Skip to: Main Content / Navigation

  • Facebook
  • Twitter
  • LinkedIn
  • Add This

All Forums > Concept Research

Price Research using Van Westendorp Technique

I'm planning to conduct a concept test and am hoping to get some direction on setting a price for a new product. I've read Keith Chrzan's article in a previous issue of Quirk's and am going to use the Van Westendorp technique. My question is - Do I suggest a "realistic" price range together with the VW pricing questions, or not provide any price range whatsoever? Can anyone comment based on their own experience?

Van Westendorp Pricing


The Van Westendorp pricing model is based upon identifying consumers'/customers' perceptions of the price and value of a product/service of interest.

Yes, you should offer price ranges.
Your price ranges should reflect the market – realistic ranges for highs, lows and everyday pricing. Utilize the marketplace or information you have derived from the market to determine the appropriate range of prices. Go beyond what is realistic for your organization on both ends (too low and too high).

Van Westendorp includes four key questions (you could add on to these if you wanted additional insights):

At what price would you say the product is INEXPENSIVE (a bargain)?

At what price would you say the product is EXPENSIVE but worth considering?

At what price would you say the product is TOO EXPENSIVE to consider?

At what price would you say the product is TOO CHEAP to be of value?

vW Model is Unaided

No! The vW model is based on an unaided response to the set of questions.

Check this site for a related paper.\

Three articles on using Van Westendorp Technique

I believe the article William Bailey is referring to is:

Pricing the space program
By William Bailey
Published in May 2001

We also have published the following two articles which discuss the topic. (these are only viewable by registered users)

A global consensus
By Shaan Rotolo and Kerry Cole
Published in December 2005

Data Use: An overview of pricing research
By Keith Chrzan
Published in July / August 2006

You can find all three articles in our article archive or by simply entering "Westendorp" into the site search at the top right side of the page.

Price Research using Van Westendorp Technique

As a long-time user of vW (think Opinion Research Corporation in the 80s), one becomes aware of problems with this approach. First, a significant proportion of participants provide inconsistent responses; for example, their “too inexpensive” price is higher than their “inexpensive” price."

Another issue is that it directs the participants attention specifically to the issue of price, rather than incorporate it as one element of the overall decision process.

At the end of the day, if the most important result is an accurate and unbiased answer, a monadic approach is least biased because none of the participants knows that other prices are being tested or that price is the object of the research.

VW Pricing

I agree that the VW technique should be done unaided. The actual model is more than the four questions listed. There is no reason that you couldn't do the VM unaided first and then aid people later and just ask the "just start to be TOO EXPENSIVE" question. Having used this technique many times with great successes, I can tell a lot by just looking at the "too Expensive" line - not having to ask all of the questions. I would also couple the question with some other elements such as are the respondents interested in purchasing said item, etc. etc.

May you data be solid and your products winners,

Concept Testing

I have been asked to do what might be considered "concept testing" for a product that has yet to be developed but can be described in some detail. This product is actually a software and services solution aimed at finance professionals. In the past we have done qualitative interviews with 15-20 people, but now I would like to try to get a larger sample and do the research in a much more quantitative way. The research needs to help our company predict market adoption and give us the information we need to know how to price the solution. Vendors have suggested conjoint techniques to me, but this doesn't make sense to me because I don't want to do trade offs between features, I just want to present the concept and ask likelihood of adoption and price. I know the best way to measure price is conjoint because it mimics real world circumstances but wouldn't Van West. be best in this scenario.

concept testing Van West vs. Conjoint

The answer depends on the respondent's ability to understand the concept. If the value of the "concept financial software" is easy to understand, and all features will always be included (e.g., software security, reliablity, etc.), then Van Westerdorf may work well, if you understand that for a product launch you will no doubt want to price well above the maximum penetration price to "skim the cream" of early adopters.

Conjoint may be best if you may not include all features. And not all conjoint analysis is equal. As I've posted before, you not only want to run the pricing analysis by all respondents, but especially by those that are most interested in buying as this will show a higher price in most cases and potentially different features will be valued. Many people leave money on the table by not looking at those who are interested (e.g., likely buyers).

Also it makes sense to ask the basic questions like "why" are people interested and "why" would they recommend the product to understand messaging along with pricing - you can't de-couple the value proposition with the price and/or features AND what people perceive as most valuable via your messaging. Concise messaging can be as important as price.

Bob Beaulaurier

Price Research

Actually I think several latest techniques are presented for determination of price research.