The Gabor-Granger method: how to build the survey and calculate the demand curve

Summary:

Pricing a new product is a guessing game until you ask the people who'd actually buy it. The Gabor-Granger method turns that guess into a number by asking a simple, repeatable question at a series of price points, then using the pattern of answers to build a demand curve for your product.

Named for economists Andre Gabor and Clive Granger, who developed the technique in the 1960s, this method has stayed in wide use in market research because it isolates one variable, price, and asks respondents to react to it directly rather than inferring price sensitivity from indirect questions.

The Gabor-Granger model works by showing each respondent a product description alongside a specific price, then asking a single, direct question: would you buy this at this price?

Based on the answer, the survey moves to a higher or lower price point and asks again. The sequence continues until the survey finds each respondent's tipping point, the price at which they switch from yes to no.

Aggregate those tipping points across a full sample, and you get a demand curve: a plot of price against the percentage of respondents willing to buy at that price. That curve gives you a way to estimate revenue at each price point, not just whether a price is "acceptable" in the abstract.

The method's biggest strength is also its most important limitation. Because it evaluates one product in isolation, it tells you a lot about how demand for your product shifts with price, but nothing about how that demand compares to a competitor's product at the same price. Pair it with a competitive pricing analysis if you need the competitive side of that picture.

Choose five price points, ascending from a price you're confident most respondents would accept to one you expect most would reject. Even spacing matters. A ladder of $10, $15, $20, $25, $30 produces cleaner data than an uneven one, since uneven gaps distort the shape of the resulting demand curve.

Every respondent needs the same clear, concise description of the product or feature before they see any price. Keep it factual. Overselling the product at this stage inflates willingness to pay and produces a demand curve that won't hold up once the product actually launches.

Present the lowest price first: "Would you buy this at $10?" If the respondent says yes, show the next price up. Continue until they say no, or until they've said yes to all five.

If a respondent says no to the starting price, some survey designs branch to a lower point to find the floor as well as the ceiling.

Each price appears on its own screen so respondents judge it independently rather than comparing all five prices side by side.

A sample that doesn't match your actual target buyer will produce a demand curve that doesn't match your actual market. Screen respondents so they match the profile of people who'd realistically consider the product, and aim for at least 100 completed responses per segment you plan to analyze separately.

For each price point, calculate the percentage of respondents who said they'd buy at that price or higher. This cumulative percentage is what gets plotted to form the demand curve, covered in the worked example below.

Gabor-Granger fits a few recurring pricing situations well, and a couple poorly.

While it has many applications, there are a couple of use cases for which you should avoid Gabor-Granger.

Here's a hypothetical Gabor-Granger study for a new project-management add-on feature, using an illustrative sample of 200 respondents.

Price ladder: $10, $15, $20, $25, $30

Respondent results (illustrative data):

Reading this table, demand drops sharply between $20 and $25, a signal that $20 sits close to a natural psychological threshold for this feature.

The revenue index (a simplified proxy that multiplies price by cumulative willingness to buy) peaks at $15 in this illustrative dataset, not at the highest price point. 

hat's the core value of the method: it shows you that the price which maximizes revenue often isn't the highest price the market will bear, and it isn't always the lowest one either.

In a real study, you'd replace this revenue index with actual projected unit volume and margin data to get a dollar figure rather than an index, but the shape of the curve, and where it bends, tells the same story either way.

Building the price ladder and question sequence from scratch takes time you don't need to spend. Start from a pricing survey template and adapt the price points to your own product, then use skip logic to build the ascending price sequence described in Stage 3. If you need a larger or more targeted sample to field the study, SurveyMonkey Audience can supply a screened respondent panel that matches your target buyer profile.