large organizations.
(such as advertising expense,
rarely
on
reduce the salesforce).
The company's budgeting
outlines.
rely
to
year as a proportion of the
total
are often just incremental adjustments to old
budgets, because organizational politics cannot cope with radical change
and because
it
is
complicated and time consuming to
item from scratch each year.
incremental,
inertial
The budgeting
adjustments, as
shown
in
justify
every budget
policy captures these
figure 6.
The budget
for the
D-3806
16
Change
Current Salesforce
in
Salesforce
Authorized Salesforce
/
Budget
for
Salesforce
Average Compensation
of
Salespeople
A-3788
Figure 5: Force Hiring and Layoff
17
D-3806
salesforce
itself is
is
represented as a fraction of
total
sales revenue (revenue
the product of price and customer orders for both large and small
systems).
The
For example,
fraction
it
is 'sticky*
(slow to change) though not rigidly fixed.
slowly increases
if
total
salesforce compensation
repeatedly exceeds budget.
Budget
for Salesforce
Fraction of
Sales Revenue
Revenue
to
Sales
A-3789
Figure 6: Budget for Salesforce
D-3806
18
DYNAMICS OF SALES PLANNING AND CONTROL
Scenario
Up
to
1
--
"Test Drive" Balanced Product Portfolio
now we have described
the
compensation and number
begin,
let's
try
a
of
large
and small systems
the salesforce generates the
a
orders, sales objectives,
account executives change over time. To
of the
'test drive'
in
understand the dynamics of
how customer
--
easy-to-interpret product scenario.
First,
sales are planned and controlled
Now we want to
datacommunications company.
sales planning and control
way
model, a simulation of a special,
The scenario assumes
yield the
same
total
same revenue
four conditions.
per sales hour, so
revenue no matter how account
executives allocate their time. Second, large and small systems are
equally attractive to account executives
--
payoff to selling either large or small systems
for large
potential
customers
over time?
Will
for both
how
conditions, imagine
large
large
for small
systems
if
so why?
simulation of the scenario
systems
(-2-)
profit
and there are many
How
will
will
evolve
account executives
—
organization model (documented
Orders
to the sales objective.
the business units' sales and revenue
revenue grow and
shows a
identical. Third, orders
and small systems. Given these
allocate their time?
Figure 7
is
and small systems are exactly equal
products are priced to generate a
Finally,
other words, the points
in
(-1-)
in
made by
running the sales
the appendix) for a period of 24 months.
begin at 60 units per month and orders for
begin at 108 units per month. Both grow smoothly and
D-3806
19
orders_smaI!
1
1
120.000
1
100.000
1
80.000
1
60.000
»
40.000
18.000
24.000
Time
1
ordersJarge
1
200.000
I
170.000
1
HO .000
1
110.000
1
80.000
6.000
T
12.000
I
I
18.000
Time
Figure 7: Customer Orders
When
Products Are
Equally Attractive to Salesforce
24,000
D-3806
20
exponentially over the 24 month simulation at a rate of approximately 30
percent per year. Figure 8 shows that the salesforce
the
same 30
with 90
of
1
(-2-) in
identical throughout the simulation).
their initial time allocation (in this
there
is
for large
no incentive
The growth
to
for small
case 10 percent
systems
is
stick with
for small
systems
systems) throughout the simulation, because
in
revenue and salesforce results from a positive
figure 9, that couples budgeting, salesforce hiring
the salesforce, on the right-hand side,
figure imagine
is
increased.
what happens
First,
if
sales effort
because there are more account executives. Then customer orders
because more customers are informed
and persuaded
revenue which
to
in
turn
expands the budget
As long as there are
who, once informed,
will
company's products
for the salesforce
and therefore
plenty of potential customers
consider buying the company's products
words, as long as the market
will
of the
buy them. More customer orders generate more sales
allows more hiring.
cycle
large
So account executives
and customer ordering. To understand the
will rise,
and
change.
of orders,
feedback loop, shown
will rise
systems
of small
the lower half of figure 8) remains constant at a value
(meaning that the points payoff
and 90 percent
at
60 account executives and ending
rate, starting with
month 24. Meanwhile the attractiveness
in
(shown as
percent
growing
(-1-) is
continue, leading to
isn't
(in
other
saturated) then the reinforcing positive
more and more orders and steady
D-3806
21
2 aulh_force
sales-force
1
y
120.000
y
100.000
J)
il
80.000
60.000
st-aw. $cer>»
d<i(.-.
^1
'
t
I
I
24000
'
'
0.0
I'll
I
I
I
i
I
I
40.000
I
'
I
18.000
12.000
6.000
Time
1
1
0.500
2
1.400
1
0.375
2
»
2
1
2
2 attract-smaH
frac_effort_sm
1
.
.200
0.250
1
.000
0.125
0.800
1
0.0
2
600
-pio dtl
'
I
0.0
'
6.000
1
'
I
'
I
'
I
'
'
I
I
12.000
'
I
'
I
'
I
I
I
'
I
'
I
stam.
:
I
'
I
18.000
'
I
'
'
I
jccirv
I
I
'
•
\
'
I
I
24.000
Time
Figure 8: Salesforce (top) and Product Attractiveness
(bottom) Wiien Products Are Equally Attractive
D-3806
22
Total
J
Per Sale
^^ Sales Effort'
•'2®,
\
Sales
Effort
Salesforce
d}
Customer
Product
Orders
Prices
V^
Budget
/
Sales
Revenue
for
Salesforce
^^
,-^^^
Average
Compensation
Per Salesperson
A-3781
Figure 9: Positive Feedback Loop Connecting
Budgeting and Salesforce Hiring
D-3806
23
exponential growth.
