Never invite me to a cocktail party. Since a customary icebreaker is to feign interest in what the person you just met does for a living, you would place each of your guests at risk of having to hear me rhapsodize, ad infinitum, about the virtues of direct mail as a medium for selling.

But today I’m going to rhapsodize about another, lesser-known way to profit from direct mail. Besides being useful for selling goods and services, direct mail is a powerful instrument for peering deep inside the mind of a customer.

Too often, researchers assume that asking people questions is a good way to find out how they’ll act. Nonsense. People cannot predict how they’ll react to your marketing, but that doesn’t keep them from giving you their best guess, and meaning it. Well-meaning respondents have assured researchers that they would buy yogurt named after a chic magazine, frozen meals named after a toothpaste, underwear named after a disposable pen logo, children’s clothing named after a hamburger chain and aspirin named after a menthol salve. Each of these products flopped.

And just as human reactions can be unpredictable, human behavior changes when people know someone is watching. If you don’t believe me, consider any personal habit you indulge in private that you wouldn’t in public. (If you won’t own up to one, consider the habits of someone you don’t like.)

As a research tool, direct response mail eliminates both problems. Respondents don’t try to predict their behavior; they simply act. And since results are privately tallied from afar, no one is influenced by the idea that someone is “watching.”

You needn’t limit direct mail testing to the mail order business, particularly in stressful economic times when you need to get the most from your marketing investment. You can use direct mail to address larger strategic questions. Here are three examples:

• Defining a point-of-purchase strategy. For years, a popular monthly digest that depended on point-of-purchase sales featured only a table of contents on its cover, with a few titles in large type at the top. Featuring the right titles was crucial to point-of-purchase sales. How did the publication know which titles to showcase? Here’s the secret. By mail, it offered to send respondents any three articles from an offering of 20, free. Then it would feature the three most-requested articles on the next cover.

• Gauging brand strength. One of my agency’s clients, a national brand in its own right, had been acquired by a larger national brand. A debate soon emerged: Should the smaller brand retain its identity or take on the identity of the parent company? We divided a representative sample of the market into three groups. Over a few months, we sent each group a series of direct mail offers. One group’s mail featured the larger company’s brand, another group’s featured the smaller company’s brand, and, as a control, the final group’s featured a local company’s brand. Otherwise, the direct mail for all groups was identical. When we counted resultant sales, we found that changing the offer impacted response a good deal, but that changing the brand made no difference. At this point, both sides of the debate had to face the unhappy — but important — possibility that their brand wasn’t as powerful as they’d supposed. (Note: This was not a test of brand power in general. But if you dare, a test like this can provide a valuable reality check.)

• Finding the most compelling claim. One late, great marketing expert related the story of how he was working on a newspaper campaign to promote a food product. He had narrowed his choices to two overall claims: One had to do with providing good nutrition for the family; the other with receiving praise for serving a tasty meal. In focus groups, moms (the target in those days) overwhelming endorsed the nutrition theme. Ever wary, the marketer ran a direct mail test. The theme focusing on praise produced far more orders. The marketing dumped the nutrition idea and successfully rolled out the campaign based on winning praise.

Try it yourself. Even if — heaven forbid — direct mail doesn’t figure into your regular media plan, perhaps you should make room for it in your research plan.

In many cases, you can use direct mail far more effectively, and for less cost, than other research methods to test a proposed, nondirect-response campaign. Divide a representative mailing list of your market into groups. Mail each group a different approach. Include a postpaid Business Reply Card and offer an incentive to those who mail it back. Use the same incentive in all the approaches, but discretely code the reply cards so that you’ll know which approach it accompanied. The approach that generates the most replies is most likely to succeed in the mass media.

It really is that simple. Therefore, expect your research firm to say it isn’t.

At cocktail parties, people who work for research firms avoid me the most.

Steve Cuno heads the RESPONSE Agency in Salt Lake City. He is the author of the book Prove It Before You Promote It: How to Take the Guesswork Out of Marketing and a popular convention speaker for the Direct Marketing Association, the American Marketing Association, the James Randi Educational Foundation and others. He can be reached at Steve@ResponseAgency.com.