1

Unit -3

Sales Forecasting & Sales Budgeting

 Introduction
2

 Sales Forecasting is the art of estimating future sales demand by

anticipating what buyers are likely to do under a given set of conditions.
 Thus, it must identify its competitors and estimate their sales.
 Forecasting is a difficult area of management.
 Most managers including sales managers believe that they are good
at forecasting. However, forecasts made usually turnout to be wrong.
Definition of Sales Forecasting- is an estimate of sales in a future period under a particular marketing
program and an assumed set economic and other factors outside the unit for which the forecast is
made.
 Sales Forecasting & Monitoring
3

Sales forecasting is all about:

 Estimating of a company’s sales for a
specified future period.
 Estimating the total size of the potential market
for a product or group of related product
 The market potential is basically an estimate of
the total sales potential for all industry
suppliers in a market (a macro market forecast)
.
 Estimating the current level of total market
demand (normally lower than the market
potential) to specified company selling is called
sales potential.
 Sales Forecasting Time Spans
4
 Types of Sales Forecasting
Methods
5

 Forecasting methods are classified into two major

groups:
 Types of Forecasting Methods
6

 Qualitative methods – judgmental

methods
Forecasts generated subjectively by the
forecaster
Educated guesses
 Quantitative methods – based on
mathematical modeling:
Forecasts generated through
mathematical modeling
 Types of Forecasting Methods
7

 Qualitative methods – judgmental

methods
Forecasts generated subjectively by the
forecaster
Educated guesses
 Quantitative methods – based on
mathematical modeling:
Forecasts generated through
mathematical modeling
 Qualitative Methods
8
 Qualitative Methods
9
 Qualitative Methods of Sales Forecasting
10
 Qualitative Methods
11

 The processes may include the following

steps:
A group of experts is sent a questionnaire by
mail and asked to express their view.
The response received from the experts are
summarized without disclosing the identity of
the experts, and sent back to the experts, along
with a questionnaire meant to probe further the
reasons for extreme views expressed in the first
round.
The process may be continued for one or more
rounds till a reasonable agreement emerges in
the view of the experts.
 Qualitative Methods
12
 Qualitative Methods
13
 Quantitative Sales Forecasting Methods
14

Time Series Analysis

 Naïve/ Ratio Method
 It is a forecasting method, which is based on the
assumption that what happened in the immediate
past will continue to happen in the immediate future.
 The formula is:

Sales for next year=

Actual sales for this year X Actual sales for this year
Actual sales for last year
 Quantitative Sales Forecasting Methods
15

Time Series Analysis

 Quantitative Sales Forecasting Methods
16

Time Series Analysis

Moving Average Technique/ Exercise
 According to this method, the forecast for the
next period represents a simple arithmetic
average or a weighted arithmetic average of
the last few periods.
 For example, assume that the sales of sugar in
Addis Ababa for the last 6 years are as given
below in quintals per year:
 Exercise-1
17

Year Actual

1991 10,350.00

1992 10,890.00

1993 10,600.00

1994 11,200.00

1995 10,700.00

1996
11,150.00

Required-1- Forecast sales for the next year(1997)

using naïve method.
Required-2- Forecast sales for the next six years
using the simple moving average of the recent
 Solution- Exercise-1
Three-year
18
Year Actual Total Average

1991 10,350.00

1992 10,890.00

1993 10,600.00

1994 11,200.00

1995 10,700.00

1996 11,150.00

1997 11,016.67 =11,200.00+10,700.00+11,150.00 33,050.00 =33,050.00/3 11,016.67

1998 10,955.56 =10,700.00+11,150.00+11,016.67 32,866.67 =32,866.67/3 10,955.56

1999 11,040.74 =11,150.00+11,016.67+10,995.56 33,122.22 =33,122.22/3 11,040.74

2000 11,004.32 =11,016.67+10,995.56+11.040.74 33,012.96 =33,012.96/3 11,004.32

2001 11,000.21 =10,995.56+11.040.74+11,004.32 33,000.62 =33,000.62/3 11,000.21

2002 11,015.09 =11.040.74+11,004.32+11,015.09 33,045.27 =33,045.27/3 11,015.09

 Exercise-2
19

b) Weighted Moving Average Technique

 Ifthe weighted moving average method is used,
weight shall first be assigned to each year's data
for the recent few years.
 The sum of the weights assigned as such shall
sum up to one (1).
 In the illustration below for example, assume a
weight of 0.6 has been assigned for the recent
data followed by weights of 0.3 and 0.1 for the
next recent figures. Forecast sales for the next
four years using the Weighted Moving Average of
the recent three years.
 Exercise -2
20
Year Sales