Scenario 2
Introduction of an Attractive
--
Imagine now a more complex scenario
new
attractive
in
Small System
which the company introduces an
workstation to complement
Market analysts expect the workstation
and
New
to
its
small system product
be well received by customers
more revenue per sales hour than
to yield
large
systems (PBXs).
Moreover, because company executives are anxious to boost sales
new
As
workstation,
before,
potential
it
customers
new product and
of th.e
offered with an attractive compensation package.
is
systems are priced
all
line.
for both
to
produce a
PBXs and
incentive conditions,
orders evolve over time and
how
will
the
new
how
profit
and there are many
workstations. Given these
will
revenues and customer
account executives allocate
their
systems
(-1-,
time?
Figure 10
shows a six-month
top) start at
units per
for large
70
units per
month by month
systems
(-1-,
simulation. Orders for small
month, quickly grow to a peak of almost 100
4,
and then gradually
bottom) remain static at about 110 units per
month. What's happening, and why? Figures
explanation.
The top
half of figure 11
compensation scheme
of the simulation,
decline. Meanwhile, orders
that drives
1
and
1 1
together provide an
shows the changing terms
of the
salesforce time allocation. At the start
compensation per hour (shown here as dollars per hour
24
D-3806
2 obj_small
orders-small
i
y
120.000
V/
100.000
II
80.000
II
60 000
y
40.000
v-nodcl- stoTn_s-cert2<^
—
'
'
'
'
'
'
I
'
objJarge
ordersJarge
\
}
y
I)
''-1
6.000
4.500
200.000
170.000
140.000
y
110.000
y
80.000
4
-4
—
-p-i
I
W
-;
r
'
I
I
I
1.500
0.0
'
3.000
'
I
3?
32
=^4^
'
I
I
I
I
-r~
6.000
4.500
Time
Figure 10: Customer Orders and Business Sales Objectives
When
Attractive
New
Small System
is
Introduced
25
D-3806
per account executive, including indirect sales support expense)
systems
for small
imbalance occurs
(-2-,
top) than for large
two reasons.
for
deliberately designed to
(-1-,
the compensation
First,
encourage sales
systems
of small
is
higher
top).
The
scheme
is
systems (workstations).
Moreover, workstations are assumed to be easier to
sell
PBXs
than
--
they
generate more revenue per sales hour. Not surprisingly then, account
executives allocate more and more of their time to selling workstations
and
PBXs. The time
less to selling
re-allocation doesn't
though, as the lower half of figure 11 shows.
allocating 10 percent of
effort to
its
The
happen
instantly
salesforce starts by
small systems
(-1-,
bottom) and
gradually increases to 12.5 percent by month 4. This small re-allocation
cause small system orders
sufficient to
to increase
30 percent
in
is
only
four months!
However, even as the salesforce
compensation scheme
Orders
for small
to halt the trend. But
systems
top) but orders for large
conditions there
(-1-,
result is
top)
systems
more
effort into selling small
(-1-,
top)
why? The answer
exceed the sales
of
the
lies in figure
objective, (-2-,
systems are below objective. Under these
pressure for planners to increase the points awarded
is
for the sale of large
The
putting
compensation planners are changing the terms
systems,
10.
is
shown
systems and decrease the points
in
figure
1 1
.
for small
Compensation per hour
systems.
for large
systems
slowly increases while compensation per hour for small
(-2-,
top)
falls.
Shortly after
month
4,
compensation
is
identical
r
r
D-3806
26
2 comp_hr_sm
compjirjg
1
1}
300.000
J)
262.500
J)
225.000
J)
187.500
J}
\-no c)«t
—
150.000
I
I
I
I
'
I
I
I
_ scc.ft2f
I'll
I
3.000
1.500
0.0
S'TOm
•
I
4.500
6.000
Time
2 altract—small
frac_erfort_sm
1
1
0.500
2
1.400
1
0.375
2
1.200
\
0.250
2
1
.000
2
0.125
0.800
1
0.0
1
2
0.600
———
>
I
I
1
I
1.500
0.0
——
'
'
'
'
I
'
'
'
Tno<i «C'.
—
I
'
'
I
•
I
sraw- SCO 2e
I
I
'
—
'
4.500
3.000
6.000
Time
Figure
1 1
:
Compensation and Time Allocation When
New
Small System
is
Introduced
Attractive
27
D-3806
and small systems, so account executives are now
for large
which product they
workstation or PBX! Accordingly they cease
sell --
so the fraction of
time,
devoted
re-allocating
their
systems
bottom) stabilizes at about 12.5 percent.
A
(-1-,
new product
the
units per
of
75
month.
units per
to
small
interesting features
in
scenario, as figure 12 shows. Orders for small systems
70
top) start at
(-1-,
effort
24 month, simulation reveals some more
longer,
indifferent
In
units per
month and grow
month 5 orders begin
month by month
10.
rapidly to
to decline,
and
more than 100
fall
to
a minimum
Then orders recover and go through
another (diminished) cycle of growth and decline between months 10 and
20.
By the end
top) are
still
units per
simulation.
of the
24 month simulation, orders
for small
systems
fluctuating slightly, but are clearly settling at about
month, almost 30 percent higher than
The sales
only gradually because
objective for small systems
it
is
an average
result of this inertia, orders for small
of recent
in
months
The lower
0-7,
and again
half of figure
objective for large systems
in
90
at the start of the
(-2-,
top) rises, but
customer orders. As a
systems exceed the sales objective
during (and immediately following) any period of rapid growth
as
(-1-,
in
orders,
months 12-19.
