 |
In one of
our recent issues, Ralph Grabowski discussed the benefits of market research
and the success that companies achieve using it as a major tool for product
development. As important as it is, it will be of little use unless it is done
correctly.
Developing a good market research program is not
exceptionally difficult, but it can be time consuming and best done correctly
the first time. Let's discuss how to do it and how to select a good contractor
if you need to outsource it.
There are two kinds of market research
to consider. The first in order of use is generally secondary research. These
are the documents you can purchase or find in libraries. Here others have
researched a market area and compiled some information on trends and sources.
This is a valuable resource for beginning the program. Not only does it give
you a general outline of the market area, it is often a source of contacts, of
whom you may ask more specific questions.
Primary research, which
comes second, will focus on asking the specific questions you need to have
answered to make decisions on the course of your business or product
development. Take some time to explore the objectives thoroughly. Long surveys
are costly and elicit small response numbers. Focus on a few questions that
will provide the information you will really use. Test your questions against
the objectives. Next you are going to need to develop a profile of the product
or service. Don't assume that this is easy because you are the developer. You
don't want to research the product directly, you really want to find out if it
is going to solve some kind of problem. Thus you need to understand thoroughly
what you think you are solving and determine how the research will determine if
it is real. Think about non-technical issues also. Distribution, the sales
model, promotion and price point are all important. Many otherwise good ideas
fail because of these issues.
Sample size and selection are critical
tasks. At the risk of some oversimplification, the more homogenous the
individual purchasers in your target market are, the smaller the sample you
need. Very small sample sizes are often quite statistically valid, easier to
reach and easier to analyze. Selection of the sample demands care to avoid
reaching only early adopters. These users will try anything once and generate
early sales, but frequently move on, leaving you with a product that does not
stimulate long-term market interest.
There are a number of ways to
conduct the research. You can contact the people by mail, by phone or in
person. Each way has its advantages in terms of cost speed and accuracy. Don't
expect to reach everyone. Mail surveys usually get a response of 2 %- 5%. Voice
mail averages four to five calls to reach an individual. In-person interviews
are very consuming of time and money. Focus groups need very objective
facilitators. I've never tried a survey by e-mail. Certain audiences and
offerings may find this method quick, cheap and accurate.
Whatever
the methods, test your research program on a small sub-sample and review the
results before you begin in earnest. It's going to take longer and cost more
than you expect. Persist and be consistent. Ask what you planned to ask.
Improvisation will compromise the integrity of the program.
When you
are done, write a formal report covering all of the preliminary work, the
research itself and an analysis. Make some recommendations. Now, get someone
else to review the report. You are too close to the action. You need an outside
opinion. If they can't make sense of the report and don't agree with the
conclusions, something may be seriously wrong. Work it out.
If all of
this is a bit daunting, you can always hire a consultant to do it for you. This
is very critical and strategic work for you so you need to exercise some care
in the selection. Find someone who is going to be easy to work with and is open
about methods and sources. Ensure that you get the First Team, that they are
flexible and will give you true primary research. Knowledge of market research
techniques, sources of information and contacts are perhaps more important than
detailed domain-specific knowledge.
In summary and stomping my foot
on a couple of points: know your objectives before you start; ask contacts
about problems, not products; and have outsiders review the results.
|
