Use Van Westendorp when the acceptable price range is unknown. Use Gabor-Granger when a plausible range or price anchor already exists and the decision is which tested price produces the strongest modeled revenue. Run Van Westendorp first and Gabor-Granger second when both range discovery and price-point optimization matter. Both methods measure stated preferences, not observed transaction behavior. Both methods answer different questions.
Method: Match the test to the unresolved pricing decision. Van Westendorp is primarily a range-discovery method. Gabor-Granger is a price-point method that compares stated purchase intent across researcher-defined prices. A sequential design connects the two without treating either result as observed market behavior.
Start with the decision, not the method
The first decision is whether the team lacks a credible price anchor or already has a plausible range to test. When no reliable anchor exists, Van Westendorp helps identify where prices may feel acceptable. Van Westendorp is especially useful when no competitive anchor exists. When the team has a baseline and needs to compare specific prices, Gabor-Granger provides a better fit for that price-point question. Van Westendorp suits exploratory work; Gabor-Granger suits projects with an existing price baseline.
The distinction matters because purchase intent is stated intent. A respondent saying they would purchase at a tested price is not the same as a completed transaction. A revenue curve produced from those responses is modeled or simulated revenue, not realized revenue.
What Van Westendorp reveals
The Van Westendorp Price Sensitivity Meter uses four price-perception questions. They ask respondents to identify prices that feel too cheap, cheap or a bargain, expensive, and too expensive. The responses are plotted as cumulative curves to identify intersections, including the point of marginal cheapness, or PMC; the point of marginal expensiveness, or PME; the indifference price point, or IDP; and the optimal price point, or OPP. The method uses four price-perception questions and identifies these intersections.
The acceptable range is generally considered the span between PMC and PME. IDP and OPP provide additional reference points within the curve structure. The output is therefore a perception-based range and set of intersections, rather than a record of actual purchases.
Van Westendorp can use open numeric entry or multiple price bins. Its practical design can use open numeric responses or price bins. This makes it useful when the research team needs to explore the shape of price acceptability before selecting a narrower set of prices for testing.
Figure 1. Van Westendorp maps the acceptable price range
Annotated Van Westendorp cumulative curves showing PMC, IDP, OPP, PME, and the acceptable price range
What Gabor-Granger estimates
Gabor-Granger presents respondents with a researcher-defined price ladder and records stated purchase-intent responses at those prices. Those responses support a demand curve and a modeled revenue curve. Gabor-Granger converts a purchase-intent ladder into both a demand curve and a revenue curve. The demand curve describes how stated purchase probability changes across the tested price points. The modeled revenue curve combines each tested price with its associated stated purchase probability.
The output can support comparison among prices inside a plausible range. It does not show observed conversion or realized revenue. For that reason, any revenue peak should be described as modeled or simulated, never as an outcome that customers have already produced.
Gabor-Granger studies commonly use five to seven price points. 5–7 price points should be treated as design guidance, not as a universal requirement. The ladder should reflect the decision being made and include prices that are plausible enough to generate interpretable variation in stated intent. A starting price can also be randomized to help address starting-price anchoring. Randomizing the Gabor-Granger starting price can help address starting-price anchoring.
Figure 2. Gabor-Granger links stated intent to modeled revenue
Two-panel diagram comparing a Gabor-Granger price ladder and its demand and modeled revenue curves
Side-by-side method comparison
The methods differ in question format and primary output. Van Westendorp asks about perceived price thresholds and produces an acceptable range with PMC, PME, IDP, and OPP intersections. Gabor-Granger asks about purchase intent at defined prices and produces a demand curve plus modeled revenue comparisons.
Their best research stages also differ. Van Westendorp fits exploratory range discovery when a reliable price anchor is unavailable. Gabor-Granger fits a baseline-led decision in which the team needs to compare a defined ladder. The first method is more open to respondents' perceptions of acceptable pricing. The second is more dependent on the prices selected by the researcher.
