Gabor-Granger Pricing Survey Guide — Finding the Optimal Price with a Demand Curve

"What should we charge for this new product?" There are plenty of pricing-research methods, but the most straightforward and direct is the Gabor-Granger method. The mechanics are simple: show a handful of prices and ask, one at a time, "Would you buy at this price?" From the answers you plot a demand curve and calculate which price makes the most money.

No elaborate experimental design, no hierarchical Bayes required. If you get the setup right, you can draw the demand and revenue curves in plain Excel. But that very simplicity — "show a price, ask if they'd buy" — carries its own traps: anchoring (respondents get pulled toward the price you showed) and inflated purchase intent. This guide walks through how to space your price points, how to design presentation order, how to read the demand and revenue curves, and when to reach for Gabor-Granger versus PSM, Conjoint, or MaxDiff — with a practitioner's feel for the work.

1. What the Gabor-Granger Method Is — Measuring Purchase Intent at Each Price

The Gabor-Granger method is a pricing-research technique established by Gabor & Granger (1966). Each respondent is shown several price levels, one at a time, and asked at each whether they would buy (or how strong their purchase intent is).

For a new product, for example, you ask:

"If this product cost $9**, would you want to buy it?" (Yes / No) "And at **$12?" "And at $15?" …

Tally the share of respondents who said "yes" at each price and you get a relationship where the higher the price, the lower the purchase rate — that is, a demand curve. Multiply "purchase rate × price" and you can see which price generates the most revenue.

That plain act of "putting a price in front of people and watching the reaction" is the essence of Gabor-Granger. Because price can act as a quality signal — a finding from Monroe (1973) — a price that is too low can actually depress purchase intent (more on this below).

2. The Demand Curve and the Revenue Curve — "The Price That Sells" Is Not "The Price That Earns"

Gabor-Granger produces two curves. This is the heart of the method.

"The Price That Sells" and "The Price That Earns" Don't Coincide

The lowest price has the highest purchase rate, but it doesn't necessarily maximize revenue. The peak of the revenue curve — the point where purchase rate × price is highest — is the revenue-maximizing price.

For example:

• $9 with a 60% purchase rate → relative revenue 5.4
• $12 with a 50% purchase rate → relative revenue 6.0 ← peak
• $15 with a 35% purchase rate → relative revenue 5.25

Here the highest purchase rate is at $9, but **revenue is maximized at$12**. The strength of Gabor-Granger is that it makes this gap — between "the price that sells" and "the price that earns" — visible.

Note, though, that this maximizes revenue (top-line sales), not profit. To get the profit-maximizing price you have to subtract variable costs from the revenue at each price. For the practical side of tallying and charting, see the survey data visualization guide.

3. How to Space Price Points — Which Prices, and How Many

The accuracy of Gabor-Granger comes down to how you design the price points you present.

Price Range and Increment

• Spread above and below your target price: if you expect around $12, fan out above and below it, e.g.$8 / $9 /$12 / $15 /$18
• Aim for 5 to 7 price levels: too few and the curve is coarse; too many and you overload respondents
• Use meaningful increments: $0.10 steps are far too fine. Increment by an amount people actually perceive as different, scaled to the price band

The Range You Present Constrains the Conclusion

The biggest caveat: you learn nothing outside the price range you present. If you ask between $8 and$18, the revenue-maximizing price can only come out somewhere in that range. Even if the true optimum is $25, you won't see it if you never showed it. The range you set constrains your conclusion, so it's safer to spread wider than you think you need to.

To get a rough fix on the range beforehand, one option is to run a Van Westendorp PSM first to gauge the psychological price band (see Section 6).

4. Presentation Order and Anchoring — Gabor-Granger's Biggest Trap

Gabor-Granger has a structural weakness: bias from presentation order. If you don't design this away, the demand curve gets distorted.

The Ascending / Descending Trap

• Presenting cheapest-first (ascending): as prices climb, respondents start to feel "this is getting expensive" partway through and drop out early
• Presenting most-expensive-first (descending): the first high price becomes an anchor, making the later, cheaper prices look like a bargain and inflating the purchase rate

Either order produces anchoring — respondents get pulled toward the first price they saw.

The Fix: Randomization

The most reliable fix is to randomize the order in which prices are presented, per respondent. If everyone sees the same order, the order effect gets baked into the results; randomize and the influence of order averages out and cancels. For the mechanics of order effects, see question order effects in survey design.

Don't Forget Inflated Purchase Intent

On top of this, stated purchase intent in surveys runs higher than actual buying ("yes, I'd buy" is free to say). Just as with concept testing, read Gabor-Granger purchase rates at a discount to their face value. Focus less on the absolute level and more on the relative slope between prices (elasticity). The handling of purchase intent is also covered in the concept test survey guide.

5. The Too-Cheap Trap — Price as a Quality Signal

Counterintuitively, there can be a range where lowering the price lowers purchase intent. As Monroe (1973) laid out, this is because consumers use price as a quality signal.

• "This price is too cheap. I'm worried about the quality." → purchase intent drops
• Especially pronounced for gifts, cosmetics, professional services, and other goods whose quality is hard to judge in advance

When you plot a demand curve with Gabor-Granger, you may see a range where dropping below a certain price fails to lift the purchase rate (and may even lower it). This is the same phenomenon as PSM's "too cheap to trust the quality" price (the PMC). The demand curve reminds you that "cut the price and it'll sell" is not always true.

6. When to Use Gabor-Granger vs. PSM, Conjoint, and MaxDiff

Gabor-Granger is one pricing-research method among several, and its role differs from the others. Here's how the main pricing and preference methods line up.

