Most new products are priced by committee, not by consumers. A founder picks a number that feels defensible. A finance team models gross margin targets backwards into a list price. A product marketer benchmarks a competitor and adds 10% for “premium positioning.” The concept launches, sales feedback comes in six weeks later, and the price changes. Pricing concept testing is the practice of finding the right number before that mistake.
The Van Westendorp Price Sensitivity Meter has been the workhorse methodology for this since Dutch economist Peter van Westendorp introduced it at the 1976 ESOMAR Congress. It is simple, it is defensible, and it produces a price corridor with a single optimal point. The reason it has survived five decades is that the four questions force respondents to articulate their own price boundaries rather than accepting an anchor from the researcher.
The catch is that Van Westendorp was designed for surveys, and surveys are the worst possible environment for a pricing question. A respondent writes a number on a screen with no context, no competitive frame, and no pressure to reason carefully. The chart looks clean in the output deck. The corridor is wrong.
This guide covers what pricing concept testing is, how Van Westendorp works, when to use it versus Gabor-Granger and conjoint, why surveys fail at pricing, and how AI-moderated interviews fix the methodology without discarding it. It closes with a setup checklist and a cost comparison against traditional pricing research. Companion reading: concept testing complete guide and concept testing cost breakdown.
What Is Pricing Concept Testing?
Pricing concept testing is the practice of measuring willingness-to-pay for a product, service, or feature concept before it launches. It sits at the intersection of concept testing and pricing research. The input is a concept stimulus (a description, a mockup, a short video). The output is a defensible price point or price corridor that the concept can launch with.
It answers three questions. What is the ceiling price above which most of the target audience will rule the concept out. What is the floor price below which buyers start to question the quality or legitimacy of the offer. And what is the optimal point where willingness-to-buy is highest relative to revenue per unit.
It is distinct from concept testing in general because it isolates the price variable. A standard concept test asks whether people want the concept at all. A pricing concept test assumes the concept is viable and asks what the market will pay for it. You usually run the concept test first, then the pricing test, then launch.
It is also distinct from price tracking research, which measures actual buying behavior after launch. Pricing concept testing is forward-looking, pre-launch, and based on stated intent rather than revealed preference. That is its power and its limitation. The stated-intent problem is why the methodology matters so much. A sloppy Van Westendorp study produces a clean chart with a wrong answer.
How Does the Van Westendorp Price Sensitivity Meter Work?
The Van Westendorp Price Sensitivity Meter (PSM) is a four-question methodology developed by Dutch economist Peter van Westendorp in 1976. It remains the most cited willingness-to-pay methodology in marketing research textbooks.
The four questions are:
- At what price would this be so expensive you wouldn’t consider it? (Too expensive)
- At what price would this be expensive but you’d still consider it? (Expensive)
- At what price would this be a bargain - great value for the money? (Bargain)
- At what price would this be so cheap you’d question quality? (Too cheap)
Each respondent answers all four. You then plot the cumulative distribution of each answer on a single chart with price on the X-axis and percentage of respondents on the Y-axis. The “too expensive” and “too cheap” curves rise in opposite directions. The “expensive” and “bargain” curves sit between them.
Four intersection points matter:
- Point of Marginal Cheapness (PMC): where “too cheap” crosses “bargain.” Below this price, enough people question the product that lowering further loses more buyers than it gains.
- Point of Marginal Expensiveness (PME): where “expensive” crosses “too expensive.” Above this price, the audience tips from reluctant-but-willing to ruling the product out.
- Indifference Price Point (IPP): where “expensive” crosses “bargain.” This is the median price at which as many people think it is a bargain as think it is expensive.
- Optimal Price Point (OPP): where “too cheap” crosses “too expensive.” This minimizes the percentage of people who rule out the product on price grounds, and is the closest thing Van Westendorp offers to a single recommended price.
The acceptable price range sits between PMC and PME. The launch price usually sits between IPP and OPP, with exact placement depending on positioning, margin targets, and competitive context.
The methodology is robust because the four questions triangulate rather than asking one question four ways. A respondent who picks $49 as “too expensive” and $39 as “expensive” is drawing their own line between the two, which is harder to game than a single buy-or-not question anchored on a researcher-chosen price.
When Should You Use Van Westendorp vs Gabor-Granger vs Conjoint?
Three methodologies dominate pricing concept testing. Each has a different purpose and a different cost profile.
