Van Westendorp Price Sensitivity Meter for New Products

The Van Westendorp Price Sensitivity Meter helps set prices for new products when there is no competitive price anchor. Buyers supply the price thresholds themselves. Those thresholds define an acceptable price range. A defensible decision still needs a clear product description, the right respondents, and testing within that range.

Why new products create a pricing problem that standard methods cannot solve

A product team needs a launch price before committing to distribution, packaging, and positioning. Yet a genuinely new product may have no market price to use as a starting point. That leaves a gap in the research design. Gabor-Granger requires a researcher to propose a price ladder for respondents to assess. Conjoint analysis also needs predefined price levels, although it treats price as one attribute among several.

Without a competitive anchor, those prices come from what the researcher assumes buyers will pay. Respondents then judge the numbers they were shown, which can steer the findings toward the starting guesses. The launch makes that mistake expensive to undo. Teams need a method that lets the market define the range.

What the Van Westendorp Price Sensitivity Meter is and how its four questions work

Dutch economist Peter van Westendorp created the Price Sensitivity Meter in 1976. The PSM remains one of the fastest, most widely used pricing research methods because it gathers four open-ended price thresholds without proposing a price. Buyers give the numbers they expect to pay. The questions come in a fixed order, though their wording can change a little as long as each one still measures the same thing.

The first question captures the quality floor by asking, “At what price would you consider this product so cheap that you would question its quality? The second asks, “At what price would you consider this product a bargain — a great buy for the money?” This captures the value sweet spot, where the offer feels attractive without seeming suspect. Cheap is not always a selling point.

The third asks, “At what price would this product be getting expensive, but you would still consider buying it? It marks hesitation. The fourth asks, “At what price would this product be so expensive that you would not consider buying it?” That captures the hard upper limit of consideration. Respondents answer freely, with no predefined list or multiple choice, and keeping the sequence fixed avoids the anchoring effects that changing it can introduce.

How the four questions become four curves and four actionable price points

Each answer set becomes a cumulative frequency curve, with price on the horizontal axis and percentage of respondents on the vertical axis. The “too cheap” and “bargain” curves start high and fall as price rises. The “getting expensive” and “too expensive” curves do the reverse. Their crossings supply the reference points for interpretation.

Take a made-up example of a home appliance that respondents understand. At low prices, the “too cheap” curve would be high while the “too expensive” curve would be low. As the plotted price rises, quality objections would decline and objections to expense would grow. Where those curves cross is the Optimal Price Point, or OPP. Equal shares object at each extreme, so this point shows where resistance is balanced, though it cannot prove which price brings in the most revenue.

A different crossing answers a different question. Where “bargain” meets “getting expensive” is the Indifference Price Point, or IPP, where equal shares perceive a deal and a premium. Where “bargain” meets “too expensive” is the Point of Marginal Expensiveness, or PME, the upper bound beyond which resistance overtakes perceived value. The lower boundary is the PMC. Together, PMC and PME define the Range of Acceptable Prices, or RAP, while OPP and IPP provide reference points for messaging and further testing.

A narrow RAP signals strong agreement about what the product should cost. A wide range can reflect uncertainty or different levels of willingness to pay. Neither shape is a failure in itself. Prices within the range are viable candidates on perceived acceptability. Positioning and revenue tests still determine where to land.

How to run a PSM study that produces a reliable range

To put the method into practice, first define exactly what respondents are pricing. Specify format, quantity, channel, and key features. “A project management tool” leaves too much unstated. “A project management tool for teams of 5–25 with task tracking, calendar view, and file sharing” gives respondents a firmer object to assess. The description must give people a realistic idea of the product because the PSM itself cannot check whether what they picture matches the offer.

Next, recruit people from the actual purchase population. Screen for category involvement, purchase recency, or decision-making authority, rather than assuming a general-population sample will do. People who would never buy can compress the range. When testing several price or pack configurations, use monadic testing, with one configuration per participant and the sample split across cells.

Then field the questions in their fixed order. Keep competitor prices and promotional price framing out of the surrounding copy so they do not supply an anchor before respondents answer. Check the currency presentation, too. An unfamiliar format or a denomination with cultural weight can reset expectations before the first response.

Before plotting, remove outliers that are orders of magnitude above or below the realistic range. Flag answers that conflict during post-processing, since blocking them during the survey can slow people down and introduce bias. Then plot all four curves on one chart. Use price on the horizontal axis and cumulative percentage on the vertical axis, and read the intersections visually to identify the range and reference points.