12 shows orders
(-2-,
bottom).
(-1-,
bottom) and the sales
One can see a
very slight
D-3806
28
2 obj_smalI
orders-small
!
J}
120.000
i)
100.000
11
80.000
2J
60.000
i)
40.000
J)
——
I
-I
6.000
I
'
I
I
12.000
24.000
18.000
Time
objJarge
ordersJarge
1
J)
J}
^)
J)
200.000
170.000
140.000
110.000
-
Tt\o«Ac^.".
i!
s(-ar^- seen ^t>
80.000
0.0
6.000
12.000
24,000
18.000
Time
Figure 12: Customer Orders and Business Sales Objectives
When
Attractive
New
Small System
is
Introduced
29
D-3806
fluctuation
in
orders for large systems. But the most striking feature of
the simulation
Remember
is
that orders are almost constant at
that in scenario
1
110
units per month.
(balanced product porfolio) orders for large
systems grew steadily from 110
units per
month
to
170
over 24 months. The only change between scenario
introduction of
a succesful new workstation. The
as before,
selling at the
should the
new
A
same
1
month
and 2
PBX system
is
PBX
in
the
is
the
same customers. Why
price, to the
workstation halt the growth
units per
same
then
sales?
close look at the system's feedback structure and the conditions
surrounding salespeople's time allocation gives insight into the puzzle.
Figure 13
shows two feedback loops
that interconnect procedures for
business planning, compensation planning, salesforce time allocation and
customer ordering. Together these procedures produce dysfunctional
competition for sales time.
in
figure 13 to understand
The reader can
how
at the attractiveness of large
is
low.
trace around the feedback loops
the competition for time occurs. Look
systems which,
Compensation per hour
for small
at the start of the scenario,
systems exceeds compensation
per hour for large systems, causing account executives to
effort to
small systems.
As a
result,
first
shift
more sales
orders for small systems increase
unexpectedly, above the sales objective, while orders for large systems
are depressed
,
below the sales objective.
In
order to correct the sales
variance, compensation planners increase the points on large systems and
reduce the points on small systems. The result of the points adjustment
D-3806
30
A-3782
Figure 13: Feedback Loops Connecting Compensation Planning,
Time
Allocation
and Customer Ordering
D-3806
is
31
shown
in
rises during
systems
look
figure 14.
more
for large
systems
top)
(-1-,
through 6 while compensation per hour for small
months
(-2-,
Compensation per hour
top) falls, Shortly after
month
4, large
systems begin
Between
attractive to the salesforce than small systems.
months 4 and 9 account executives reallocate
to
their time increasingly in
favor of large systems. Orders for small systems decline while orders for
large
systems increase. By month 10 the sales variance
entirely reversed
variance problem
--
small system sales are
compounded because
is
small systems has been revised upward
and large system sales are below
engage
in
in
now below
situation
objective
(
is
the
the business sales objective for
in light
objective.
another round of points adjustment
of the
initial
sales sucess)
Compensation planners then
in
their
quest to bring orders
line with objective.
The competition
performance.
for sales time
First,
fluctuates, playing
addressed
has two dysfunctional effects on business
sales effort allocated to large and small systems
havoc with manufacturing schedules (these
specifically in
effects are
a report D-3807 dealing with the dynamics
of
production scheduling). Second, the average compensation of the
salesforce rises. Each round of points adjustment bids up the firm's
selling
expense, because compensation planners are more
increase points than they are to decrease them.
expense
stifles
shown
figure 15.
in
The increase
willing to
in selling
growth, by reducing the growth rate of the salesforce, as
As salespeople's compensation
rises, the existing
D-3806
1
J)
3
J)
3
J)
3
1)
32
2 comp_hr_sm
comp-hrJg
3 est_time_sale_sm
300.000
24.000
262.500
18.000
225.000
12.000
167.500
6.000
150.000
3
0.0
0.0
6.000
12.000
18.000
24.000
Time
!
frac_efrort_sm
attract_sma!I
0.500
1.400
0.375
2
1.200
1
0.250
2
1
2
1
2
1
.000
0.125
0.800
0.0
0.600
12.000
18.000
Time
Figure 14: Compensation and Time Allocation
Attractive
New System
is
When
Introduced
24.000
D-3806
1
sales-force
\]
120.000
IJ
2.800e+6
J}
51
33
2 auth_force
3 budget
4 total-comp
100.000
2.600e+6
80.000
51
2.400e+6
J)
60.000
I)
2.200e+6
J)
5)
aTn-j<:cft 2b
40.000
2.000e+6
6.000
12.000
18.000
Time
Figure 15: Salesforce and Budget
New
Small System
is
When
Attractive
Introduced
24.000
34
D-3806
force absorbs an increasing proportion of the sales budget, leaving
for force
expansion. The process reinforces
expansion, revenue growth
is
itself.
less
With lower force
suppressed and the sales budget
grows
itself
less quickly.