• Gabor-Granger: measures purchase intent against price directly and delivers a single revenue-maximizing price point. The simplest and most direct. Well suited to setting the price of a single product
• Van Westendorp PSM: uses four questions to measure the psychologically acceptable price range (the too-expensive / too-cheap boundaries). Good for exploring the price band
• Conjoint analysis: treats price as one of several attributes and measures price sensitivity amid attribute trade-offs. Good for preference that includes competitors and specs
• MaxDiff: measures the priority ranking of elements rather than price itself. Good for prioritizing what to emphasize

Choosing in Practice

• Price band still fuzzy → PSM to grab the acceptable range
• Then pin down the single best point within that range → Gabor-Granger for the revenue-maximizing price
• Want to see relative preference against competitors or feature specs too → Conjoint
• The PSM → Gabor-Granger one-two punch is one of the classic routes to a pricing decision

Gabor-Granger is most efficient when you want a simple optimal price for a single product on price alone. Conversely, if you need to untangle complex preferences involving multiple attributes, Conjoint is the better fit.

7. Editorial Perspective — Five Things Not to Do with Gabor-Granger

From a vantage point of continuously tracking industry cases and the voices of practitioners, here are five accidents that recur with Gabor-Granger.

1. Not Randomizing the Presentation Order

The most common accident, and the one that distorts results the most. Show everyone cheapest-first (or most-expensive-first) and anchoring plus order effects get baked into the demand curve. Always randomize the order of prices, per respondent. A Gabor-Granger study that skips this is structurally biased data.

2. Too Narrow a Range, Missing the Optimal Price

You present only a tight band around the expected price, and the peak of the revenue curve lands at the edge of the range. When the peak sits at an edge, the true optimal price is very likely outside the range. Spread wider than you expect, and design a range where the peak falls near the middle.

3. Mistaking "The Price That Sells" for the Optimal Price

Reporting the lowest price — the one with the highest purchase rate — as the "optimal price." The value of Gabor-Granger lies in the peak of the revenue curve (purchase rate × price). Look only at the purchase rate and grab the cheap price, and you leave revenue on the table. Always plot the revenue curve and decide at the peak.

4. Taking Purchase Intent at Face Value

Loading the demand curve's purchase rate straight into a business plan as the "actual purchase rate." Stated purchase intent in surveys is always inflated. Look at the relative slope between prices (elasticity), not the absolute level. To convert to a real purchase rate, build a coefficient from your own historical results.

5. Conflating Profit and Revenue

Explaining the revenue-maximizing price as the "most profitable price," when what Gabor-Granger delivers is maximized sales (revenue), not profit. Once you factor in cost of goods and variable costs, the profit-maximizing price is often higher than the revenue-maximizing one. Make it explicit that revenue and profit are different, and if needed, plot a profit curve net of costs.

8. Running a Gabor-Granger Survey with the Survey Tool Kicue

A Gabor-Granger survey splits into a phase of "presenting several prices and collecting purchase intent" and an analysis phase of "plotting the demand and revenue curves to find the optimal price." Kicue handles the former.

• Designing the price-presentation questions: you can build a set of questions asking purchase intent at each of several price levels (question types)
• Randomizing the presentation order: as an anchoring countermeasure, it supports a design that randomizes the order of prices per respondent
• Respondent screening: screening questions to narrow to your target customers (screening question design guide)
• CSV export with respondent IDs: outputs the purchase-intent data at each price point in structured form, ready to feed straight into the demand- and revenue-curve calculations

⚠️ What Kicue Can't Do

• No automatic plotting of demand or revenue curves: plotting price × purchase rate and computing the revenue curve's peak (the optimal price) is done by exporting the CSV and working in Excel / R / Python
• No price-elasticity or profit-maximizing-price calculation either: deriving elasticity or a cost-adjusted profit curve belongs in external tools
• Constraints on dynamic price presentation (showing the next price based on the answer): the classic Gabor-Granger staircase (changing the next presented price based on the prior answer) allows simple branching via display conditions, but complex adaptive presentation has to be designed on the external-tool side

As companion reading, Van Westendorp PSM design guide, conjoint analysis in practice, MaxDiff design guide, concept test survey guide, and survey data visualization guide together bring the full picture of pricing research into view: exploring the price band, deciding the optimal price, and measuring preference.

Summary — Six Points for Mastering Gabor-Granger

1. Measure purchase intent at each price directly — the simplest pricing survey. Plot the demand and revenue curves
2. "The price that sells" and "the price that earns" differ — the peak of the revenue curve (purchase rate × price) is the optimal price
3. Always randomize the presentation order — cancel anchoring and order effects. The single most important design choice
4. Spread the range wide — you can't see outside the range you present. Design so the peak lands in the middle
5. Watch for the too-cheap trap — price is a quality signal; there's a range where cutting it won't sell
6. Don't conflate revenue and profit — what comes out is the revenue-maximizing price; profit maximization needs costs factored in

Because Gabor-Granger is as plain as "show a price and ask if they'd buy," anyone can derive the optimal price from a demand curve at a strong cost-to-value ratio — as long as you don't skip randomizing the presentation order and reading the revenue curve correctly. Explore the price band with PSM, decide the optimal price with Gabor-Granger — this one-two punch turns pricing from guesswork into numbers.

If you want to design a pricing survey, why not try Kicue, a free survey tool? From multi-price-level purchase-intent questions and randomized presentation order to respondent screening and CSV export with respondent IDs, you can start building the input data for a Gabor-Granger survey from a single account (plotting the demand and revenue curves, computing the revenue-maximizing price, and calculating price elasticity are handled in combination with Excel / R / Python).