Van Westendorp answers: what price range is acceptable for this concept? You have one concept. You need a price corridor. You do not need to model price against other features. It works on samples as small as 30-50 per segment. It is the right first move on most pre-launch pricing decisions.
Gabor-Granger answers: what is the demand curve at specific price points? You have one concept and a set of candidate prices (say $19, $29, $39, $49). You ask each respondent how likely they are to buy at each price in sequence. The output is a demand curve and a revenue-maximizing price point. It requires a bigger sample (100-200) and is most useful when you want to estimate volume at specific prices, not just the corridor.
Conjoint analysis answers: how does price trade off against other attributes? You present bundles that vary in price, features, brand, delivery speed, and so on. Respondents pick their preferred bundle across 10-20 scenarios. You model the full utility function. Conjoint is the gold standard for complex decisions, but it needs 200-400 respondents, takes 4-8 weeks, and costs $15,000-$50,000 on traditional platforms.
The decision tree is straightforward. One concept, single pricing axis, tight budget: Van Westendorp. One concept, need demand curve at candidate prices: Gabor-Granger. Multi-attribute decision with price as one variable: conjoint.
Most pre-launch pricing decisions in B2B SaaS, DTC, and subscription products belong to the first two categories. Conjoint is over-engineered for a pricing corridor. See best concept testing platforms for how each platform supports these methodologies.
Why Do Surveys Fail at Price Concept Testing?
Van Westendorp was designed in 1976 for paper surveys and has been used on digital survey platforms ever since. The methodology is fine. The delivery is where it breaks.
Three failure modes show up in every survey-based pricing study.
No context. A respondent reads a 40-word concept description and types a number into a box. There is no use case to ground the pricing. No competitive alternative to anchor against. No time to think. The number is a reflex, not a considered judgment.
No probing. The survey platform accepts $49 as an answer without asking why. If the respondent is comparing against a $100 competitor, $49 is cheap. If they are comparing against a $20 competitor, $49 is expensive. The survey has no way to surface that frame.
No quality screening. Survey panels are polluted with respondents who rush through for the incentive. Pricing questions are especially vulnerable because there is no obvious “wrong” answer, so even professional survey-takers produce clean-looking data. Nielsen Norman Group’s guidance on qualitative surveys documents how much signal is lost when follow-up probes are not possible in a self-serve survey environment. The Van Westendorp chart is statistically tidy and substantively meaningless.
A pricing corridor built on this foundation feels authoritative because the methodology is respected. It is the corridor that looks right in the deck but breaks on contact with actual customers. Product teams learn about the break six weeks after launch, when sales motion friction and win-rate data arrive, not at the point of pricing decision.
Fixing this does not mean discarding Van Westendorp. It means changing the delivery.
How Do AI-Moderated Interviews Improve Pricing Research?
AI-moderated interviews run the same four Van Westendorp questions, but inside a 30-minute conversation instead of a two-minute survey. An AI moderator asks each question, probes the answer with follow-ups, holds the respondent to specifics, and surfaces the reasoning that surveys never capture.
The structure of a pricing interview looks like this.
First five minutes: warm-up and use-case walkthrough. The moderator asks about the respondent’s current workflow, the alternatives they have considered, and the last time they made a buying decision in this category. This establishes the competitive frame.
Next fifteen minutes: concept walkthrough and the four Van Westendorp questions. Each price answer is followed by three probes. Why that number. What are you comparing against. What would move your answer up or down by 20%.
Final ten minutes: unbundling, switching cost, and objection handling. The moderator asks whether the concept should be bought as a whole or modularly, what would have to be true for the respondent to switch from their current solution, and what the single biggest concern is at the stated price.
The output is the same Van Westendorp chart a survey platform would produce, plus verbatim justification for every data point on that chart. The price corridor is defensible to an exec team because every line item can be traced to a quote from a target buyer explaining why.
User Intuition delivers these interviews at $20 per interview across a 4M+ participant panel in 50+ languages, rated 5/5 on G2, with 98% respondent satisfaction. A 30-interview study is $600 and ships in 48-72 hours. See qual at quant scale for how the economics work.
What Questions Should Pricing Concept Interviews Include?
A complete pricing concept interview has six question blocks. The four Van Westendorp questions sit in the middle. The blocks before and after them do the work of making those four questions land.
Block 1: Current-state walkthrough. What problem does the respondent currently have in this space. What are they using today to solve it. What are they paying for that solution, including hidden costs.