POUCIES TO IMPROVE SALES PLANNING AND CONTROL
The
net result of the system's adjustment
introduces an attractive
new product
more revenue per sales hour than
success creates competition
of selling
Particularly
expense, which
an
inflation of selling
orders, revenue
is
has the potential
expense
earn
company
causes an
inflation
and revenue growth.
stemming from the
for salesforce time,
that the
in
to
existing products. But the product's
an extraordinarily succesful product
size of the salesforce, setting
It
line that
turn slows sales
in
The company
quite curious.
for salesforce time that
severe competition
introduction of
is
is
line,
can cause such
forced to reduce the
motion a spiral of decline
sales
in
effort,
and salesforce budget!
useful to reflect on the policy arrangements
most
encourage
likely to
growth of orders for both large and small systems.
is
It
important to avoid excessive competition for salesforce time.
obviously
One
idea
is
to
curb the rate at which compensation planners increase the points value
of
a product whose sales are below
aggressive hiring policy.
quickly
enough
objective.
Another idea
The company should
to allow sales of small
systems
hire
to
is
to
adopt an
account executives
grow without
cutting
35
D-3806
into the time available for selling large
Figure 16
shows a
dramatic change
in
systems.
simulation that incorporates these two ideas. There
on a scale that runs from 40
the scale runs from
12).
The
to
240
to
to
320
fluctuations
eliminated. Orders begin at
smoothly
units per
120
figure 12 orders are plotted
-- in
month, whereas
units per
in
figure 16
units per month, four times the range of
in
70
month
second year, orders continue
to
orders for small systems are entirely
month and grow quickly and
units per
in
the
year of the simulation.
first
grow, but at a slower
orders for large systems decline slightly
grow
a
behavior, which the reader can see by comparing figure
16 with figure 12 (note the scale change
figure
is
until
month
15,
rate.
In
the
Meanwhile,
and then begin
to
slowly.
Figure 17
shows why orders
Compensation per hour
before.
large systems.
to small
systems
for small
new compensation
starts higher than for
to allocate
more
to large
below objective and
of their time
policy allows this re-allocation
because compensation planners are slower
awarded
much more
systems grow so much more than
So account executives begin
systems. The
to continue,
points
for small
to increase the
systems, even though large systems sales are
falling.
The
net result
time selling small systems,
in
is
fact
that the salesforce
spends
more than 30 percent
of
its
time by the end of the run, compared with only 12 percent under the
original
compensation
policy.
From a corporate perspective
this result
—
r
I
D-3806
1
36
2 obj-small
orders_smal!
i]
J}
J)
J}
320.000
240.000
160.000
60.000
mod «l
W
:
_
sVo-»T»
poL
24.000
18.000
12.000
Time
1
2 objJarge
orders-large
^}
200.000
J]
170.000
2)
140.000
y
110.000
D
80.000
———
-1
0.0
1
I
-r—
6.000
,
I
—
12.000
i-nodtl: S^or^'Y^i'
I
'
I
I
18.000
I
I
I
I
I
'
I
I
[
24.000
Time
Figure 16: Customer Orders and Business Sales Objectives
With Improved Sales Planning and Control Policies
T
I
D-3806
37
comp_hr_sm
comp_hr_Ig
1
3
300.000
24.000
X]
262.500
5,}
3
Z esl_time_sale_sm
18.000
225000
J)
3
12.000
187.500
J}
3
6.000
150.000
J)
0.0
T
0.0
6.000
'
r—
'^
'
'
I
I
I
———
'
'
24.000
18.000
12.000
Time
2 atlract—small
frac_efforl_sm
1
1
2
0.500
^
.400
^
1
1
0.375
2
1.200
1
0.250
2
1.000
1
2
1
2
0.125
0.800
0.0
I
I
I
'
I
'
I
0.600
I
'
I
I
I
18.000
12.000
'
'
'
'
I
24.000
Time
Figure 17: Compensation and
Time
Allocation with Improved
Sales Planning and Control Policies
D-3806
38
makes sense. Since
small systems generate
than large systems (by assumption), the
more time
its
more revenue per sales hour
company earns more revenue
the
salesforce spends selling small systems.
Of course, large system sales are depressed, but two factors are at work
to limit the decline. First,
small systems)
is
larger. This factor,
because
more than
by
itself,
total
revenue (from both large and
before, the salesforce budget
allows
is
more account executives
to
therefore
be
hired.
So, even though the proportion of time allocated to selling large systems
is
smaller than before, there are more
because the new
expanded more
available.
hiring policy
is
rapidly than before
more aggressive, the salesforce
The reader can see the boost
budget and
for salesforce).
grows from $2.45
simulation.
million to
In
By contrast
hiring policies, the
in
in
(the figures
sales hours are
use the same scales
budget
almost $2.8 million
steadily from
figure 15,
total
is
salesforce growth clearly by
figure 18, the
The salesforce grows
period.
sales hours available. Second,
and so again, more
comparing figure 18 with figure 15
same
total
in
60
for the salesforce
the
to
for
first
year of the
70 people over the
under the old compensation and
budget grows from $2.45
and the salesforce stays almost constant
at
million to only $2.5 million
60 people
in
the
first
year
of
the simulation.
Several general policy lessons emerge from the model.
that rely
on a
direct sales force to
First, in
businesses
generate orders, the success of a given
D-3806
1
sales-force
39
D-3806
product
40
line
depends not only on customers' preferences, but also on
account executives' preferences. Customers of course choose between the
products of several competing companies
market comparison
of factors
--
such as
a choice based on an
and
price, quality
Account executives however choose between the product
company, a choice based on an
internal,
lines of their
The
distinction
not only the external attractiveness of
but also their internal attractiveness to the salesforce.
competitive prices, a product line
scheme
fail
to sell
if
own
of the
crucial.
is
its
A
products,
Even
with
the compensation
deters account executives from showing the product to customers.
The design
All
may
availability.
non-market comparison
payoff from selling different product lines.
company must manage
external
of the
compensation scheme therefore deserves close
compensation schemes tend
to
attention.
equalize the payoff to account
executives of the company's different product lines
--
if
they didn't then
eventually account executives would allocate 100 percent of their time to
the product line with the highest payoff. This tendency to equalize payoff
must be understood and managed. One doesn't want the equalization
occur too quickly, otherwise the growth of promising
may be
stifled
new product
to
lines
and sales expense may escalate. On the other hand, one
doesn't want the equalization to occur too slowly, otherwise the factories
may be
flooded with orders for the product with the highest payoff, and
salestime for the company's staple product lines
The simulations show two
policy
may be
cut excessively.
parameters that should receive
D-3806
41
management
points
when
--
attention
the speed at which compensation planners adjust
sales are below (or above) objective, and the speed with
which new account executives are hired when the budget authorizes a
salesforce expansion.
When
a company's product lines are changing
frequently, then points should adjust slowly
be hired aggresively
The discussion so
to
and account executives should
encourage sales and revenue growth.
far is indicative of
how
the model can be used for policy
design. There are undoubtedly further simulation experiments that can be
carried out to explore other aspects of sales planning
and
control policy.
For example, one might explore the effect of increasing or reducing the
rigidity of
corporate sales objectives and of increasing or reducing the
size of the 'stretch' objective.
One might
explore the effect of
strict
budget controls on compensation planning. The model should be viewed as a
laboratory for testing
outcome
with
new
management
policy proposals
intuition.
and comparing the simulated
D-3806
42
BACKGROUND READINGS IN SYSTEM DYNAMICS
Forrester J. W. Industrial
Dynamics
,
M.I.T. Press,
Cambridge, MA, 1961.
Forrester J.W. 'Market Growth as Influenced by Capital Investment', Sloan
Management Review, Vol 9, No 2, pp 83-105, Winter 1968.
Hall R.I. 'A
System Pathology
Old Saturday Evening
185-211, 1976.
Post',
an Organization - The Rise and Fall of the
Administrative Science Quarterly, Vol 21 pp
of
,
Lyneis J.M. Corporate Planning and Policy Design:
Approach, M.I.T. Press, Cambridge, MA, 1980.
Mass
A System Dynamics
Model Behavior', System Dynamics Group
Working Paper D-3323, Sloan School of Management, M.I.T., Cambridge, MA,
N.J. 'Diagnosing Surprise
October 1981.
Morecroft J.D.W. and Mines J.H. 'Strategy and the Design of Structure',
Sloan School Working Paper, WP-1721-85, Sloan School of Management,
M.I.T., Cambridge, MA, September 1 985.
'The Feedback View of Business Policy and Strategy',
System Dynamics Review, Vol 1 No 1 pp 4-19, Summer 1985.
Morecroft J.D.W.
,
,
Morecroft J.D.W. 'Strategy Support Models', Strategic
Vol 5, No 3, pp 21 5-229, August 1 984.
Pugh
A.L.
DYNAMO User's Manual,
Management Journal,
Sixth edition, M.I.T. Press, Cambridge,
MA,
1983.
Richardson G.P. and Pugh A.L. Introduction to System Dynamics Modeling
with DYNAMO, M.I.T. Press, Cambridge, MA, 1981.
D-3806
43
Richmond B.M. A User's Guide to STELLA, High Performance Systems, P.O.
Box 811 67, Hanover, NH 03755, August 1985.
Roberts E.B. Managerial Applications of System Dynamics, M.I.T. Press,
Cambridge, MA, 1 978.
44
D-3806
CXXUMENTTATION OFTHE SALESFORCE
TIME ALLOCATION MODEL
Policy Structure of Salesforce
STELLA Diagram
of
STELLA Diagram
of
Allocation
Sales Planning and Control
STELLA Diagram
STELLA Diagram
Time
of
for
Model
Large Systems
Revenue Procedures
of Sales Planning
and Control
for
Small Systems
Budgeting and Hiring (Also Showing Performance
Indicators)
STELLA
Description of Parameter
Equation Listing
Changes
for Simulation
Scenarios
;
45
D-3806
^'' ^4\
.::!,::
,
Effort
4
•••>LARGE
,
Soles
SYSTEMS
r---i. <^-^
j
Fraction of
:::::::»::::::::;'
Soles EHort
to Small
":
SMALL
f—
:::
SYSTEMS::!
Systems
::t:::>
'
•
'
;:'
:x
ij/
'
I
;
>
H
iiiiV
Soles
Plonning S
Objectives
J
:
,::::::::;::::i::::::::::.
»
\
\
Stretch
V
^
::f
:\ Margin
\
/
V—^^JJ
(i^
C^
A-37n?