1992 9,100.00

1993 11,200.00

1994 13,600.00

1995 15,100.00
 Solution- Exercise 2
21
Year Sales

1992 9,100.00

1993 11,200.00

1994 13,600.00

1995 15,100.00

1996 14,260.00 (0.6*15,100)+(0.3*13,600)+(0.1*11,200)

1997 14,446.00 (0.6*14,260.00)+(0.3*15,100)+(0.1*13,600)

1998 14,455.60 (0.6*14,446.00)+(0.3*14,260.00)+(0.1*15,100)

1999 14,433.16 (0.6*14,455.60)+(0.3*14,446.00)+(0.1*14,260.00)

 Exponential Smoothing
22

 Exponential smoothing is where extra weighting is given either to

earlier or to more recent data (depending on which data is seen
as more likely to be representative of the future sales pattern), in
an attempt to take account of more significant changes in the
pattern of sales.
 In exponential smoothing forecast, results are modified in the
light of observed errors.
 In this method, the next period's forecasted sales for a product
are made based on the forecast made previously for this period
and the actual sales for the same period.
 Hence, the difference between the forecasted and the actual
sales results for this period is smoothed for a certain constant (α)
alpha, which is smoothing constant using the following formula:
 Exponential Smoothing
23

FD (t+1) = FDt + α (ADt - FDt),

•FD (t+1) is sales forecasted for the next

period,
•FDt is forecasted sales of period t, and
•ADt is actual sales of time t.
 Exercise -3
24

The value of alpha shall be between zero and

one (including the boundaries of course) usually
some decimal point such as 0.65. If the
difference is minor for instance, it may be zero if
the difference is worth considering it may be
one.
The exponential smoothing method in effect
adjusts the forecasted sales of this period to
forecast sales of the next period for some
percentage of the forecast error (difference
between actual and forecasted sales).
 Solution -Exercise -3
25

 Assuming that sales forecasted for 1996 for

Coffee was 18,000 quintals per year and actual
sales for coffee was 15,000 quintals and
smoothing constant of 0.70, what will be sales
forecasted for the year 1997?

 FD (t+1) = FDt + α (ADt - FDt)

 FD1997 = 18,000 + 0.7(15,000 - 18,000)

 FD1997 = 18,000- 2,100 = 15,900

 Solution -Exercise -3
26

 Assume actual sales for 1996 was 19,800 quintals

instead of 15,000 quintals. Then forecasted sales for
1997 should be as follows:

 DD (t+1) = FDt + α (ADt - FDt)

 DD1997 = 18,000 + 0.7(19,800 - 18,000)
 DD1997 = 18,000 + 1,260 = 19,260
 Statistical Demand Analysis:
27
 The weakness in time series forms of analysis and forecasting
is that it treats sales only as a function of time, taking no
account of any other demand-influencing factors that affect
sales, such as:
 the effects of price and promotion,
 income levels, changes in population or
 the mix of population where relevant,
 age structures, sex mix, education levels, and so on or
 the introduction of new technology or product varieties.
 Various statistical demand techniques (often referred to as
causal analysis techniques) are available, for those with access
to computers and sufficient interest and expertise-large
companies.
 Statistical Demand Analysis:
28
Regression Analysis
 Where equations are developed that relate the volume of sales to a
number of independent variables known to impact on sales
performance (such as:
 Advertising,

 Salesperson call rate,

 Number of distribution points,

 Promotional expenditure,

 Display activity, etc.)

 A line of regression (line of ‘best fit’) describes in quantitative terms the

underlying correlation between any two sets of data.
 This is a very popular sales forecasting tool in practice.
 It involves extrapolating the past trend of sales with identified factor
affecting the sales such as income to project the future consumption.
 It measures the average amount of change in the dependent variable
associated with a unit change in one or more independent variables.
 A simple linear regression
29 model
A simple linear regression model consists of two variables
that are thought to be related. One of these is the variable to
be forecasted and it is referred to as the dependent variable,
its value depends on (is a function of) the value of another
variable. The other variable which will be used to explain or
predict the
In order to value
make of the the
a forecast, dependent variable
two variables is expressed
must be referred to aslinear
in a the
independent variable.
equation of the form:

Y= a + b(x)
Where,
y= sales for the year (dependent variable),
x=Independent variable,
a = intercept of the relationship
b= Slope of the relationship
 Y=a+bX
30


 xy  nx y
x 2
 nx 2

y  b( x )
 Exercise -1
31
Example: KAKI S.C is engaged in whole selling business that imported

Isuzu car for domestic use. Over time the company has found that its sales
volume is dependent on the advertising expenditures. During one executive
meeting, the marketing manager gives his information about the
advertising budget for the Isuzu car. The following are sales and advertising
Month Sales (thousands of Advertising (thousands
data for the past five
units)months. of birr)

1 18 7
2 11 6
3 26 18
4 30 20
5 40 24
 Exercise -1
32

 The marketing manager says that next month the S.

