Management Perspectives: The Do-It-Yourself Guide to Valid Marketing Research, Part 1
Mar 1, 2006 8:00 AM, By Don Kreski
Tools for structuring your own marketing research studies.
A critical problem in the AV industry is a lack of good information. We regularly make decisions that can make or break our companies, but we rarely take a step that would be standard practice anywhere else. That is, we don’t do market research to make sure our decisions are sound.
It’s not a question of expense. It’s very possible to do good research on your own. I know, because I did my own research for years when I was director of marketing at Chicago-area AV integrators.
If you’re going to do your own research, however, there are a few simple steps you can take to ensure that your results are valid and useful. In this article and one to follow, we’re going to outline those steps and hopefully get you started.
Backward marketing research
As an in-house researcher or research director, the first question is what you want to learn. The answer is usually not as obvious as it seems. For that reason, the best place to start your project is at the end, by outlining a hypothetical report. The fact is you won’t get all the information you need unless you know how you’re going to use it. Write down the decisions you need to make, then the pieces of information you’ll need to make them. Consider all outcomes, not just for the one you’re hoping. Work through the consequences of at least two, if not three or four possible research results. For example, suppose you’re considering the church market, and you’re planning to create a survey asking pastors if they expect to buy projection systems. If 80 percent say yes, you’ll ask your salespeople to call on every church in your area. But is that one question enough? What about timeframe? And won’t your salespeople want to know who is likely to be in charge of the decision?
What if only 10 percent plan to buy this year? You may still want to enter the market, but you might not want your guys to call on every church. Now, you need to predict which churches are most likely to buy, and you’ll want to add questions about the size of the church, whether they offer a contemporary service, and perhaps the location.
Alan Andreasen’s “Backward Marketing Research” is an outstanding article on this approach. It’s available from Harvard Business Online.
Looking for ideas
Perhaps another way to look at our hypothetical survey is to ask the question, what are the factors that predict which house of worship is a good candidate for a presentation system?
Generally, to answer that question you’ll want to do some sort of qualitative or secondary research to refine the information you’ll need to make your marketing decision.
There’s probably good research already available on a market as large as HOW. Start with an Internet search to review what’s out there. Be sure to look at InfoComm International, NSCA, Wainhouse Research, and Frost & Sullivan. It’s possible you’ll find everything you need right there.
But perhaps you’re looking at a less-explored market, or you want to refine what you learned from a purchased report. The obvious next step is to talk to some potential customers.
Consider running your own focus groups. Research companies charge thousands to put these groups together, which they’ll gather in a special room and videotape. In essence, they’re bringing five to twelve people together and asking questions. You can do that yourself for next to nothing and get good results.
I like to take a group out to breakfast or lunch and simply audiotape the discussion. You need the tape because you won’t have time to take notes, and you want to capture every word so you can go back and take a fresh look at what was said. When you start the tape rolling, have each person identify himself or herself so you can associate a name with a voice later on. You’ll want to have someone transcribe the tape for you: It’s a lot easier to deal with the conversation if it’s in printed form.
When you choose participants, remember there’s no way you’re going to have a statistically meaningful sample, so forget about randomness. You’re looking for ideas and that’s it. The best participants are people who are opinionated and talkative. It’s generally okay to invite loyal customers, though if you want to know about service problems, find people who are unhappy. Consumer marketers always pay focus group participants, but that’s because retail customers feel little connection with the companies they buy from. I’ve never offered a participant payment. Instead, I’ve always taken the group to a nice restaurant, and I’ve always sent a gift as a thank you afterwards.
It’s important to recognize potential problems with do-it-yourself focus groups:
1. Most company managers will argue with clients rather than listening to their sometimes-disparaging remarks. If you can’t ask a question that matters and simply listen without comment, then you need someone else to moderate your group. The reason focus group rooms have one-way mirrors is not so participants don’t see the company executives. It’s to keep the company executives from screwing up the discussion.
2. There’s a real skill to moving a discussion along, eliciting comments from every group member, and keeping one or two people from dominating. You’re a rare person if you can ask the questions in a detached manner, think quickly enough to move the discussion into unexpected, yet interesting avenues, and keep your opinion out of it.
All that said, it’s likely that someone in your organization has the necessary skills.
Telephone interviews, either short and to the point, or longer “depth interviews” can also be enlightening. Generally they’re easier to set up and cheaper to conduct than a focus group, but of course they don’t allow participants to build on each other’s comments.
The biggest problem with qualitative research is representativeness. Managers like to take the ideas from a focus group and project them out onto a larger customer base, but that’s not valid. To find out how prevalent group members’ opinions may be, you really have to do a survey, and it has to be scientifically structured. Find out how to put together in our April web exclusive.
Questions? You can reach Don Kreski at www.kreski.com/contact.html.