Figure 19: Policy Structure of Salesforce
Allocation
Model
Time
D-3806
46
Figure 20:
STELLA Diagram
Control for Large
of Sales Planning
Systems
and
47
D-3806
Figure 21
:
STELLA Diagram
of
Revenue Procedures
D-3806
48
Figure 22:
STELLA Diagram
for
of
Small Systems
Sales Planning and Control
49
D-3806
pr.cc o^
S>4.i^«i"i
Figure 23:
STELLA Diagram
(Also
of
Budgeting and Hiring
Showing Performance
Indicators)
D-3806
lI'
50
8C0SS = acoss
+
coss
INIT(acoss) =
ilj
asr = asr + casr
INIT(asr) = (coss*apss)+(cols*ap1s)
n
C
D
bols = bols + cbols
INIT(bols) = ((tse*.5)+(sels*.5))/etsls
boss = boss + cboss
INIT(boss) = sess/lsss
frs = frs + cf rs
INIT(frs) = .2
[Z
f sess
=
f sess
INIT(fsess) =
n
D
O
.1
pis = pis + cpis
INIT(pls) =
25000
pss = pss + cpss
INIT(pss) =
5000
sf = sf + csf
INIT(sf) =
O
O
O
C
+ cfsess
60
8CS = tSC/sf
aplsr 100000
opss
=
20000
QSf = bsf /acs
bsf = asr*f rs
O
G
bsols = bo1s*(l+sm)
bsoss = boss*( +sm)
1
casr
G
=
(sr-asr)/tasr
cbols = (co1s-bo1s)/tecso
cboss = (coss-boss)/tecso
O
cf rs = (rfrs-frs)/tebf
cfsess = fsess*cfpa*(1-f sess)
C' chls r ph1s*dvp
G' chss = phss*dvp
O
G
cols = sels/tsis
coss = sess/tsss
STELLA
Equation Listing of Salesforce Time Allocation Model
3
D-3806
O
51
cpls = (ipls-pls)/tcp
cpss
(ipss-pss)/tcp
=
csf = (asf-sf)/tasf
L^..')
dvp =INIT(bsf)/((INIT(coss)*INIT(pss)WlNIT(cols)*
INIT(pls))*(1+igb))
O
O
6
O
O
etsls = tsls
etsss = lsss*wts+xtsss*(1-wts)
igb =
.1
ipis =
p1s*mppls
ipss =
pss*mppss
pass = (pss/etsss)/(p]s/etsls)
O
O
G
O
O
O
phis = pls/etsls
phss = pss/etsss
pols = cols/bsols
poss
coss/bsoss
=
rcrls = rhls/chls
rcrss = rhss/chss
C' rfrs = tsc/sr
rhls = apls/etsis
O
O
O
O
O
O
O
O
O
O
O
O
C
O
O
O
O
rhss = apss/etsss
sels = tse*(1-fsess)
sess = tse*fsess
shsm
sm
=
1
50
=
sr = (coss*apss)+(co1s*apls)
tasf = 3
tasr = 6
tcp = 3
tebf =
24
6
tecso =
tsc = ((coss*pss)+(cols*pls))*dvp
tse = sf *shsm
tsls =
tsss
=
75
1
xtsss = 13
STELLA
Equation Listing Continued
52
D-3806
cfpa = graph(pass)
=
grapKposs)
0.500 -> 1.500
0.0 -> -2.500
i'-'")
mppss
0.200 ->- 1.700
0.400 -> -1.100
0.600 -> -0.600
0.600 -> 1.450
0.800 -> -0.250
0.900 -> 1.100
1.000 -> 0.0
1.000 -> 1.000
1.200 -> 0.250
1.100 -> 0.960
1.400 -> 0.600
1.200 -> 0.930
1.600 -> 1.100
1.300 -> 0.9 ro
1.800 -> 1.700
1.400 -> 0.900
2.000 -> 2.500
1.500 -> 0.900
mppls = graph(pols)
wts
0.700 -> 1.400
0.800 -> 1.250
= graph(acoss)
0.500 -> 1.500
0.0 -> 0.0
0.600 -> 1.450
100.000 -> 0.095
0.700 -> 1.400
200.000
300.000
400.000
500.000
600.000
700.000
800.000
0.800 -> 1.250
0.900 -> 1.100
1.000 -> 1.000
1.100 -> 0.960
1.200 -> 0.930
1.300 -> 0.910
-> 0.200
-> 0.335
-> 0.500
-> 0.800
-> 0.970
-> 1.000
-> 1.000
900.000 -> 1.000
1000.000 -> 1.000
1.400 -> 0.900
1.500 -> 0.900
STELLA
Equation Listing Continued
D-3806
53
Deflnitions
ACOSS
ASR
BOLS
BOSS
FRS
FSESS
PLS
PSS
SF
ACS
APLS
APSS
ASF
B SF
BSOLS
BSOSS
CASK
CBOLS
CBOSS
CFRS
CFSESS
CHLS
CHSS
COLS
COSS
CPLS
CPSS
CSF
DVP
ETSLS
ETSS
IGB
IPLS
IPSS
PASS
PHLS
PHSS
POLS
POSS
RCRLS
RCTRSS
RFRS
Accumulated Orders fw Small Systems (Orders)
Average Sales Revenue (Dollars/Month)
Base Orders for Large Systems (Orders/Month)
Base Orders for Small Systems (Orders/Month)
Fraction of Revenue to Sales (Dimcnsionless)
Fraction of Sales EffcHt to Small Systems (Dimcnsionless)
Points for Large Systems (Points/System)
Points for Small Systems (Points/System)
SalesfOTce (Account Executives)
Average Compensation Per Salesposon (Dollars/Account Executive/Month)
Average Price of Large Systems (Dollare/System)
Average Price of Smsdl Systems (Dollars/System)
Authorized Saleforcc (Account Executives)
Budget for Salesforce (Dollars/Month)
Business Sales Objective for Large Systems (Orders/Month)
Business Sales Objective for Small Systems (Orders/Month)
Change in Average Sales Revenue (Dollars/Month/Month)
Change in Base Orders fcH- Large Systems (Orders/Month/Month)
Change in Base Orders for Small Systems (Ordcrs/Month/Month)
Change in Fraction of Revenues to Sales (Fraction/Month)
Change in Fracticm of Sales Effort to Small Systems (Fraction/Month)
Compensation Per Hour fw Large Systems (Dollars/Hour)
Con^)cnsation Per Hour for Small Systems (Dollars/Hour)
Customer Orders fw Large Systems (Orders/Month)
Customer Orders for SmlU Systems (Orders/Month)
Change in Points for Large Systems (Points/System/Month)
Change in Points for Small Systems (Points/System/Month)
Change in Salesforce (Account Executives/Mcmth)
Dollar Value of Points (Dollars/Point)
