How to Interpret Van Westendorp Results
Running a Van Westendorp survey is the easy half. Reading what the results actually mean, and knowing what to do with them, is where most of the real value gets left on the table. This guide breaks down the four price points a completed survey produces, what a narrow versus wide acceptable range tells you about your market, why the optimal price point is a starting point and not a final answer, and what to do when your results don't come out clean.
This is a companion piece to How to Run a Van Westendorp Survey, which covers setting up a Van Westendorp survey in Zorin and reading the three summary outputs at a practical level. This post goes one level deeper into what those numbers actually mean and how to act on them correctly.
The Four Price Points a Van Westendorp Survey Produces
Each of the four original Van Westendorp questions, too cheap, a bargain, getting expensive, too expensive, produces its own cumulative response curve when plotted across all respondents. Where those curves cross defines four specific price points, each with a distinct, well-established meaning.
Point of Marginal Cheapness (PMC) and Point of Marginal Expensiveness (PME)
The Point of Marginal Cheapness is where the "too cheap" curve crosses the "expensive" curve. This is the lower bound of your acceptable price range. Price below this point and quality doubt starts to dominate, enough respondents start wondering what's wrong with a product priced this low that it works against you rather than for you.
The Point of Marginal Expensiveness is where the "too expensive" curve crosses the "cheap" (bargain) curve. This is the upper bound. Price above this point and cost resistance starts to dominate, enough respondents rule the product out on price alone that you're losing more sales than the higher price is worth.
Together, PMC and PME define your acceptable price range, the corridor where price is broadly perceived as fair by your respondents. This is the range Zorin surfaces directly from a completed survey.
Optimal Price Point (OPP) and Indifference Price Point (IPP)
The Optimal Price Point is where the "too cheap" curve crosses the "too expensive" curve. At this specific price, the share of respondents rejecting the product as too cheap and the share rejecting it as too expensive are equal, which makes it the point of lowest overall resistance across your sample.
The Indifference Price Point is where the "cheap" (bargain) curve crosses the "expensive" curve. At this price, roughly equal numbers of respondents see the product as a good deal versus starting to feel expensive. It sits inside the acceptable range and represents a kind of psychological midpoint, neither clearly a bargain nor clearly pushing into expensive territory.
These four points, PMC, PME, OPP, and IPP, are what a completed Van Westendorp analysis produces. Everything else in reading the results well comes from understanding what these numbers do and don't tell you.
| Point | Where it comes from | What it marks |
|---|---|---|
| PMC | "Too cheap" × "getting expensive" | Lower bound of the acceptable price range |
| OPP | "Too cheap" × "too expensive" | Point of lowest overall price resistance |
| IPP | "A bargain" × "getting expensive" | Psychological midpoint inside the acceptable range |
| PME | "A bargain" × "too expensive" | Upper bound of the acceptable price range |
What a Narrow vs Wide Acceptable Range Actually Tells You
The width of your acceptable range, the gap between PMC and PME, is itself a signal worth reading carefully, not just a boundary to note and move past.
A narrow range, something like $39 to $49, means your respondents largely agree on what the product should cost. That agreement is useful, but it also means the market is more price-sensitive within that band: a price move even a few dollars outside the range is likely to trigger a real, fairly sharp reaction, since there isn't much room for disagreement about value to absorb the change.
A wide range, something like $29 to $79, is less straightforward to read, and it can mean one of two fairly different things. It can genuinely reflect pricing flexibility: your product might reasonably appeal to different segments (some buyers wanting a stripped-down version, others willing to pay more for a premium tier or bundle), and a wide range means you have real room to price differently across those segments or channels without breaking anyone's sense of fairness. Alternatively, a wide range can reflect market confusion, respondents genuinely don't have a clear read on what a fair price for this product is, either because it's a new category with no established reference points or because the product concept itself wasn't clearly communicated in the survey.