C will spend 30,000 birr on advertising for the
product.
Based on the given information:
 developing linear regression equation (which
show the functional relationship b/n advertising
and sales) and a forecast for the product.
 Solution-Exercise -1
33

Sales (thousands Advertising X2 xy

of units) (thousands of
birr)
Y
X
18 7 49 126

11 6 36 66

26 18 324 468

30 20 400 600

40 24 576 960

Sums ∑y = 125 ∑x = 75 ∑x2=1385 ∑xy = 2220

Averages
 Solution-Exercise -1
34


 Exercise -2
35

To illustrate

the linear sales equation, assume that sales for a commodity

in a town are a function of number of households (HHs).

The equation to forecast sales called the linear regression

equation and the forecast results will be as shown below.

(Note: The formula to compute a and b in the equations are

the same as the linear time series analysis except that the
 ………Exercise -2
36
No of HHs (X) Sales (Y)

80 200

92 210

96 240

98 300

99 340
 Solution-Exercise -2
37

 The linear regression sales equation is Y = – 303.91 +

6.042(X) based on the following computations.
 The regression line minimizes the sum of the
squared vertical differences from the data points
(the various points on the graph) to the regression
line.
 Goodness of fit test indicates the strength of the
relationship between the variables.
 When the regression (co-relation) trend projection
method is used, the most commonly employed
relationship is the linear relationship (one
independent variable).
 Solution-Exercise -2
38

No of HHs (X) Sales (Y) X.Y X2

80 200

92 210

96 240

98 300

99 340

Sums

Averages
 Solution-Exercise -2
39

No of HHs (X) Sales (Y) X.Y X2

80 200 16000 6400

92 210 19320 8464

96 240 23040 9216

98 300 29400 9604

99 340 33660 9801

Sums 465 1, 290 121, 420 43, 485

Averages 93 258 24,284 8, 697

b=

Solution-Exercise -2
40

=  xy  nx y
=
121 , 420  5 * 93 * 258
6.042
b x 2
 nx 2 43 , 485  5 * 93 * 93 =

=
a= y  b( x ) 258  6 . 042 ( 93 ) = -303.91

The results of the sales equation should be

interpreted with diligence:
The explanatory variable must make sense –
There should be plausible relationship between the
dependent and independent variables
 Solution-Exercise -2
41

 The results of the sales equation should be interpreted with diligence:

 The explanatory variable must make sense – There
should be plausible relationship between the dependent
and independent variables
 The right model must be selected
 The results should be interpreted with due care
 Outliers (observations that are very far from the
majority observation) may be disregarded in order to
avoid their effect on the regression results.
 Analysis is difficult to handle manually and requires
using computer software to produce the slope, the
intercept, and the sales equation.
 Other Considerations in Forecasting
42

 Apart from the foregoing commentaries there

are other factors that may warrant consideration
when preparing a forecast, such as:
1) Inflation,
2) Seasonal trends,
3) Cyclical trends,
4) Random fluctuation &
5) Product life cycles.
 Multiple linear Regression
43

Multiple linear regression refers to a statistical technique

that uses two or more independent variables(predictor
variables) to predict the outcome of a dependent
variable(Sales).
Where:
•yi is the dependent or predicted variable
•β0 is the y-intercept, i.e., the value of y when both xi
and x2 are 0.
•β1 and β2 are the regression coefficients representing
the change in y relative to a one-unit change
in xi1 and xi2, respectively.
•βp is the slope coefficient for each independent
variable
•ϵ is the model’s random error (residual) term
 4.1 Sales budget
What is a Sales Budget?
 It includes estimates of sales volume and selling
44

expenses
 Sales volume budget is derived from the company
sales forecast – generally slightly lower than the
company sales forecast, to avoid excessive risks.
 Selling expenses budget consists of personal
selling expenses budget and sales administration
expenses budget.
 Sales budget gives a detailed break-down of
estimates of sales revenue and selling expenditure.
Budget says about the present and future for the
sales.
11/1/2022
 Sales Budgeting
45

 It is the estimate of expected sales volume

and selling expenses for the company’s
products and services, for the budget period.
 The sales manager is responsible for
preparing three detailed budgets.
(i)Sales volume budget
(ii)Selling expense budget
(iii) administrative budget of the sales
department
 Sales Budgeting
46

(i) Sales Volume Budget derived from the sales

forecast, is broken down into:
(a)Product-Wise Quantities :
The average selling price per unit, and sales
revenue.
(b) Territory-Wise:
Quantities to be sold and sales revenue
( c) Costumer-Wise and Salesperson-Wise:
Sales volume quota during yearly, quarterly and
monthly budget time
 Sales budgeting
47

(ii) Selling expense budget

includes expenditures for personal selling
activities.

(iii) Administrative budget of the sales

department
the budget should be include operating
expenses
 Planning of Sales Budget
48

 Sales budgets are often divided into first,

second, third and fourth fiscal quarter
estimates.
 The critical components of a sales budget are
estimated unit sales, price per unit and the
allowance for discounts and returns.
 Estimated unit sales multiplied by the price per
unit equals budgeted gross sales.
 Budgeted gross sales less estimated sales
discounts and returns is the budgeted net
sales for the period.
 Example of a Sales Budget
49

 ABC Company plans to produce an array of

plastic pails during the upcoming budget year,
all of which fall into a single product category.
 Assume the increase demand in the second
half of the year will allow it to increase its unit
price from $10 to $11 and also the company’s
historical sales discounts and allowances
percentage of two percent of gross sales will
continue through to budget period.
 Its sales forecast is outlined as follows:
 ABC Company
Sales Budget
50
For the Year Ended December 31,
2021
Quarter 1 Quarter 2 Quarter 3 Quarter 4

Forecasted unit 5,500 6,000 7,000 8,000

sales
x Price per unit $10 $10 $11 $11

Total gross sales $55,000 $60,000 $77,000 $88,000

- Sales discounts & $1,100 $1,200 $1,540 $1,760

allowances
= Total net sales $53,900 $58,800 $75,460 $86,240
 Method used for deciding sales expenditure
budgeting
51

1.A Percentage Of Last Year’s Sales.

 multiplying the sales volume budget by various
percentages of each category of expenses.
 It assumes that increasing sales will generate increasing
promotion and vice versa,
 whereas the converse might be the remedy, i.e. a cure for falling
sales might be to increase the advertising spend;

2. Parity With Competitors,

 Whereby smaller manufacturers take their cue from a larger
manufacturer and adjust their advertising budget in line with the
market leader.
 It assumes status quo within the marketplace;
 Method used for deciding sales
expenditure budgeting
52

Executive Judgement Method

uses judgement to decide the budgeted selling
expenses for each category.
 Method used for deciding sales expenditure
budgeting
53

3.The Affordable Method

 Where expenditure is allocated to advertising after other
cost centers have received their budgets.
 In other words, if there is anything left over it goes to
advertising.
 It does not really commend it because the assumption is
that advertising is a necessary evil and should only be
entered into when other expenditures have been met.
 It quite often happens in times of company squeezes that
advertising is the first item to be cut because of its
intangibility.
 The cure for the company ailment might rest in increased
promotional awareness.
 4. Objective and task Method
54

 1. Look at the sales volume objective

 2. Task and action are decided that are
required and to be carried out.
 3. Estimate the costs of carrying out the
task.
 Method used for deciding sales expenditure
budgeting
55

5.The Return On Investment Method

 Assumes that advertising is a tangible item
that extends beyond the budget period.
 It looks at advertising expenditures as longer
term investments and attempts to ascertain
the return on such expenditures
 Method used for deciding sales expenditure budgeting
56

6.The Incremental Method

 This method is similar to the previous method;
 It assumes that the last unit of money spent on
advertising should bring in an equal unit of revenue.
 (5) and (6) seem to make sense, but the main
difficulties are in measuring likely benefits such as
increased brand loyalty resulting from such
advertising expenditures, and determining when
marginal revenue equals marginal expenditure.
 In practice, firms often use a combination of methods,
e.G. Methods (d) and (e), when deciding their
advertising budget.
 Method used for deciding sales expenditure budgeting
57

7.The Administrative Budget

Represents the expenditure to be
incurred in running the sales office.
Such expenses cover the costs of
marketing research, sales
administration and support staff.
 Purpose of sales budget
58

 Planning
 Coordination
 Control
59

!! !
o u
k Y
an
Th