Sales Forecast

Definition

Estimation of sales, in a future period under an assumed set of economic and other factors

Sales forecast help an organization to determine accurately the market demand for products, customer tastes & usage patterns It predicts, how much of a companys particular product can be sold during a future period under a given market program & assumed set of factors
Sales forecasting, according to Cundiff and Still, is an estimate of sales during a specified future period which is tied to a proposed marketing plan and which assumes a particular set of uncontrollable and competitive forces.

Steps in Sales Forecasting

1. Defining the objectives to be achieved.

2.
3. 4. 5. 6. 7.

Dividing various products into homogeneous groups.

Analysing the importance of various factors to be studied for sales forecasting. Selecting the method. Collecting and analysing the related information. Drawing conclusions from the analysis made. Implementing the decisions taken.

8.

Reviewing and revising the sales forecasting from time to time.

Cont.

Sales Forecasting Methods

Qualitative Methods

User Expectations Jury of Executive opinion Method Sales force Composite Delphi technique Market test

Quantitative Methods

Time series analysis Moving average Exponential Smoothing Regression and correlation analysis

Qualitative Methods..

Users expectation

Normally carried for industrial products, having less number of customers & product is well defined. Here customer requirements are found out by directly meeting the customers Through simples questionnaires

Advantages:

Direct contact with costomers

Disadvantage:

Under/overestimate the requirement without considering the changes in business environment

Sales Force Composite

This method is also called BOTTOM-UP approach or GRASS ROOTS approach Derived by taking an estimate of expected sales in the forecast period from each salesperson

Forecast here is based upon experience and expectations of the sales person

Advantages:

Done by salespeople who are closest to market Detailed estimates(customer,product,territory) salesperson might sometimes over or under estimate The salesperson might not consider the overall environment while forecasting

Disadvantage:

Jury of Executive opinion Method

Oldest,simplest & most widely used method. This method includes getting the views of TOP EXECUTIVES regarding Sales. Sales forcasts are either taken by average of all individual opinions or through discussions

Executive opinions are based either on some forcasting method or based on experience,judment & intuition.

A study of 150 companies found that 86% cos use this method.

Advantages:

Quick and easy Less expensive Popular among small & mediun type cos.

Disadvantages:

Unscientific Subjective Difficult to break down the sales into sub units (region,branches)
inaccuracies may be there as these people are not in direct contact with the market

Delphi Technique
This method is developed by Rand Corporation during late 1940.

Experts(within & ouside the organisation) are asked to forecast the sales of an organization Experts are usually from universities, govt. institutions, industry etc Opinion of all experts are combined and an average figure is taken out Experts are kept informed about the general opinion of the group, so that they an modify their decision This continues till consensus is reached

Advatages:

Useful for new products

Disadvantage:

Difficulty in getting a panel of experts Longer time for getting consensus Break down into product territories is not piossible

Market Test

used for forcasting sales for new product,where no historical data is available.

Here, the product is tested in a limited area to find out about consumer acceptance of the product
Based on sales in that particular market, future sales are forecasted Generally those cities are chosen which represent the country as a whole Customers reaction in that particular market is taken as a base for forecasts of overall sales of the product in the country as a whole

3 major methods used are:

Full blown Test Markets Controlled Test Marketing Slimulated Test Marketing

Full Blown Test Markets

Co. chooses 2 to 6 representative cities,does full promotional campaign,similar to what would be done at National level. Duration varies from few months to 1 yr,depending upon repurchase period of new product buyers surveys are carried out to know about consumer attitude,usage and satisfaction towards the product.

If results show high trails & repurchase rate product is launched nationally If results show high trail & low repurchase rate product is redesigned or dropped If results show low trail & high repurchase rate product is acceptable If results show low trail & low repurchase rate product is left out permanetaly

Controlled Test Marketing:

Co.hires a research firm & gets the panel of stores at specified geographical location Research firm delivers the new product to the panel stores,arranges for the promotion at stores & measures the sales also. Research form also interviews the sample consumers to know the perception about the new product.

Stimulated Test Marketing:

In ths 30-40 shoppers are selected,based upon their brand Familiarity,preferance in a paticular product category eg Babycare They are shown print advertisements & commercials of well know brands and also of new product. The shoppers are given small amt of money & asked to buy any item in the store. Researches co.notes how many buy the new product & how many competing product

Later consumers are intervied to find reasons for buying or not buying,after usage of product their satisfaction Level &repurchase intention New product is not exposed to competitors

Advantages:

Forcasting sales for new products Helps co. to decide whether to launch the product nationally

Disadvantage:

If the repurchase period is long,it is difficult for the co.to wait for results

For industrial products the test marketing is done by:

Alpha testing(within the organisation) Beta testing(outside the organisation)

Example:infosys did beta testing for its banking software,to check if its fit for multiple billion dollar US market Another method cos can use is Industry Trade Shows

Quantitative Methods.

Time Series Analysis/Decompostion method

Here future trends are estimated based on organization's past performance Method normally used for long term forecasts i.e. 10yrs & above Sales are broken down into 4 major components

Trends Cyclic variations Seasonal Erratic events

Sales = T X C X S X I where T=Long term variations, C=Cyclical variations, S=Seasonal changes, I=Irregular/unexpected changes in environment Companies like Coca Cola use this method.

Assuming that previous yrs sales have been broken down as follows:

Growth of 3% due to tech.,population(trend) Increased terroriost activities sales are expected to reduce sales by 5%(erratic event) 10% reduction in sales due to recession in demand(cyclic) 15% increase due to festive season in last quater Sales for 2009 were 956 million

Forecast for 2010 sales are:

Trend component shows sales will be985(956*1.03) Sales reduced due to erratic component,will be 936(985*.95) Sales reduced due to cyclic component will be842(936*.90) Quaterly sales will be 210(842/4) Increase in sales in last quarter 242(210*1.15) Sales in rest of 3 quarters 200 million (842 Conceptually sound Difficult to break the data into various components

Advantages:

Disadvantage:

Moving Average

Sales are forecasted based on sales of previous year. Here average of sales for several periods is used for projecting future sales. when a forecast is developed for next period,the sales in the oldest period is dropped from the average and is replaced by sales in the newest period,hence the name is moving avg Formula:

Sales forcast for nxt yr =actual sales for past 3 or 6 yrs/no of yrs(3 or 6)

If co operates in stable environment 2 or 3 yr avg is most useful If a firm. In a industry operates in cyclic varitions,the moving avg should use data equal to length of cycle

Moving Averages forecast

Year Actual sales 3 yr moving avg 6 yr moving

1997 1998

840 880

1999 2000
2001 2002 2003 2004

864 832
862 948 956

861
858 852 880 922 871 890

Advantages:

Relatively simple Easy to calculate Widely used for short term/medium term sales forcasts

Disadvantages:

Cannot perdict long term sales forecasts accurately Historical data is required Unable to predict the upturn or down turn in market

Exponential Smoothing

It is refinement of moving average method. Under this method greater weightage is attached to sales in recent periods compared to sales of earlier periods Best suited for short term forecasting when market is relatively stable. Usually of great help in updating quarterly forecasts. Sales forcast for next year =(L)actual sales this yr + (1-L)this yrs sales forecast where L is smoothing or probability weighting factor

Sales of 2004 will be 0.2*956+80.8*880=895

Advantages:

Simple to operate Useful when data has a trend or seasonal pattern

Disadvantage:

Smoothing constant is arbitary Long term & new product forcasting is not possible

Ratio/Nave method

Based upon the assumption that what happened in immediate past will happen in immediate future

Sales forcast fo next year=actual sales for this year*(actual sales of this yr/actual sales of lat yr) Sales for 2004 will be 956*(956/948)=964 million

Advantages:

SIMPLE TO CALCULATE Requires less data Good for short term forecasts Accuracy will be less if past sales have fluctuated considerably

Disadvantage:

Regression & Correlation Analysis

These are used for forecasting sales of a firm. Regression analysis is used to identify the factors that influence sales. If there is single independent variable, its called Simple regression analysis and in case of two or more variables its called multiple regression analysis. Simple regression analysis is measured by Least Square Method Y=a+bX

where Y = dependent variable(sales) Where a =the Yintercept value(the value of Y when X is 0) Where b = avg increment of sales change Where X = independent variable

Correlation analysis is used to measure the degree of relationship between sales(Y) due to change in X

Multiple regression Model

When there are several independent variables. YF=a+b1*x1+b2*x2. where x1xn are independent variables

Selecting a forcasting method

Accuracy

For short term forecasts exponential method is accurate.for long term regression analysis is useful

Costs Type of data available Requirement of software Experience of co.

How to increase forcasting accuracy

Use multiple forecasting Methods Identify suitable method

Regression analysis

Obtain a range of forecasts

Minimum estimate Maximum Intermediate

Use software tools

Difficulties associated with Forecasting

Lack of qualified &trained personnel Changing consumer attitudes Fashion & fads Lack of adequate sales history

Factors affecting or Influencing sales forecasting

1.

Business Environment Conditions within the industry Internal Conditions of the business Enterprise Socio Economic Conditions Factors Affecting Export Trade

2.

3.

4.

5.

Basic terms

Market Potential/Industry sales forecast

It is the maximum expected sales of a given product or service for the entire industry in a given mkt for a specific period of time Eg:the mkt potential for Mobile phones in India for the Yr 2010-11 is estimated to be 4 million number

4 major things to be included:

Item marketed eg product service Sales estimate in units/value Description of mkt by geographical area or type of customers A specific time period eg a paticular yr

Market forecast/Mkt size:

It is the expected industry sales for a given product or service at one specific level of industry in a given mkt for a specific period of time. Eg : mkt forecast for Mobile phones in organised sector in india for yr 2010-11 is 700 crore

Sales potential

It is the estimated sales of a given product or service fo a company in a given mkt for a specific period of time Eg: Sales