During our consulting work, we consistently find the best development teams spending ample time in the field talking to their customers to better understand customer requirements. "We do that, too," claim other companies, but they really don't. The best teams have up to 80% of their members participating in field interviews. Sadly, we find companies asserting that such customer contact is unnecessary, impractical, or simply impossible. Let's review the common objections.
Some companies claim that it isn't the job of the engineering department to talk to customers. Rather, if marketing did its job correctly, engineers wouldn't need to contact the customers.
In reality, teams relying on a single individual to speak for the customer inherently subject themselves to being misled. Redundancy in this critical communications link reduces the inevitable noise that misguides development teams. Also, a design engineer wrestling with choices of technical implementation will ask different questions, which, in turn, will expose different data.
Other companies state that conducting field interviews wastes valuable engineering time and delays the project. Actually, two weeks of field interviews doesn't add two weeks to a project. For every day spent in the field, you save at least this much time because of faster decision-making and a more stable product requirement.
Then, there's the cost objection. Companies think that it's just too expensive to send engineers into the field. But do the math on this one. The cost of misunderstanding requirements far exceeds the price of executing field interviews. Remember that a change made late in the design process could cost 1000 times more than a change made early.
Taking another angle, some companies are under the impression that engineers don't possess the proper training to talk to customers, and that it isn't a part of their job. Supposedly, the engineers might irreparably damage the customer relationship by telling the truth. For example, if a customer asks, "Have you ever done this before?" the engineer might bluntly respond, "No." In contrast, a suave marketeer will provide some helpful but misleading answer like, "I assure you that we understand this problem better than anyone else in the world."
In fact, most engineers can acquire social skills. Provide adult supervision and competent coaching during early visits, and let them solo when they're ready. After all, many marketing people were engineers before they turned to the dark side.
Sometimes engineers think that with their vast and deep knowledge of the product, the technically primitive customer couldn't possibly tell them anything new.
Actually, the developer's knowledge revolves primarily around the cost of providing certain functionality. The customer is the ultimate expert on the value of such functionality. You cannot make good design choices without understanding both cost and value.
Finally, engineers sometimes fear that they will have to interview hundreds of customers. But research done by Griffin and Hauser of MIT indicates that 90% of all requirements will be identified within the first 30 interviews.
Of course, if you really don't want to conduct customer interviews, any excuse will be sufficient to stop you. But if you're serious about improving product development, invest time in field interviews.