Skip to main content

Price Testing: Gabor Granger in Upsiide

Written by Leigh Greenberg
Updated today

The Gabor Granger method is a common way to determine price elasticity. It can be set up in Upsiide, with some tweaks to the traditional methodology: to accommodate how logic flows in Upsiide, the first price shown is not randomized.

The general process is as follows:

  1. Introduce the concept unpriced, with all relevant information.

  2. Follow with a series of questions showing 5-7 pre-determined price points and ask purchase likelihood, starting at the middle price point.

    1. If a respondent indicates that they would not purchase a concept at the presented price, a lower price is shown and the exercise repeats.

    2. If a respondent indicates that they would purchase a concept at the presented price, a higher price is shown and the exercise repeats.

    3. This process is repeated until their highest price is identified.

Dig Insights’ standard for “would purchase” is defined as T2B (Definitely/Probably would buy), as we find that including “might or might not buy” in the buying group introduces noise and inflates willingness-to-pay. Your organization may define “would purchase” differently. Whatever your criteria, what’s most important is keeping it consistent when making decisions.

To set up your Gabor Granger in Upsiide, you will create your initial question, which everyone will see, with your middle price point:

If they would purchase, we use logic to route them around the following question to one that shows the next highest price:

If a respondent would not purchase, they move without disruption (i.e. without logic) to the next question, which shows the next lowest price:

This process is repeated until the respondent's highest price is identified. This process can also to be applied in an Idea Split if the prices for all items in that split are the same.

Other pricing exercises in Upsiide

If you have another pricing exercise in mind, please reach out to our support chat for our suggestions on your methodology of interest.

Did this answer your question?