Telling these two readings apart usually comes down to context you already have. If your product sits in an established category with clear competitor pricing, a wide range is more likely confusion worth investigating, possibly by refining how the product was described to respondents. If you're pricing something genuinely new, or something that legitimately serves distinct customer segments differently, a wide range is more likely real flexibility you can put to use. This is directly relevant if you're pricing the same product across multiple channels, a wide acceptable range is exactly the kind of signal that supports pricing differently on Shopify versus Amazon or Etsy without alienating buyers on any one channel, since the survey data itself suggests the market tolerates more than one price point.
The Optimal Price Point Is a Starting Point, Not a Final Price
The single most common misreading of Van Westendorp results is treating the OPP as the number to charge. It isn't, and understanding why matters more than any other part of this guide.
The OPP tells you where resistance to your price is lowest, based purely on how your respondents perceive fairness and value. It has no information about your margin requirements, your channel structure, your competitive position, or the psychological pricing conventions in your category ($X.99 vs a round number, for instance). It's a genuinely useful anchor for a pricing conversation, not a finished answer to it.
Turning an OPP into an actual list price means layering several things on top of it: whether the price still clears your cost floor and target margin, whether it needs to shift for a specific channel's fee structure, whether a promotional or launch strategy calls for starting somewhere else temporarily, and whether nudging the number to a more conventional price ending improves conversion without meaningfully changing perceived value.
This is also exactly why Zorin treats survey results and elasticity data as separate signals rather than blending them into one number. The OPP tells you what customers say feels fair, a stated-preference read, useful especially before you have any sales history to work from. Elasticity data, once a product has enough real sales history, tells you what customers actually do at a given price, a revealed-preference read. Neither one alone is the final price. The OPP, your margin, your channel context, and (once available) your elasticity data together are what a real pricing decision is built from.
See what Zorin's elasticity model says about your own catalog.
Start free trialWhen Your Curves Don't Intersect Cleanly
Sometimes a Van Westendorp analysis doesn't produce the clean, textbook crossing pattern the four-curve diagram suggests it should. This usually happens with a small sample size, or with a product respondents genuinely find hard to price, an unfamiliar category, an unclear concept description, or a product with very few comparable references in the market.
This isn't a failure of the method. It's a signal to read the result more cautiously, not to throw it out. A few things help: if curves are close but don't cross exactly, you can approximate the intersection point through interpolation rather than treating the absence of a perfect crossing as meaningless. If curves are nearly parallel rather than crossing at all, that itself is informative, it suggests the market doesn't have clear price boundaries for this specific product yet, which is a legitimate finding, not a broken survey. In either case, increasing the sample size is the most direct fix, since noisy, unstable intersections are far more common with small samples than large ones.
This is precisely why Zorin shows no confidence tier at all under 5 responses and only a low confidence tier between 5 and 19. A messy or inconclusive result at a low response count isn't a sign something went wrong with your survey setup, it's an expected consequence of not yet having enough data for the curves to settle into a clean, reliable pattern. Wait for more responses, or treat an early low-confidence read as directional rather than final.
Common Mistakes When Reading Van Westendorp Results
A short list of the misreadings that come up most often, several of which are covered in more depth above:
Treating the OPP as the final price. Covered in detail above. The OPP is an anchor for a pricing decision, not the decision itself.
Ignoring range width as a signal. A narrow range and a wide range mean genuinely different things about your market's price sensitivity and flexibility. Reading only the range's boundaries, without considering what its width implies, leaves useful information on the table.
Over-trusting a small sample. A result from 6 or 7 responses can look precise on a chart while actually being highly unstable. Treat low-response results as directional, and let the sample grow before making a significant pricing decision based on the curves alone.
Forgetting that stated preference isn't the same as actual behavior. Van Westendorp measures what respondents say they'd pay, not what they've actually paid. Research on this gap has found stated price thresholds tend to run 10-20% lower than real purchase behavior, covered in more depth in the companion post on running the survey. Treat the results as a strong starting signal, not a guaranteed final number.
Run a Van Westendorp survey on your own catalog and read the results alongside your elasticity data. Start a free trial to see both signals on the same product.
