Main methods of developing forecasts - Sales Management

There are a number of approaches to forecasting,at the macro level of the total market or industry, and at the micro level of the company down to the individual customers. Forecasting normally needs to be undertake nat both levels, and then a considered approach by the sales manager to the outcomes of macro forecasting and micro forecasting should help come to a reasoned view of realistic and achievable sales forecasts or targets.

These can be looked at in relation to the overall goals and objectives for the market,and the range of marketing and sales strategies and tactics to be employed in marketing plan to ensure their achievement.

Figure illustrates a number of the major macro and micro forecasting techniques,split under two headings that group those dealing with future demand separately

Macro forecasting Macro forecasting involves looking at the overall market for a product or category of products. It is about:

  • studying company historical performance in relation to the environment(political, economic, social, legislative), industry performance,demographic factors related to company or market performance, and product demand
  • relating industry demand to national levels of production, income levels,interest rates, employment, imports, demographic factors.

Micro forecasting The typical approach to market micro forecasting is to:

  • study the performance of each existing and potential customer account on a product-by-product basis over the past few years or recent sales periods and preparing forward sales estimates for the next forecast period
  • build up to territory, area and national sales forecasts for comparison with the independently prepared macro forecasts.


Some of the techniques are quantitative, in that there are often accessible statistical records that will form a firm basis for estimate sand forecasts, and others are qualitative,in that they rely heavily upon requirements of the forecasters and their market knowledge and assessments of trends. The quantitative techniques would mainly be compiled using appropriate computer program sand spreadsheets.

Current demand

The starting point should be to make some estimate of market potential for each market sector and the overall market for company products, and then see how this is being satisfied, i.e. current levels of demand, and the company share of the current demand can then be factored in to develop marketing strategies to develop the company’s sales and market share.

Total market potential

The market potential is basically an estimate of the total sales potential for all industry suppliers in a market (a macro market forecast).This will invariably be a larger number than current actual levels of demand .Quantitative techniques can lead to estimates of volume or value being made. First the forecaster must make some assumptions about what qualifies a person as a prospective buyer.

It is always difficult to make a reasonable estimate of the number of prospective buyers for a product. The target market should be profiled or qualified in any practical way that relates to product consumption or use, in order that estimates of the number of potential buyers have some rational basis.

For example, a basic profile of the scotch whisky drinker might be a male, over 35,with certain levels of income. If we have knowledge of the average consumption of the typical whisky drinker we can then add this to the equation, and find a first estimate of market potential as in the following table.

With industrial products this kind of macro market estimate is often difficult to develop with any accuracy, although some good estimate scan be built up from micro-level data(discussed shortly).

Starting point is often to attempt to establish the number of persons employed in an industry or market sector that uses the products,and apply a ratio of usage based on experience or rational criteria. Supplier of computers might obtain market estimates of numbers of persons employed in industrial and commercial activities(excluding agriculture) and apply a reasonable ratio (possibly based on experience in similar markets) of computers to employed persons (say one computer per 20persons) to estimate maximum likely market potential.

A more refined approach to estimating total market potential involves taking a base number considered to represent the maximum possible level of consumers, and then applying a sequence of percentage ratios(chain ratio approach) that narrows down the market and to which some value or volume factors can be applied. Readers interested in this technique will find it explainedin detail in more specialist texts on forecasting methodology.

In many developed countries government organizations monitor industry statistics of production and consumer expenditure, and desk research to access this data can help in developing estimates of market potential and market demand for products.

Manufacturing companies are normally classified by standard industrial classification(SIC) codes, and if a company’s industrial products have applications in certain industries then accessing government statistics and lists of companies in appropriate SIC classifications will assist the sales manager in estimating market potential and in subsequent sales forecasting.

Industry sales and market shares

Making quantitative estimates of industry sales and market shares is often easier for consumer goods producers than for industrial products suppliers. There are some starting points in developing industry sales estimates, and competitor shares, including from the following sources:

  • trade association data
  • company accounts for competitors,suppliers and customers
  • market research studies (e.g. A. C. Nielsen market audits, and the numerous other sources of published market research into product categories and industries)
  • import and local production statistics
  • user surveys (a ratio of usage can be developed for survey respondents that might then be applied to other non-respondents).

Multiple factor index method

This approach to forecasting regional demand is most commonly used by consumer goods companies wanting to breakdown national forecasts to regional forecasts,or trying to build national forecasts from known regional data about a product category.

It involves having access to a reasonable range of relevant demographic data on population, in order to build an index relating to demand and potential for growth. The method is probably a little more complex than most sales forecasters will normally choose to involve themselves in, unless for markets where a lot of relevant index data is readily available (often to be purchased or developed by a market research company),and, once again, those readers with particular interest can refer to other specialist texts on forecasting methodology.

In some markets where government departments monitor consumer expenditure(as in the UK), statistics may be available to show national expenditure on a range of product categories and regional variations in expenditure, and sales forecasters may be able to use this data to make forecasts of regional sales as a percentage of national forecast sales.

Market build-up method

This approach to forecasting is commonly used for both industrial and consumer goods as a quantitative technique, building up from the micro level to give an overall macro assessment of sales or forecast. It does require the sales forecaster to have access to lists (e.g.from trade directories) identifying the potential buyers for a product.

With an industrial product, an initial profile can be developed for typical potential users, and use might be made of any listings of companies by standard industrial classification(SIC) codes to identify other companies matching the user profile. If there are a large number of potential customers in the product category then data obtained from a sample of users can provide broad criteria to build market demand estimates (see the earlier section on user surveys).

