How can polling be helpful for small businesses?

The era of seat-of-your-pants decision-making is gone. I'm willing to accept the fact that there are some people who just have great instincts, but here is a powerful tool - opinion research, polling - that can either underscore or defy seat-of-the-pants thinking. And so, if it's available and it's scientific, you use it.

What can small business owners use it for?

What is your market, and what does your market want? How much are they willing to spend? Customer satisfaction: Are you doing a good job or a bad job? The work we've done over the years suggests there are two things you're looking at when you're doing customer satisfaction. You're looking for score: good job, bad job. But among the clients that say bad or so-so job -- why.

What's one of the secrets to getting the most out of polling and opinion research?

We've done customer satisfaction for banks, for retailers. You can get a 95% [positive-experience] score either overall or in some specific item. But then when you ask why among the 5%, if there's one person who says, "it was the worst experience of my life, I'll never go back there again," you have to know. You have to find that out. And you only find that out if you ask.

Most polling is done by telephone, but the Do-Not-Call Registry limits business owners when considering this venue for its polls. How do you suggest business owners reach their customers?

You find that even in this era of the Do-Not-Call Registry and people working an average of 50 to 60 hours a week, there are still people willing to answer questions [on the telephone]. But interactive services is one of fastest growing [areas of polling], and very useful and accurate.

Each methodology offers something that another methodology doesn't. The telephone, for starts: We ask our people, don't just give us the yes/no, agree/disagree, the scale 1-5 -- tell me what the tone was. On the telephone, you can get the real attitude, the real tone. E-mail allows you to ask more questions in more detail.

Are there areas where polling works better than others, or those that don't work at all?

Always, customer satisfaction. But also, if you're going to do it yourself, you're not going to get accurate information. Why is that? Because sometimes people don't want to tell the actual vendor. They'll tell us. Sometimes they don't want to tell about a bad experience - you know, do unto others as you'd do unto you.

Can a business be too small and polling irrelevant to them?