Smart research

Click here for free tools and know-how materials from the Mud Valley™ strategy & brand marketing community.

Question - What will market research do for me?

Answer - If you have never used market research before, then the chances are that it will do very little for you immediately. Most people start out being very suspicious and sceptical of market research, and it is only with experience that they learn how to use it. However, once this skill has been acquired, using market research will put you on a level playing field with the best of your competition, and ahead of the game with the rest. And, whereas market research was once slow, expensive and inaccurate, modern electronic / Internet techniques bring solutions to your door very much cheaper and faster and, because you can therefore do a lot more of it, you can be much more accurate in your insights and assessments

Key points

Yesterday, market research was:

• Complicated
• Expensive
• Slow
• Inaccurate
• Trying to answer too many questions at once
• Poorly communicated
• Divorced from the business purpose of delivering sales and profits growth

But things are changing. Market research will become (and is already):

1. very fast
2. very cheap
3. more impressionistic than statistical
4. relationship building

And it will remain inaccurate and imprecise in many cases.

Brand marketers today can gain both the published desk-research and the primary market research they need within 36 hours for less than $2,000. That makes it cheap enough and timely enough to lead decisions rather than to lag them.

In more detail…………..

There are many sources of error in market research, of which these are but a few:

1. there is an incorrect analysis of the underlying business problem, so the research is designed to answer the wrong questions
2. the wrong people are designated to be interviewed – or at least some of the wrong people are interviewed, obscuring the real results
3. the right people are interviewed, but their responses are not weighted to reflect the influence they have within the market ($1 buyers are treated equally with $1million buyers)
4. the wrong interviewing technique is used, e.g. in-depth research is sought through postal surveys
5. questions are phrased in a way that they are misunderstood by the interviewees (the language of the market researcher is not the language of the interviewee)
6. the question as designed leads the interviewee into a specific answer
7. the literal meaning of a question is not its psychological meaning – e.g. Total Research’s “Relative Perceived Quality” question (“How do you perceive the quality of the following brand”) has nothing to do with product quality or with quality processes; it measures the overall trustworthiness of the brand
8. terms are used differently by different people – e.g. “graphics” will mean something very different to a printer and to an electronics engineer
9. the interviewee gives an inaccurate answer, either deliberately or by mistake – some interviewees say that they give random answers just to get through the questionnaire. More hidden is the fact that people are very poor at knowing what is important to them (although excellent at judging what they like and dislike), so any stated importance rating is likely to be wrong. If you ask people what brands they use, there is always a level of over-claiming for the brand leader
10. the interviewer misunderstands the interviewee’s response, and writes it down incorrectly
11. the interviewer’s (or interviewee’s) written record is misinterpreted during the data input process
12. the data are not cleaned so as to provide consistent analysis
13. the data are not normalised to reflect cultural biases
14. the analyst simply makes an error in, say, transposing numbers from a table into a chart
15. the wrong statistical treatment is used on the data
16. the right statistical treatment is used, but the results are misunderstood
17. the writer of the report/presentation allows his/her own assumptions to colour the presentation of the data, conclusions and recommendations
18. the audience misunderstands the results because of its own presumptions most of the facts of a presentation are simply forgotten. In market research presentations, too many facts are usually delivered in one session
19. the client brand marketer relays the findings of the survey incorrectly or unconvincingly to colleagues within the decision making unit
20. members of the decision making unit misunderstand, or refuse to understand, the true findings of the research

…..and there are, in fact, very many more sources of error in market research.

The response of traditional market researchers to this problem has been to try to design the errors out of the process. Some errors can be avoided (e.g. the interviewee directly enters the data onto a computer rather than relaying them to the interviewer), but most errors are intractable.

This has led traditional market research process to look something like this:

• it becomes something of an academic process
• market research gurus delight in spinning mysteries around sample size, sample structure and statistical significance. That is before they even get to the complications of questionnaire design, research methodologies and incentives for government employees
• it is still a capital investment. It is planned over a long period to be perfect and, worse, encyclopaedic. The communications industry knows that people can only take in one message at a time. When was the last time you saw a research presentation that only gave you one fact?
• it is painfully slow. In many cases the business decision is taken months before the market research is delivered
• market researchers regularly ambush their client brand marketers with the results. In marketing consultancy work, the consultants spend weeks preparing a consensus around the conclusions, and don’t enter the showdown presentation until every vote is counted. In market research, the clients are shanghaied by the results and, as marketer researchers prefer bad news to good, the clients inevitably react defensively. “I didn’t spend good money for you to tell me I am an idiot!”, as the old cartoon goes
• it must not be used in any direct sense to build relationships and generate sales (MRS Code of Practice)

But things are about to change.

Why?

The reason things are changing is fundamentally to do with technological development.

Market research will always be a flawed process, with scope for many errors, but what it can deliver, thanks to the technology, is very fast, good-enough information.

In the world of smart research, brand marketers:

• link into sites set up to provide country and market statistics. Many of these are free sources of information (try Eurostat for European economic statistics, and Kompass to find the numbers and names of companies related to 23 million products and services in 53,000 product/service classes, or Europages)
• search the Internet for information, rapidly gaining skills in keyword searching (try Google, of course, but also Clusty, which categorises its links)
• have access to market research aggregators who tell them what published research is available (try MindBranch)
• have paid subscriptions to key market knowledge sites, specific to their industry, all of which are searchable online
• have access to their own company’s databases of information that has already been collected, collated and analysed
• can use an e-survey tool that will usually get enough responses within 24 hours to start the analysis

So, given perhaps a couple of hours of searching for published information, a couple more hours to develop and distribute an e-survey, and a responsive advisory panel incentivised to respond quickly to e-surveys, there is no reason why many brand marketers could not take at least an interim well-informed decision within 36 hours for under $2,000.

This compares favorably with a cost of $5,000+ for a relatively small market research program taking typically over 3 months to discuss, budget, commission, execute, and analyse.

Some of the conclusions drawn from such research will inevitably dismay a traditional market researcher, but this approach is very much in line with how the expensive marketing consultants operate – create a working hypothesis, then test it by collecting the minimum amount of data necessary to make people comfortable with the conclusions.

© 2004, Mud Valley ™ brand marketing community.

Related answers