Future demand

When forecasting the future, a much more difficult period to forecast, many more variables come into play, not least of which must be technological change and changes in consumer or user preference. Much future forecasting activity is directed to macro market forecasts, using both quantitative and qualitative techniques.

Time series analysis

Bearing in mind the target readership for this text, practicing sales managers, we will not attempt lengthy expositions of complex statistical forecasting techniques, most of which are based on analyzing data over a longer time period (time series analyses) but refer interested readers to more specialist texts. We will look briefly at some of the main macro forecasting quantitative techniques.

Time series data studies, which would normally make use of computer facilities such as spreadsheets, include the following main statistical techniques.

  • Trend fitting is a technique where historical actual data is plotted and the trend projected.
  • Moving annual totals or moving annual averages are practical techniques that smooth data in a time series, showing at rend that is not distorted by serious seasonal, cyclical or random fluctuations.

An example of constructing a moving annual total trend follows in this chapter.

A disadvantage of moving annual or moving average techniques is that they do not respond quickly to unexpected or significant changes in the pattern of sales.

  • 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.
  • Forecasting from time series data using standard deviations. Again, this quantitative technique, which can be used at the level of industry or company sales,is probably unlikely to be used by most sales forecasters except possibly within major corporations serving major market sectors.

But it is a technique that can be developed quite easily using a computer spreadsheet package, and also it is technique most suited where there is repetitive seasonal sales pattern (i.e. monthly or quarterly). The objective is to measure the seasonal fluctuations,typically by month or quarter of the year,in terms of their deviation from the average trend, then to project the trend forward, adding back in the seasonal deviation factors for each quarter.

Statistical demand analysis

The weakness in time series forms of analysis and forecasting is that it treats sales only as 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, e.g.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, but we will only mention them here by way of an introduction,leaving the reader to study specialist texts where appropriate.

These causal studies, often found in use only in the largest companies, include:

  • 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.) and a line of regression (line of ‘best fit’) describes in quantitative terms the underlying correlation between any two sets of data
  • econometric models, which look at the interdependent relationship of a number of factors that affect sales and profits
  • input/output models, which might be very useful for ingredients or component forecasting where projections will encompass the demand for inputs in relation to outputs in user industries. This technique requires some knowledge of the expected level of output from users of the forecaster’s products.

Market sales tests

Normally in market sales tests for new consumer products, the products are distributed into a limited geographical sales area whose consumer profile matches closely the national pattern, and where it is judged that sales performance can be monitored closely while any marketing communications can also be tightly focused to the target audience. National forecasts can be extrapolated from the test market results in this quantitative technique.

Expert opinion

In this mainly qualitative approach to macro forecasting, panels of experts on the industry (both internal and external) give their reasoned opinions on trends and estimates of market volumes in this qualitative technique. From the perspective of the sales forecaster,an expert panel might include himself/herself,trade distributors, trade dealers or retails, major end users (for industrial products), marketing consultants, international market researchers, and possibly trade association representatives.

However small or large the panel is, it will have a weakness in that each participant will have limited market exposure and knowledge.
That does not mean the approach should be discarded, but does suggest that it should be only one of the approaches to fore-casting adopted by a company, and not the only one.

Any known historical market or company data should be shared between all the panel of experts, and each expert will add to this from his or her own perspective of market knowledge or insight. The experts should all state the assumptions upon which they are basing their estimates. The sales forecaster can then take account of panel inputs in planning marketing strategies and developing sales forecasts.

In some markets local research firms or specialist economic forecasters will develop macro market models of expected trends and demand levels for certain industries, and larger companies often avail themselves of data from these sources.

Marketer's opinions

In this approach the sales forecaster uses his or her experience to make a primarily qualitative demand in, or sales to,each market sector based upon the known data and historical performance, and a considered view of the trading environment. Winsome instances, particularly with major branded consumer products, market research reports may indicate consumer attitudes,buying patterns, and product preferences,thereby providing help in observing or predicting trends.

The weakness in relying only on judgmental in practice is that they often rely too much on the marketer’s own recent sales data, taking little account of the total market and competitive activity. If an agent or distributor is under-performing against the overall market this may be missed, or even if recognized, might not be addressed through a corrective action plan. As ales manager may be more influenced by his or her distribution or sales limitations in estimating sales potential or demand than by independent data on the market.

Another factor that produces a weakness in basing forward forecasts on sales team opinion sis that the team usually have no reasonable basis to anticipate trends that will emerge from significant technological changes. The more data that sales forecasters have available on markets and existing and potential customers, the more meaningful will beamy sales estimates they produce. Later in this chapter we will look at an example of building up sales estimates from customer sales records.

Surveys of future buying plans

Surveys of future buying plans are also a key means of obtaining information that can bemused in preparing future forecasts of market demand and potential. This is a qualitative technique commonly used for industrial products or larger consumer products (such as household appliances and cars), where users are asked about expected product category purchases to meet their own company needs, and the company estimates its likely share of these user estimates.

Most industrial products have a certain life in use, and user swill need to replace equipment over time,either with an identical item where it is purely a consumable product in production processes or a wearable part, or usually with more recent technology where it is equipment. With industrial products advance knowledge of likely buying intentions and patterns can be important to plan market coverage and marketing strategies.

All rights reserved © 2018 Wisdom IT Services India Pvt. Ltd Protection Status

Sales Management Topics