Named after economists André Gabor and Clive Granger, who developed it in the 1960s, the Gabor-Granger method measures how demand for a single product changes across a set list of candidate prices, called a price ladder.
Van Westendorp asks respondents to name their own prices. Gabor-Granger instead tests the exact prices your business is considering (say £19, £29, £39, £49 and £59) and maps the demand curve, the revenue curve and the price cliffs between them.
How a Gabor-Granger survey works
You describe one product and set a ladder of prices from lowest to highest. At each price shown, the respondent answers a five-point purchase-intent question:
Definitely would buy
Probably would buy
Might or might not buy
Probably would not buy
Definitely would not buy
Full ladder or adaptive ladder
In a full sequential design, every respondent rates every price. With seven prices, that is seven questions, which adds fatigue and lets earlier prices colour later answers.
In an adaptive ladder, the survey searches the ladder:
The respondent starts at a price in the middle of the ladder.
If they would buy, the survey moves to a higher price; if not, to a lower one.
It continues until it has found where their answer changes from yes to no.
Most respondents answer three to five prices instead of the full ladder. The prices they were not asked are filled in on one assumption: a lower price never reduces someone's willingness to buy. So someone who would "probably" buy at £39 is counted as at least "probably" at £29. You can switch to the full ladder if that assumption may not hold, for example for a luxury product where a very low price could put people off.
Why stated intent overstates demand
Saying "I would probably buy this" in a survey costs nothing; paying at the till does. The gap between stated intentions and actual purchases is well documented (Juster, 1966; Morrison, 1979; Jamieson & Bass, 1989). If you count everyone who answers "definitely" or "probably" as a buyer (the top-2-box share), the demand curve will overstate what you will actually sell.
Calibration converts each answer into an expected purchase probability. Ordeen's default weights are 0.70 for "definitely would buy", 0.30 for "probably would buy" and 0 for the other three answers. These weights are an assumption, not a law of nature, so Ordeen shows the uncalibrated top-2-box demand alongside the calibrated curve.
Reading the results
Averaging the calibrated purchase probabilities at each price gives:
The demand curve: the share of your target audience expected to buy at each price. Look for cliffs: steps where one move up the ladder loses an unusually large share of buyers.
The revenue curve: price × demand at each step, which shows the revenue-maximising price on your ladder.
Arc elasticity: how sensitive demand is between two neighbouring prices $p_1$ and $p_2$:
$$E = \frac{(D_2 - D_1) \,/\, \tfrac{1}{2}(D_1 + D_2)}{(p_2 - p_1) \,/\, \tfrac{1}{2}(p_1 + p_2)}$$
An elasticity below −1 means demand is elastic between those prices: the higher price loses more in volume than it gains per unit.
Remember that the revenue-maximising price is not the profit-maximising price; that depends on your costs.
How Ordeen runs Gabor-Granger
Ladders of 4 to 9 prices, adaptive by default, with the full ladder available.
Filling in unasked prices carefully. A price cheaper than a respondent's cheapest "yes" takes the answer they gave at that "yes" (so a "probably" stays a "probably"). A price dearer than their dearest "no" takes the answer they gave at that "no". Prices between the two that the respondent was never asked are left out of the calculation for that price.
Calibrated demand by default, with uncalibrated top-2-box demand shown alongside as a dashed line and a table column.
Intervals where the data supports them. Every price answered by at least 20 respondents gets a 1,000-sample bootstrap 95% interval; below that, the result is marked as directional only.
A revenue curve with the revenue-maximising price highlighted, and arc elasticity between neighbouring prices.
Related reading: What is the Van Westendorp Price Sensitivity Meter? · Van Westendorp, Gabor-Granger or conjoint? · What is willingness to pay?
Sources
Gabor, A., & Granger, C. W. J. (1966). Price as an indicator of quality: report on an enquiry. Economica, 33(129), 43–70.
Jamieson, L. F., & Bass, F. M. (1989). Adjusting stated intention measures to predict trial purchase of new products: a comparison of models and methods. Journal of Marketing Research, 26(3), 336–345.
Juster, F. T. (1966). Consumer buying intentions and purchase probability: an experiment in survey design. Journal of the American Statistical Association, 61(315), 658–696.
Morrison, D. G. (1979). Purchase intentions and purchase behavior. Journal of Marketing, 43(2), 65–74.
