Pricing and Revenue
Forecast Model
Multivariate Solutions
Applications and Goals
•
This pricing and revenue forecast model is used primarily to
determine optimal pricing of a product/service, and market share
penetration of a given product at specific price points.
•
Using that information, a model of revenue projection can be
built using simulation methods.
•
The goal is to give clients the tools to make an informed
decision about pricing a given product or service based on the
highest likelihood of maximum revenue.
2
The Monte Carlo Method
•
While it is a relatively straightforward matter to develop
confidence intervals for each of the revenue parameters taken
alone, what is really at issue is the confidence interval for
projected differences taken jointly. Such a problem is best
addressed through the use of so-called Monte Carlo methods.
•
In a Monte Carlo simulation, a model in spreadsheet format is
set up and the cells whose values come from the survey results
are identified (and which are therefore subject to sampling
error). For each of these cells, a distribution of possible values
using the appropriate means and standard errors is specified. A
series of trials is then generated, each one of which represents
a possible outcome of the process.
3
The Monte Carlo Method
(cont.)
•
On average, most of the trials will yield values close to the
mean, since the distributions are typically bell-shaped and high
values for some parameters are likely to be offset by low values
for others. Expected revenue represents a best-case outcome
given the survey results and confidence intervals associated
with the individual parameters. But if the simulation is
performed, say, 1,000 times, such a best-case outcome would
represent only a small fraction of the total trials.
•
The 1,000 outcomes of these trials can be arrayed in a
cumulative distribution, so that the probability of percentage
growth falling into any given interval can be read off as the
number of trials with outcomes in that interval.
4
The Product / Service
•
The product or service must be defined before the survey is
fielded. This model does not test different features for product
construction—that is left to tradeoff methodologies.
•
This example is a cleaning foam that is used in the everyday
cleaning of kitchens and bathrooms.
•
During the survey the product’s features and benefits are
described in detail to the respondent. Possible sale points are:
– $3.19
– $3.49
– $3.79
– $4.09
5
Constructing the Pricing Module
•
•
•
Read the product/concept description.
The pricing module on the questionnaire is constructed as
follows:
‘Given the features of this product, would be willing to pay
$3.49 for it?’
 If yes, then ask, “Would you pay $3.79 for it?”
 If no, then ask , “Would you pay $3.19 for it?”
 Continue with Yes until the respondent says, ‘No.’ Record
the highest Yes.
 Continue with No until the respondent says, ‘Yes.’ Record
that Yes.
 If Yes at top price ($4.09), record top price.
 If No at bottom price, ($3.19), record No Answer.
Rotate the starting points so that all price points begin equally.
6
Predicted Market Penetration
•
Cumulative line chart
– The assumption is that if a respondent is willing to purchase the
product at $4.09, he/she will be willing to purchase that product at
$3.79 and all lower price levels.
Probability
74%
68%
61%
57%
$3.19
$3.49
$3.79
$4.09
7
Input for Revenue Forecast
•
Input based on market penetration estimates taken from survey.
•
To project revenue, information must be supplied by the client:
– Size of potential market (range OK)
• Our market is said to be 2 million foam-using households, +-10%
– Fixed costs, e.g. production, marketing, dist., etc. (range OK)
• Fixed costs are said to be $2.5 million, +-10%
Revenue Forecast
Projected Size of Target Market (Households)
Projected Fixed Costs
Price Point
2,000,000
$2,500,000
$3.19
Market Penetration
Number of Interviews
74%
300
Projected Revenue
$2,221,200.00
8
Revenue Forecast
Probability That Revenue Is Greater Than or Equal to Forecast
$3.19
Probability
100%
$3.49
Probability
1626
100%
1688
2030
2014
2094
2084
80%
80%
2136
2153
2179
2200
60%
60%
2221
2270
2264
2293
40%
40%
2306
20%
2363
20%
2349
2409
2434
2479
2668
0%
1600
2875
0%
1700
1800
1900
2000
2100
2200
2300
2400
2500
Forecast Revenue=$2,221,000
2600
2700
1600
1700
1800
1900
2000
2100
2200
2300
2400
2500
2600
Forecast Revenue=$2,270,000
Estimated Revenue for Cleaning Foam Sales at $3.19 and $3.49 Price Points
9
2700
Revenue Forecast
Probability That Revenue Is Greater Than or Equal to Forecast
$3.79
Probability
100%
$4.09
Probability
1366
1317
100%
1846
1863
1972
1972
80%
80%
2023
2054
2073
2108
60%
60%
2124
2163
2200
2217
40%
40%
2250
20%
2272
20%
2301
2353
2402
2490
2882
0%
2953
0%
1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900
Forecast Revenue=$2,124,000
1600
1700
1800
1900
2000
2100
2200
2300
2400
2500
2600
Forecast Revenue=$2,163,000
Estimated Revenue for Cleaning Foam Sales at $3.79 and $4.09 Price Points
10
2700
Revenue Projection Results
•
At $3.19, mean expected revenue is $2,221,220. One can be
90% sure that revenue will exceed $2,029,800, but only 10%
sure that it will exceed the higher amount of $2,433,867.
•
At $3.49, the expected revenue is $2,269,667. The chances of
exceeding the revenue at $3.19 are 59%.
•
At $3.79, expected revenue is $2,123,800. At this price the
chances of exceeding the revenue at $3.19 are 36%, and at
$3.49, 27%.
•
At $4.09, expected revenue is $2,162,600. One can be 90%
sure this revenue will exceed $1,862,667, but only 10% sure it
will exceed $2,489,800. The chances of exceeding the
projections at $3.19 are 37%; at $3.49, 29%. Revenue at the
latter price is expected to be higher than at $3.79; the chances
of that are 57%.
11
Price the Foam at $3.49 . . .
Mean Expected Revenue at Tested Price Points
x000s
$2,300
$2,250
$2,267
$2,221
$2,200
$2,163
$2,150
$2,124
$2,100
$2,050
$3.19
$3.49
$3.79
$4.09
12