Two key questions unlock the door for the strategic plans which marketing directors have to prepare on any product or division within their company.
- Where are we going (with the product/division)?
- How are we going to get there?
In order to answer these questions, answers are needed to a host of supplementary questions and, within the “where are we going?” poser, market size is an important component. Knowledge of market size lies at the nub of strategic planning.
|Market size solves these strategic questions…||By answering these specific questions…|
Of course, a market size calculation alone will not give the red or green light on an investment decision; other research inputs will be needed on the strengths and weaknesses, of the competition, anticipated profit margins and return on capital, etc.
Furthermore, market size calculations are of no benefit in answering the question “how will we get there?” This must be solved by qualitative research designed to show amongst other things, why products are chosen or rejected, how the distribution routes can be penetrated, and which large customers can be won and how?
Before the market researcher decides on the approach to assess market sizing, a decision must be taken on:
- How accurate the assessment should be;
- How fast it is needed
The cost of obtaining the data is, of course, closely linked to both these issues.
Speed vs. Accuracy Of The Market Size Calculation
The speed with which a study is required is usually self evident. A board meeting or an investment committee is likely to indicate the deadline and the researcher must decide the feasibility of being able to complete by this date.
It is the management sponsoring the study rather than the researcher who must decide on the degree of accuracy required and here common sense is a guide. It may be necessary for example, for management to know, not that a market is worth £85m annually, but simply that it is worth over £50m. It may be enough for them to know that a market falls between the upper and lower estimates of £250m to f350m per annum. Equally, there will be occasion, when high levels of precision are needed, perhaps because the investment is large within the context of the total market.
In general a wide tolerance on market sizing is permissible under the following conditions:
- When an investment is very small within the total market
- When the study is a preliminary scan of the market
- When the chief objective is to answer the question “how are we going to get there?” rather than “where are we going?”
On the other hand, high levels of accuracy are sometimes necessary where:
- The investment is large within the total market and the investor aims to achieve a significant share within it
- Market sizes from different years are needed to show a trend
- It is necessary to split out sub cells of the market which could be attractive targets.
In industrial market research it is seldom possible to state the precise accuracy of a market assessment, because methods of arriving at market size are generally subjective and not derived from statistically valid samples. Realistically, most assessments of industrial market sizing have a fairly generous tolerance; + 15 to 20 per cent is the norm, while +10 to 15 per cent is quite accurate. Researchers who claim accuracy levels of +5 per cent will in most cases be deluding themselves and their sponsors.
How To Arrive At Market Size: Getting To A Market Size Formula
Naturally, it is every researcher’s aim to obtain the greatest accuracy at the lowest cost. Unfortunately there are no firm and fixed rules for doing so. For one product the market sizing may be handed on a plate in a Business Monitor or by a trade association; in others the figures will have to be dug out, possibly by a special survey. Assuming it is necessary to assess a market size, here are some steps to consider.
Step One – A Top-Down Approach
Search all government and trade association figures. Using the Central Statistical Office’s Guide to official statistics (available from Her Majesty’s Stationery Office), the search should begin for government publications which contain the market size data. The most obvious starting point is Business Monitor but, depending on the product, it could be another government publication such as Housing & Construction Statistics, Agricultural Statistics, etc.
If it is not immediately apparent that the data are available in published form, it is worthwhile double checking by telephoning one of the government statistical departments.
A search through the Directory of British Associations (published by CBD) will identify any trade associations which could be a repository for sales or production statistics. Cooperation from a trade association, even if it does collect data, cannot be assured. Very often their terms of reference are to maintain confidentiality on all figures for the benefit of members.
Finding a published source of data from a multi client study is like striking gold for the researcher. It is quick, inexpensive and usually accurate. Nevertheless it is worthwhile carrying out a few simple cross checks since government departments (or the firms which supply the data) have been known to make some unintentional errors in their records. The publication Marketsearch from Arlington Press is a comprehensive listing of published market research reports.
The published data which the researcher finds need not necessarily be in a convenient form. However, with a little manipulation it may be possible to link to the subject of interest. There are, for example, no published figures on the market for high quality labels used on bottles of gin and whisky but the sales of spirits are so well documented they could very easily be derived.
If still unsuccessful in assessing the market size, we move on to step two.
Step Two – Looking At The Supply-Side
Build up a picture of the supply side of the market, by adding together the sales of the companies who sell within it.
The researcher is likely to need to consult a variety of sources to obtain suppliers’ sales. A starting point is Companies House where for a small search fee for each company the registrar will send a microfiche of the whole history of the report and accounts. Unfortunately The Companies Act does not require small and medium sized companies to state their tunrover. Even for larger companies which have to submit turnover figures, some adjustment will be necessary. The researcher needs to strip out (often with no greater sophistication than simply making intelligent guesses) those sales which are derived from other products.
Other sources which may provide clues as to a company’s sales turnover may be found on web sites in articles in the trade press or in company’s sales literature. The references may be oblique, referring to the number of employees or productive capacity rather than sales. However they provide a position which can be refined down. With all the guesswork which is applied to this method of market size assessment, it should be clear that accuracy levels may well be hovering around the 20 to 30 per cent mark.
An option which provides a quick and often more accurate means of assessing market size than juggling turnover figures is to find someone (or preferably more than one) with an overview of the market. Insights from these market gurus help sharpen the estimates and add to the researcher’s wider appreciation of the market. Overview interviews can be sought from suppliers themselves; journalists; important distributors and companies selling raw materials or components to the suppliers.
Step Three – Building From The Bottom-Up (Demand Side)
Assessing the demand of consumers (or distributors) of a product usually requires an interview programme. Buyers of products seldom publish figures which can be used as a guide to their purchases – this needs to be elicited from an interview. In the case of a narrow market sector with just a few large and medium sized customers, the researcher may well attempt a virtual census. Where there are many buyers of varying sizes the researcher may sample most of the large ones and a proportion of the medium and small ones to arrive at a picture. Whether telephone of personal interviewing is used (or even a postal survey) depends on the complexity of the data to be collected as well as its cost.
A problem often arises in surveys of demand when the sample data has to be grossed up. The basis for grossing up needs to be reliable. For example, a researcher may gross up by taking the average consumption per company (derived from the survey) and multiply by the number of companies in the market. And yet there are very few markets where we know how many companies there are. Directories are notoriously unreliable as they include double counting, omissions and the inclusion of irrelevant companies.
Another method of grossing up is to obtain employment statistics for the people employed in the industry sector of interest. These can now be used to calculate market size using the following market size calculation:
Market size formula = Total purchases by sample firms X Total people employed in sector
Of course, the above is just one example of a market size calculation that could be used to determine the likely volumes bought or sold in a given sector. In any market size calculation, there are likely to be a series of key variables (or analogues thereof) that when combined together give a sense of market size.
Market assessment can take place within existing markets or new industry sectors or new geographical markets. For many companies, the first place to look for more sales is amongst existing customers. Current customers have already made the ultimate gesture of approval and paid money to buy your products. A bit more persuasion and they may buy more. Also, existing customers know and trust the company sufficiently well to do business. So much so, they may give serious consideration to buying a new product or service from the company. However, every company has a product that can travel. New markets wherever they are – new countries or new segments – carry risk. Market research is a must in making all these decisions.
Readers of this white paper also viewed: