How Much Will People Pay for That?

A new twist on a decades-old analytic technique promises to give marketers a better tool for measuring their customers' willingness to pay (WTP), outperforming common survey-based methods. That's according to a forthcoming Marketing Science Institute working paper by Klaus Wertenbroch and Bernd Skiera, who introduce the procedure and explain how they tested it.

To use the process, researchers first ask consumers at the point of sale to state the most they would pay for a particular product. Then the product's price is set at random — for example, by having someone draw a ticket marked with a price. Participants who offered more than the amount drawn have to use their own money to purchase the product at the randomly determined price. Individuals who offer less have no chance to buy. The authors label the new approach BDM, after social scientists Gordon Becker, Morris DeGroot and Jacob Marschak, who invented a similar method in the 1960s to evaluate attitudes toward risk.

In three studies involving inexpensive consumer goods (a soft drink, a slice of cake, a ballpoint pen), Wertenbroch and Skiera found that BDM displayed significant advantages over price matching (asking consumers to state their WTP) and a choice-bracketing technique (asking consumers to make a series of hypothetical buy/don't-buy decisions across a narrowing range of price points).

When asked directly for their WTP, participants' answers tended to collect around major price points, whereas BDM generated a smoother and presumably more accurate distribution. The BDM distribution also was more stable across different subsamples and showed a stronger correlation with factors that might drive purchasing decisions, such as how thirsty or hungry individuals felt.

Most striking, the BDM estimates of mean WTP were 21% to 59% lower than those produced by alternative methods. “If you simply ask people how much they're willing to pay, they don't think hard enough about the problem,” Wertenbroch explains. “But if you force consumers to put their money where their mouth is, they tend to be more conservative.”

More broadly, says Wertenbroch, BDM mimics the actual decision-making process, and it is relatively simple to implement at real points of purchase, giving it an advantage over lab-based simulations.

Like all market-research methods, BDM is vulnerable to strategic misrepresentation, albeit to a lesser extent. For example, participants may understate their WTP in the hope of influencing future prices.