Where the PSM output is reliable and where it breaks down for new products

The PSM records stated price perceptions, provided respondents understand what they are pricing. Established categories with clear comparables make that assumption easier to defend. Genuinely new products make it harder. The absence of a competitive anchor creates the need for PSM, but unfamiliarity also increases the chance that people imagine different offers. Open-ended answers leave out the researcher’s price ladder, but buyers still have their own prices in mind.

Premium personal care illustrates the distinction. If a respondent mentally prices a mass-market alternative instead of the intended premium offer, the answer describes the wrong product. The same thing can happen with a home appliance if its description leaves people unsure of what is on offer. These are interpretation problems. Clean curves cannot repair a misunderstood stimulus.

Even with shared understanding, perceptions are not purchases. Stated answers can diverge from behavior in either direction, and PSM alone cannot resolve that gap. Its named intersections describe crossings of price-perception curves. Procurement-driven B2B pricing needs particular caution because tenders, incumbent bias, and contract terms may outweigh perceptions. The practical correction is to establish what participants currently buy and pay before asking about expense, while keeping researcher-supplied price anchors out of the stimulus.

What to add to the PSM before committing to a launch price

To address those limits, use PSM as the first layer of research. The Newton-Miller-Smith purchase intent extension, developed in 1993, adds purchase likelihood questions at one or more prices within the range. This produces a rudimentary demand curve and revenue estimate. However, asking about likelihood is still not the same as observing a purchase.

Gabor-Granger becomes more useful once PSM has supplied a defensible range. Instead of testing an arbitrary ladder, it can test specific prices within that range to measure demand elasticity and estimate revenue curves. Sequence changes the quality of the question. Conjoint analysis helps when configurations or tiers call for feature-price trade-offs, showing how much the features raise the upper price limit beyond what buyers think of the price alone.

A/B price testing then shows what buyers do. In the worked example, a higher price reduced conversion but increased average revenue per user enough to lift gross profit. The team launched at a lower price with an annual plan and a premium upsell tier for high-engagement cohorts, all within the PSM corridor. The decision concerned the offer structure.

Purchase intent adds a stated likelihood of buying, Gabor-Granger tests demand at selected prices, conjoint examines trade-offs, and A/B testing observes behavior. The range comes before the heavier optimization work. That order prevents later research from inheriting an unsupported price assumption.

How AI-moderated interviews and synthetic panels change what the PSM can surface

Static PSM surveys show where buyers start to resist a price but reveal little of why they resist it. AI-moderated platforms can embed Van Westendorp as a native question type in a conversational interview and explore the reasoning behind each threshold in real time. The numbers gain an explanation in the same instrument. This addresses a weakness of the survey without replacing its four-question mechanism.

Synthetic consumer panels add a preflight stage before human fieldwork. A synthetic preflight can test the concept and stimulus, suggest where the range is likely to start and end, and flag when people may base their answers on the wrong comparison product. Those outputs are directional bounds. Their role is to narrow a broad option set quickly and cheaply before a team commits to a full study with human respondents.

Checks with real people are a required second phase. AI-moderated interviews with real respondents from the purchase population can test the provisional range and uncover the reasoning behind their thresholds. This is especially useful when an innovative product does not fit an existing category. The conversation can expose what participants believed they were pricing, rather than leaving the team to infer it from the chart.

A verified, representative panel is as important as the moderation method because sophisticated interviews cannot turn non-buyers into the target market. Synthetic preflight narrows the field, and human conversations validate the surviving options and explain their appeal or resistance. Neither phase substitutes for the other.

Turning a PSM range into a defensible launch price decision

To make the launch decision, connect the acceptable range to a positioning strategy. Penetration pricing sits near the PMC to get more people to buy, with a plan to raise the price once buyers have formed a view of the quality. Value pricing sits near the IPP and is the most defensible position when there is no clear premium differentiator. Premium pricing sits near the PME to bring in more revenue per customer while accepting fewer sales. That premium choice depends on a stimulus that distinguishes the offer from the mass-market reference respondents might otherwise use.

The OPP should not become a default simply because its name sounds decisive. Choosing it without testing can leave revenue unused or underinvest in premium positioning. A launch also needs more than a headline price. Tiers, annual plans, and upsell paths can occupy different parts of the acceptable range, rather than forcing one price to serve every buyer and revenue objective.

Before committing, confirm that participants understood and priced the intended product, checking for references to cheaper or different products. If the stimulus was ambiguous, revise it before relying on the resulting range. After launch, plan a repeat study once real purchase experience can inform perceptions, because the market’s understanding of a new product changes as people learn what it does.