Download "Top 10 Mistakes in Market Research"


How do you know how to price?

price tagHow do you optimize your product pricing? It’s a hot topic, because pricing is so dynamic these days.  There are many approaches; one is to use custom research (as opposed to syndicated data analyses) to assess price sensitivity and calculate price elasticities.   Here’s a summary of  tools, and criteria to use in selecting the one best suited to your issues.

For most marketers, the choice comes down to conducting a Conjoint Study (usually Discrete Choice) or a van Westendorp Price Sensitivity Analysis.  They are quite different in their requirements, and in what they deliver. While Conjoint is the more sophisticated tool, it is complex and costly.  Often, the simple van Westendorp may be an adequate and efficient research investment; for one of our clients, that investment led to strategic price changes across a line that added over $1,000,000 in profit.  Now that’s a great ROI!  Please see my blog on the van Westendorp model for an in-depth explanation of how it works, and graphic output. 

 This topic of pricing research methods is controversial; contact me to discuss your situation, I’ll provide my perspective in more depth, and we can determine the right approach for you. 

 Van Westendorp Price Sensitivity Analysis:

Description: 4 short questions ask respondent to state at what price this product is too cheap to be any good, a bargain, getting pricey, or too expensive.

Advantages:

  1. Intuitive, easy for respondents.
  2. Quick, can embed in surveys with other issues, or ask pricing Q’s for a few products.
  3. Output is easy to interpret, no “black box”.
  4. Economical – not costly to design, execute, or analyze.
  5. Result: optimal price range, can create demand curves.  

Disadvantages:

  1. Not appropriate for products that are so novel, respondent has no sense of pricing at all.
  2. Focused on price, so no tradeoffs with other variables, no competitive context, no market modeling.

 Conjoint Analysis:

Description: There are several versions of Conjoint; Discrete Choice is commonly used for pricing but consult with us or your statistician to see what’s best for you. Price is one of several variables that describe the product. Each variable has several levels. Respondent makes  choices among products that have different levels of the variables. Respondent does this for a small number of iterations.  Result is utilities for price (and the other variables).

Advantages:

  1. Intuitive, easy for respondents
  2. Puts price in context with other product variables, generates a richer knowledge set
  3. Flexible in the number of variables, and levels, you can test.
  4. Can incorporate competitive products, see your product in context
  5. Generates a model with simulator, see the impact of alternative product designs.

Disadvantages:

  1. Takes a lot of survey time.
  2. There are limits to the number of variables, usually 4 – 8.
  3. Costly – a complex design is sophisticated to program, execute, and analyze.

 Gabor Granger Price Wheel:

Description: Start with a high price. Respondent is given a price and asked would you buy at this price?  Move price down and repeat, until get a “yes, I’d buy”.  Can also start at a very low price and move up, until you get a “no, not at that price”.  Output is a curve with the % who’d buy at each price point.

Advantages:

  1. Intuitive, easy for respondent.
  2. Quick, can embed in surveys with other issues, or ask pricing Q’s for a few products.
  3. Output is easy to interpret, no “black box”.
  4. Economical – not costly to design, execute, or analyze.

Disadvantages:

  1. “anchoring” effects due to starting at the low or high end can impact output, in ways that you can’t assess – this reduces the validity, and so the usefulness, of the technique.
  2. Not widely used today, so not a lot of case studies on it.
  3. Not appropriate for products that are so novel, respondent has no sense of pricing at all.
  4. Focused on price, so no tradeoffs with other variables, no competitive context, no market modeling.

 Other Approaches:

There are other quantitative survey research approaches that are used, depending on your situation.  They may or may not provide you with demand curves.  These tend to include pricing as one of the variables being tested, not the sole focus of the study.  Please contact Julia for more information on these:

  • Brand/Price tradeoff: This is used to determine the amount of value the brand name delivers. There are various ways to deploy this, but basically we are looking to see, at what price does a person switch from one brand to another? What’s the price premium various brands will command? 
  • Concept Testing: One can test the same concept at various price points in a monadic survey test design, but this tends to be rather inefficient, can be costly. One benefit is that it actually measures purchase intent, not all methods do.
  • Regression:  Price or perceived value may be part of variables in a regression model, modeling customer loyalty for instance.  This won’t tell you precisely how to price, but it will show how much price is driving loyalty.
  • Other options: This memo is about primary survey research – pricing analyses using syndicated data are another powerful route, if that’s an option for your business.

 Key questions to determine what pricing research technique makes sense for your business situation:

  1. What’s at risk? If the business risk or opportunity is substantial, you may want to invest in a more sophisticated Conjoint approach.
  2. What variables are in play? If you are designing a new product, a Conjoint that includes price may be appropriate.  If you are focused on price, the van Westendorp may suffice. 
  3. What’s your budget? If the Brand situation simply won’t support a major research investment, a van Westendorp may give you good guidance.

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine

Leave a reply