In Made to Stick, The Heath brothers recount a story about Herb Kelleher, the founder of Southwest Airlines, and one of his marketing people. Tracy from marketing says that her customer surveys show that passengers would like a light meal instead of the peanuts offered on their Houston to Las Vegas flight. Kelleher responds, saying, “Tracy, will adding that chicken caesar salad make us the low-fare airline from Houston to Las Vegas? Because if it doesn’t help us become the unchllenged low-fare airline, we’re not serving any damn chicken salad.”
I take the point of that story to be that the customer isn’t always right. Sometimes the customer is completely wrong, or, more precisely, you didn’t ask the customer the right question. If Tracy had asked in her survey whether people would like to have a chicken salad or the absolute lowest fares, I’m guessing she’d have gotten different results. I see these kind of polling errors all the time. My first exposure was when I was in high school. My math teacher, Mr. Burns, was studying to be an auctioneer. Some friends and I went to his first auction. It was an estate auction so there was lots of junk going on the block. At one point, Mr. Burns held up a rusty old set of roller skates and asked for bids starting at ten dollars. He worked his way down to twenty-five cents without getting a bid. At that point, he turns to a lady in the front and asks, “You’d pay twenty-five cents for these old skates, wouldn’t you?” The lady nods her head. Burns shouts, “Sold!”
A recent exposure to this idea of how you ask the question came when I read that well over 90 percent of French citizens are organ donors, while the number in the US is closer to 30 percent. That didn’t seem right so I read on. It turns out that in France, you are automatically an organ donor unless you opt out, while here in the US, you must choose to opt in.
Before you accept the results of surveys, find out how they asked the questions. See what I mean.