The Gabor-Granger pricing method determines the price elasticity of products and services. Developed by two economists, André Gabor and Clive Granger, it has been used since the 1960s. It is particularly useful when:
The price elasticity of demand curve shows customers' willingness to pay for your product at different price points. The steeper the demand curve, the more price-sensitive customers are in relation to your product.
The “revenue vs. price” curve helps identify revenue-maximising price points.
For example, this chart below suggests that the revenue-maximising price is around $24.
Each respondent is given a series of almost identical questions such as “Would you buy product X at price Y?”. In each of the several questions, the price shown to a respondent is different: it is adapted based on the respondent's previous answer with the aim to find the maximum price each respondent is willing to pay for a product.
In this example, we have price points from $50 to $170 (incremented by $10):
You need to specify several price levels (ideally, between 5 and 15 price levels). For example: $10, $20, $30, $40, $50, $60.
In Conjoint.ly, you can add as many Gabor-Granger exercises in a single experiment as you need (to test different products). Gabor-Granger can be used as a separate experiment or an additional question added to conjoint, Van Westendorp, or another survey.
Technically speaking, Gabor-Granger is a type of a randomised sequential monadic test where respondents are sequentially given one option at a time in which they will make a decision upon.
When you need to examine product attributes other than price (e.g., design, quality, power) or you also need to look at competitive brands, it would be more appropriate to use conjoint analysis, where price and brand are only two of the attributes among many. Conjoint studies can also provide more precise estimates for total willingness to pay and are less prone to understating the acceptable price by research participants.
There are two biases in the Gabor-Granger methodology which mostly offset each other: