You can ask customers what they want, but it doesn't do to listen

It may sound like heresy, but the customer is not always right. Harassed call-centre and front-line staff know this to be true. But there is one set of customers who are more dangerous than those giving grief to the receptionist. They're the ones who try to help us. We ask them questions for some market research. We tabulate their replies and act on them. And then disaster strikes. The customers are making us do the wrong thing.

It may sound like heresy, but the customer is not always right. Harassed call-centre and front-line staff know this to be true. But there is one set of customers who are more dangerous than those giving grief to the receptionist. They're the ones who try to help us. We ask them questions for some market research. We tabulate their replies and act on them. And then disaster strikes. The customers are making us do the wrong thing.

There are three reasons for this.

* Customers do not always know what they want. We did not know we wanted the internet, frappuccinos, mobile phones or MP3 players until they were offered to us. Research can inhibit innovation. Try the following questions:

"Do you want to pay £2 for a coffee instead of getting it for 50p elsewhere?"

"Do you want to pay £10 to send your mail instead of 28p?"

"Do you want your airline to cancel its inflight service, reduce seat size and pitch, fly from secondary airports at awkward times, and give you neither a ticket nor a seat reservation?"

These questions would have led to dumb disbelief on the part of the consumer. And in the process Starbucks, FedEx and discount airlines would never have got started. In each case they were taking on entrenched competition that should have known what their customers wanted. But the entrenched competition were asking the wrong questions, getting the wrong answers and missing the growth markets. No manager is ever fired for missing a new market. Instead, whole divisions are fired when senior management belatedly discover that upstart competitors have taken their market from them.

* Customers try to be polite to service providers. Or rather, they try to be polite to the market researchers. Most humans do not enjoy conflict or embarrassment, so instead we damn with faint praise. We give three out of five for barely adequate service. We are reluctant to admit that the service we have received was abysmal. If the service was average, we give it four out of five. We are being polite, but we are not telling the truth. Important exceptions are overlooked: the enthusiasts who give the service a perfect five and say they want more are averaged out by the service disasters which score a one and lose the company business.

Most market researchers recognise this problem, although it remains difficult to persuade management that they have a crisis when they are scoring an average of three out of five. For example, I had a problem with one water utility in demonstrating the shortcomings of its service; it had scored more than three on a five-point scale and it felt pretty good about that. I then spent a day with a service crew and a video recorder. The film of an 80-year-old lady in tears because sewage was running through her kitchen as a result of a delayed response to a callout had a more powerful effect on the executive committee than all the mountains of market research. It had an emotional impact that dull statistics and reports cannot achieve.

* Customers try to be rational. We all like to think we are rational. We can justify buying sports cars for the traffic jam and off-road monsters for the city jungle. We can justify impulse purchases in the sales.

My research showed customers thought they were rational purchasers of video cameras: it was all about price and performance. But when I interviewed them as they left a store, they were often unclear about how much they had actually paid for the model after splashing out for warranties, batteries and accessories and sorting out a financing plan. They were also very confused about the relative performance of different models.

They were buying a story as much as a product, but they needed to show they were rational. They wanted a salesman to give them a story they could relate to their friends and colleagues to show they had made a smart purchase. They needed a story like: "I got free financing, I got free connector cables, it was a special discount on a discontinued model, it has a super macro facility that I need."

We can never guarantee to avoid being misled by research. But here are three simple tips to help companies avoid the traps:

* Customers cannot predict what they will want. Spend a day in the life of your customers and see what frustrates them. Michael Bloomberg attacked the financial information company Reuters because he knew how traders worked: they needed analysis, not just data. Akio Morita came up with the Walkman after seeing kids with big sound systems and imagining a better solution for them.

* Beware of polite customers. Use benchmarks, either against the com-petition or against yourself over time. A score of three out of five may look average by itself, but it would be a disaster if the competition scored higher.

* Beware of "rational" customers. Watch the product in use. Watch how people actually go about buying in the store. If you must ask questions, ask about behaviour – not opinion.

Finally, make a real effort with your research. Capital One conducts more than 64,000 micro market tests for its credit card offerings. Each test gives real responses, real data. The result: it's the fastest-growing credit card company because it knows exactly what each customer wants.

Jo Owen is the author of 'Hard Core Management', published by Kogan Page. jo.owen@virgin.net