Estimated Time to Sell Large Systems (Hours/System)
Estimated Time to Sell Small Systems (Hours/System)
Initial Growth Bias (Dimensi(xiless)
Indicated Points for Large Systems (Points/System)
Indicated Points fw Small Systems (Points/System)
Perceived Attractiveness of Small Systems (Dimcnsionless)
Points Per Hour for Large Systems (Points^our)
Points Per Hour for Small Systems (Points/Hour)
Performance Against Objective for Large Systems (Dimcnsionless)
Performance Against Objective for Small Systems (Dimcnsionless)
Revenue to Con^)ensation Ratio for Large Systems (Dimcnsionless)
Revenue to Con^)ensation Ratio for Small Systems (Dimcnsionless)
Reported Fraction of Revenue to Sales (Fraction/Month)
Definition of Variable
Names
54
D-3806
RHLS
RHSS
Revenue Per Hour for Large Systems (Dollars/Hour)
Revenue Per Hour for Small Systems (Dollars/Hour)
SELS
SESS
Sales Effort for Large Systems (Hours/Month)
Sales Effort for Small Systems (Hours/Month)
Standard Hours Per Salesperson Month (Hours/Account Executive/Month)
Stretch Margin (EHmensionless)
Sales Revenue (Dollars/Month)
Time to Adjust Salesforce (Months)
SHSM
SM
SR
TASF
TASR
TCP
TEBF
TECSO
TSC
TSE
TSLS
TSSS
XTSS
CFPA.
MPPLS
MPPSS
WTS
Time
to Adjust Sales
Revenue (Months)
Time to Change Points (Months)
Time to Establish Budget Fracti(n (Mcmtfas)
Time to Establish Caporate Sales Objective (Months)
Total Salesforce Conq)ensatioo (Dollars/Month)
Total Sales Effort (Hours/Mrath)
Time per Sale of Large Systems (Hours/System)
Time Per Sale of Small Systems (Hours/System)
Expected Time Per Sale of Small Systems ^ours/System)
Change in Fracticm firom Perceived Attractiveness ( 1/Month)
Multiplier from Performance on Points for Large Systems (Dimensionless)
Multiplier fiiom Performance on Points for Small Systems (Dimensionless)
Weight for Time Per Sale (Dimensionless)
Definition of Variable
Names
--
continued
.
D-3806
55
DESCRIPTION OF PARAMETER CHANGES FOR SIMULATION SCENARIOS
Scenario
1
--
"Test Drive" Balanced Product Portfolio (Model stam_scen1)
Four conditions are required
1
to
produce the test drive scenario:
Large and small systems yield the
Revenue per sales hour
for
a given product
product's price by the time per sale.
must arrange the following
same revenue
So
in
line is
per sales hour.
computed by
dividing the
order to set up condition
1
,
one
equality:
(average price of large systems /time per sale of large systems) =
(average price of small systems/time per sale of small systems)
or (apls/tsis) = (apss/tsss)
The parameter values used
in
the paper were:
apis = $1 00,000 per large system tsis = 75 hours per large system
apss = $20,000 per small system
tsss = 15 hours per small system
With these parameter values, revenue per sales hour
large
2.
is
$1333
for both
and small systems.
Large and Small Systems are Equally Attractive
For condition
2,
systems.
The parameters one must
set to generate this condition are:
Initial
value of points for large systems INIT(pls)
Initial
value of points for small systems INIT(pss)
sale of large systems tsIs
Time per sale
Account Executives
account executives must earn equal points per hour
selling large or small
Time per
to
of small
systems tsss
for
D-3806
56
The parameter values chosen were:
INIT(pls) =
25,000 points per large system
INIT(pss) = 5,000 points per small system
= 75 hours per large system
tsis
tsss = 15 hours per small system
With these parameters, account executives receive 333 points per hour for
selling either large or small
3.
Orders
Consider
for
systems
(e.g. INIT(pls)/tsls
= 25,000/75)
Large and Small Systems Equal to the Sales Objective
first
the case of large systems. For condition 3 to apply to large
systems then customer orders
for large
systems
business sales objective for large systems bsols.
the model's algebra to find the
initial
cols
must equal the
One can
trace through
parameter values that generate
this
equality.
cols = sels/tsis
and bsols = bols *(1+sm) where:
cols
-
customer orders
sels
-
sales effort to large systems
tsIs
business sales objective
-
bols
base orders
So
-
-
for large
for large
systems
systems
stretch margin
for condition 3,
bols = sels*(1+sm)/tsls
If
systems
time per sale for large systems
-
bsols
sm
for large
the stretch margin
and INIT(bols) = sels*(1+sm)/tsls
sm
is
set to zero, which
it
is
in
all
the runs
57
D-3806
described
of
in
the report, then condition 3
base orders
systems
for large
is
satisfied
when
the
initial
value
is:
INIT(boIs) = sels/etsis
4.