Block 2: Concept introduction. The AI moderator describes the concept in 100-200 words and asks the respondent to play it back in their own words. This confirms comprehension and surfaces misunderstandings that would poison the price answers.
Block 3: Use-case walkthrough. The respondent describes a specific scenario in which they would use the concept. This grounds the pricing in real context instead of abstract hypotheticals.
Block 4: The four Van Westendorp questions, each with three probes.
- Too expensive: Why? What are you comparing against? What would bring that number down?
- Expensive: Why? At what point does “expensive but worth it” become “rule it out”?
- Bargain: Why? Does “bargain” signal value or make you suspicious?
- Too cheap: Why? What does the price say about the product?
Block 5: Competitive alternative. Who else provides something like this, and what do they charge. This surfaces the reference prices that drove the four answers.
Block 6: Unbundling and willingness-to-switch. Would the respondent buy the whole concept or only parts of it. What would have to be true to switch from their current solution at the stated price.
Concept testing complete guide has the full question bank. The point of the six-block structure is that the Van Westendorp answers sit in context. A respondent who has already walked through their current workflow and named a competitor gives different price numbers than one who has not.
How Do You Set Up a Van Westendorp Study in AI-Moderated Interviews?
Setup on User Intuition takes under an hour. The workflow has five steps.
Step 1: Write the concept stimulus. 100-200 words describing the concept, the problem it solves, the core mechanism, and the target user. Include a mockup or short video if available. Avoid feature lists; focus on the value proposition as a buyer would hear it.
Step 2: Define the audience. Use the screener to filter to your target segment. For B2B, this usually means role, company size, and responsibility for buying decisions in the category. For DTC, this is demographics, category usage, and purchase frequency. The platform’s 4M+ panel covers most segments.
Step 3: Paste the discussion guide. The six-block structure from the previous section. For the four Van Westendorp questions, add the three probes explicitly so the AI moderator knows to ask them.
Step 4: Set sample size. 30-50 interviews for a single-segment pricing decision. 60-120 for a multi-segment comparison (e.g., SMB vs enterprise, US vs Europe). The methodology is stable at small samples because it is distribution-based, not mean-based.
Step 5: Launch and monitor. Interviews complete in 48-72 hours. Review the first three transcripts to verify the moderator is probing effectively, and adjust the guide if needed before the full sample fields.
The deliverable is the Van Westendorp chart with the four intersection points, a synthesized summary of why respondents chose their numbers, verbatim quotes tagged to each price band, and a segmented view if you ran the study across multiple cohorts.
How Much Does Pricing Concept Testing Cost?
Pricing concept testing ranges from $600 to over $200,000 depending on methodology and delivery channel.
Full-service pricing consultancies (Simon-Kucher, Nielsen, Kantar): $50,000-$200,000 per study. 6-10 week timelines. Includes custom sampling, bespoke analysis, and executive readout. Justified when a billion-dollar pricing decision is on the table. Over-engineered for most B2B SaaS and DTC pricing calls.
Conjoint platforms (Sawtooth, Conjointly): $10,000-$30,000 per study plus programming time and panel fees. 3-5 week timelines. The right tool for multi-attribute tradeoff analysis, not for a single-concept Van Westendorp.
Survey platforms (Qualtrics, SurveyMonkey) with Van Westendorp modules: $1,000-$5,000 for the platform plus $3-$8 per panel respondent. 2-3 week timelines. Cheap on paper but suffers from all three survey failure modes described earlier.
AI-moderated interviews (User Intuition): $20 per interview. 48-72 hour turnaround. A 30-interview Van Westendorp study costs $600 and a 100-interview segmented study costs $2,000. Includes the Van Westendorp chart, the verbatim justifications, the synthesized summary, and the raw transcripts.
The math that usually ends the debate: a $600 pricing study that prevents a $10,000 pricing error on a product that sells 1,000 units in year one returns 16x on the research spend, and that is a modest example. Most pre-launch pricing mistakes cost more than that.
For a deeper cost breakdown across all concept testing methods, see concept testing cost.
Pricing concept testing is one of the highest-leverage research investments a product team can make. The wrong price at launch is expensive to fix, slow to diagnose, and corrosive to sales motion. The Van Westendorp methodology gives you a defensible corridor in small samples. AI-moderated interviews give you the qualitative depth that makes that corridor trustworthy. The combination costs less than most teams spend on a launch event.