Key Takeaways
- A completed survey produces four price points: PMC and PME define the acceptable range's lower and upper bounds; OPP marks the point of lowest overall resistance; IPP marks the point where bargain and expensive perceptions balance.
- Range width is itself a signal. A narrow range signals strong consensus and higher price sensitivity; a wide range can mean genuine flexibility across segments, or it can mean the market is unclear on the product's value, and telling the two apart usually depends on context you already have.
- The OPP is a starting point, not a final price. It needs to be layered with margin requirements, channel context, and (once available) elasticity data before it becomes an actual price you'd charge.
- Curves that don't intersect cleanly aren't a failure. They're most often a small-sample or hard-to-price-product signal, fixable with a larger sample or read cautiously as directional.
- Stated preference and revealed preference aren't the same thing. Survey results tend to run lower than actual purchase behavior, so treat them as a strong starting signal rather than a guaranteed number.
Frequently Asked Questions
What do PMC, PME, OPP, and IPP stand for?
PMC is the Point of Marginal Cheapness, the lower bound of the acceptable price range. PME is the Point of Marginal Expensiveness, the upper bound. OPP is the Optimal Price Point, where resistance to being too cheap and too expensive are equal. IPP is the Indifference Price Point, where roughly equal numbers of respondents see the price as a bargain versus getting expensive.
Is the optimal price point (OPP) the price I should actually charge?
Not directly. The OPP tells you where price resistance is lowest based purely on how respondents perceive fairness, but it doesn't account for your margin requirements, channel-specific costs, competitive positioning, or common pricing conventions in your category. Use it as a well-informed starting point, then adjust for those factors to arrive at an actual price.
What does a narrow acceptable price range mean?
A narrow range, for example $39 to $49, means your respondents largely agree on what the product should cost. That consensus also means the market is more price-sensitive within that band, moving even slightly outside the range is likely to trigger a noticeably sharper reaction than it would in a market with a wider range.
What does a wide acceptable price range mean?
A wide range, for example $29 to $79, can mean one of two things: genuine pricing flexibility, where different customer segments have different willingness to pay and you have room to price differently across them or across channels, or market confusion, where respondents don't have a clear sense of what the product should cost. Whether it's flexibility or confusion usually depends on whether your product sits in an established category with clear reference prices or a newer, less familiar one.
What should I do if my Van Westendorp curves don't intersect cleanly?
This usually happens with a small sample size or a product respondents find genuinely hard to price. You can approximate the intersection points through interpolation if curves are close but don't cross exactly, or treat nearly parallel curves as a signal that the market doesn't yet have clear price boundaries for this product. In either case, growing your sample size is the most reliable fix, and a low-confidence result should be treated as directional rather than final.
Why does Zorin keep Van Westendorp results and elasticity data separate instead of combining them?
Because they measure different things. The survey measures stated preference, what customers say they'd pay, which is useful even for a brand-new product with no sales history. Elasticity measures revealed preference, what customers actually do when a real price is in front of them, which requires real sales data to calculate. Keeping them separate lets you see when the two signals agree, which builds confidence, or disagree, which is worth investigating rather than quietly averaging away.
Do I need a large sample size to trust my Van Westendorp results?
Larger samples produce cleaner, more stable curve intersections, but you don't need hundreds of responses to get directional value. In Zorin, 20 or more responses is labeled good confidence, enough for the acceptable range and price points to be reasonably reliable for a single-product decision. Below that, treat results as an early, directional signal rather than a number to commit a final price to.
The four price points a Van Westendorp survey produces tell you how your market perceives fairness at different prices, which is genuinely useful information you don't have before running one. Turning that into an actual price still takes margin data, channel context, and, once real sales history exists, elasticity data layered on top. Zorin keeps the survey results and the elasticity model side by side on the same platform, so both pieces of the decision are in front of you when you're ready to set the price.
Written by Dexter
Dexter is part of the team at Zorin, building tools that help ecommerce merchants price with data instead of guesswork.