Products Priced to Generate a
Condition 4
really
is
priced to generate a
small systems
revenue
is
This sub-condition
is
and Many
Potential
two separate sub-conditions.
profit.
set so that
to the sales
Profit
This
means
Customers
First,
products are
that the price of both large
and
each additional account executive adds more
budget than he takes from the budget
guaranteed by the equation
in
compensation.
for dollar value of points,
dvp
dvp = INIT(bsf)/((INIT(coss)*INIT(pss))+(INIT(cols)*iNIT(pls))*(1+igb))
where:
dvp
bsf
dollar value of points
-
-
coss
pss
budget
-
for salesforce
customer orders
for small
points for small systems
-
customer orders
cols
-
pis
points for large systems
igb
-
-
initial
When
of
.1
systems
the
in all
for large
systems
growth bias
initial
growth bias igb
the simulations
greater than zero
is
shown
in
points dvp takes a value which
(it
is
set at
a value
the report), then the dollar value of
ensures that account executives'
)
D-3806
58
compensation
is
less than the revenue contribution they
budget. This sub-condition ensures that the salesforce
because the sales budget
is
to
more customer
revenue, a larger salesforce budget and more
The second sub-condition, many
if
the
salesforce
is
-
many
initially,
orders,
is
true.
potential
When
more
more
hiring.
customers,
time per sale
growing, then customer orders
words, there are
Scenario 2
grow
is
the time per sale of large and small systems
if
sub-condition
first
will
to the sales
greater than the salesforce expense, so
account executives are hired, leading
automatically
make
will
is
generated
is
fixed
fixed
also grow
~
and
and the
in
other
potential customers.
Introduction of an Attractive
New
Small Svstem (Model
stam_scen2a and stam_scen2b)
Scenario 2
is
generated by lowering both the expected time per sale
of
small systems xtsss and the time per sale of small systems tsss and also
by
re-initializing
base orders
for large
systems bols
INIT(bols).
The new
parameters are shown below.
expected time per sale of small systems xtsss = 13 hours per system
hours per system
in
scenario
1
time per sale of small systems tsss = 13 hours per system
system
in
scenario
(15
(15 hours per
1)
INIT(bols) = ((tse*.5) + (sels*.5))/etsls
(sels/etsis in scenario 1)
59
D-3806
The
reduction of time per sale of small systems has two principal effects.
First,
revenue per sales hour
instead of $1333.
So
of small
account executives
selling small
initially
more
333 points per
hour).
still
of
the
business sales objective
weighted average
large
systems
for large
new
of the orders that
other words
if
systems
initial
large
and the orders
introduced,
systems
less time
is
still
yield
will
it
will
exceed customer
value of base orders
if
is
a
100 percent
of
the total sales effort tse) were devoted to
that
would be expected from the current
inertia in the sales objective for large
is
to
would be expected
sales effort devoted to large systems (sels).
system
systems
find
value of base orders for large systems INIT(boIs) causes
orders for large systems. The
(in
generate $1333 per hour). Second,
attractive to sell than large (large
initial
sales effort
per sales
333 points per hour. So they now
The new
initial
$1538 (20,000/13)
receive 385 points per hour (5000/13) for
systems instead
small systems
is
now generate more revenue
small systems
hour than large (large systems
systems
The idea
systems.
is
When
to capture the
new
the
small
absorb sales time, but the sales objective
for
not decline to account for the fact that correspondingly
being spent selling large systems. The result
is
a
'tension' in
the company's sales objectives.
Policies to
Improve Sales Planning and Control (Model stam_pol)
Three parameter changes are used
to represent the policies to
improve
D-3806
60
sales planning and control.
with the values used
in
The changes are shown below and contrasted
scenario
2.
time to adjust salesforce tasf = 2 months (3 months
time to correct points tcp = 12 months (3 months
in
in
scenario 2)
scenario 2)
time to establish corporate sales objective tecso = 3 months
(6
months
in
scenario 2)
The
reduction
in
the time to adjust the salesforce (tasf) from 3 months to
2 months represents a more aggressive
to correct points (tcp)
The increase
hiring policy.
in
time
from 3 months to 12 months represents a less
responsive compensation planning procedure that only gradually increases
product points
when
sales are below objective.
The
establish corporate sales objective (tecso) from 6
reduction of time to
months
to 3
months
represents a procedure for more frequent review and adjustment of sales
objectives.
Scenarios for Unexpectediv
It
is
Good
or Unexpectediv
Bad New Products
possible to use the model to examine the introduction of
products which are unexpectedly good (they are easier to
expected) or unexpectedly bad (they are more
expected).
To do so one needs
to set the
parameter
difficult
for
sell
new
than
to sell than
expected time per
sale of small systems xtsss at a value that differs from the time per sale
of small
is
systems
tsss.
introduced which
Consider
many people
first
in
an "Edsel" scenario
the
company expect
the market, but which turns out to be a flop.
--
to
a new product
be a success
in
.
D-3806
61
xtsss = 13 hours per system
people expect the product to be easy/quick
--
to sell
tsss = 17 hours per system
the product turns out to take longer to
--
sell
than expected
Note that a small system
takes 15 hours to
sell,
is 'neutral'
(neither
a success or
in
when
it
because then the new system generates the same
revenue per sales hour and the same number
large systems, as
failure)
scenario
of points per sales
hour as
1
Consider next a new small system which turns out
to
be a surprising
success.
xtsss = 15 hours per system
tsss = 13 hours per system
expected.
--
--
people expect the product to be
the product turns out to
sell
'neutral'
quicker than
G83G
U
1
5
Dates' I^EMEMT
=1DflD
DDM
Qfl4
3bl