A slightly more sophisticated form of the Willingness to Pay question (“How much would you be willing to pay for this?”) is the Van Westendorp PSM (Price Sensitivity Meter), which is a battery of four questions. After being presented a description of the product or service, respondents are shown a scale of prices, then asked:
- At what price on this scale would you consider the product a good value? [“Cheap”]
- At what price would you say the product is beginning to get expensive, but you would still consider buying it? [“Expensive”]
- At what price on this scale would the product be so expensive that you would never consider it? [“Too Expensive”]
- At what price on this scale would the product be so inexpensive that you would doubt its quality? [“Too Cheap”]
When these questions are administered on paper, up to 20% of the responses have to be discarded for failing validation. When fielding it online, make certain that the following relationships hold:
- Expensive > Cheap
- Too Expensive > Expensive
- Too Cheap < Cheap
The basic Willingness to Pay question produces unnatural results in that it returns low price points which everyone is willing to pay. In the real world, price is a signal of quality, and there are in fact low prices that are too low to be acceptable: would you buy a steak dinner if it cost $5 or a new car if it cost $5,000? While I would be fine with eating a Grade E steak dinner in a car powered by a go-kart engine, most consumers are more discriminating. One of the advantages of the PSM is that it helps estimate this low end through its “Too Cheap” question.
Peter Van Westendorp’s analysis of these questions was idiosyncratic, and I won’t go into it here, as it has been debunked by Newton, Miller and Smith (1993), David Lyon (2002) and others. When analyzing the results from these questions, instead look for prices that are in the normal range (between Cheap and Expensive) for as many people as possible; if that is too narrow, look for prices in the acceptable range (between Too Cheap and Too Expensive) for as many as possible.
The Van Westendorp question battery has spread by word of mouth ever since it was first presented at an ESOMAR conference in 1976. That said, it offers the illusion of rigor while suffering from many of the same problems as the basic Willingness to Pay question: respondents tend to lowball their answers, and those answers are not likely to reflect actual behavior. As with Willingness to Pay, this technique is best used as an input to further pricing research and a small sample size is sufficient. As David Lyon writes in “The Price is Right (Or Is It?)”, “PSM is used far more widely, and its results taken far more literally, than they should be.”
If you’re struggling with pricing research, I encourage you to attend David’s workshop at the Applied Research Methods conference – where you can also attend my classes on Panel & Community Management and on Pragmatic Social Media Research.
See also:
- Relevant Insights – Making The Case Against The Van Westerndorp Price Sensitivity Meter
- Willingness to Pay