Neither direct method captures competitive substitution or feature trade-offs as conjoint analysis does. Neither accounts for competitive substitution or feature trade-offs. If the decision concerns how customers exchange features, packages, and prices, conjoint is the more relevant method to consider. This does not make Van Westendorp or Gabor-Granger interchangeable with conjoint. They answer different pricing questions.
Use both in sequence when the research questions differ
Many teams run both in sequence when they need to answer both range and price-point questions. First, use Van Westendorp to establish an acceptable range. Then use that range to define the Gabor-Granger price grid. A sequential design can use the Van Westendorp acceptable range to define the Gabor-Granger price grid.
This sequence reduces the risk of choosing a price ladder without a useful range context. The ladder still needs to be designed carefully. If every tested price is too low, responses may be all yes. If every tested price is too high, responses may be all no. Either pattern can leave the modeled revenue curve uninterpretable because the ladder does not reveal useful differences in stated purchase intent.
Figure 3. Choose the test that matches the unresolved pricing question
Decision flow choosing Van Westendorp, Gabor-Granger, sequential testing, or conjoint based on the pricing question
A practical study design checklist
Define the decision before selecting the method. If the decision is range discovery, specify that objective rather than presenting a narrow price ladder prematurely. If the decision is price-point comparison, define the plausible range and the price points to be tested.
Make the product concept clear enough for respondents to evaluate the price perception or purchase intent being measured. Qualify respondents against the target audience defined by the research plan, then apply data-quality checks before interpreting curves.
For Van Westendorp, choose between open numeric entry and multiple price bins deliberately. For Gabor-Granger, use a researcher-defined ladder, commonly containing five to seven price points, and consider randomized starting points to address starting-price anchoring. Analyze the demand and modeled revenue curves together rather than treating a single stated-intent percentage as a transaction forecast.
Finally, interpret the methods together when both are used. The Van Westendorp range provides context for the Gabor-Granger grid. The Gabor-Granger responses then show how stated purchase intent changes across selected prices. The two outputs should not be merged into an assertion about observed conversion.
Limits: stated intent is not observed behavior
Both methods rely on stated preferences. They can be affected by overstatement of purchase intent, anchoring, sample noise, and instability in the intersections produced by the Price Sensitivity Meter. They also lack the competitive substitution and feature trade-off structure that conjoint analysis is designed to address. Both methods rely on stated preferences and do not capture those trade-offs.
A modeled revenue peak is not observed revenue. Illustrative example, Simulated: if a researcher multiplies a tested price by a stated purchase-probability estimate, the result is a modeled revenue comparison expressed in currency per potential purchase opportunity, not a measured sales outcome. No simulated value should be presented as a realized conversion rate, transaction count, or revenue result.
Interpretation: Van Westendorp answers where prices feel acceptable, while Gabor-Granger estimates how stated purchase intent changes across researcher-defined prices and supports modeled revenue comparisons. They are complementary when the research questions differ, but neither alone establishes actual market behavior or competitive substitution effects.
Limits: The evidence supports directional method selection and design guidance. It does not establish causal validity, observed conversion, realized revenue, or a universal optimal sample size. Numeric illustrations remain Illustrative example and Simulated unless directly measured behavior is available.
Next step: match the test to the unresolved question
Next step: Use Van Westendorp when the acceptable range is unknown. Use Gabor-Granger when the range is known or plausible and the decision concerns tested price points. Use both sequentially when range discovery and point optimization are separate needs. Consider conjoint when the decision involves feature trade-offs or competitive substitution.
Sources
- Survey pricing methodologies: Gabor-Granger vs. Van Westendorp
- Gabor-Granger Pricing Method
- Smart Pricing Research: Van Westendorp, Gabor-Granger & Conjoint
- Gabor Granger vs van Westendorp Models
- A look at three survey-based methods for pricing research
- Van Westendorp vs. Gabor-Granger: Price Sensitivity Methods
- Van Westendorp vs. Gabor-Granger