Your survey responses
The four questions: at what price would this be so cheap you doubt its quality; a bargain; getting expensive but still considered; too expensive to consider. Commas, tabs, semicolons or spaces separate the numbers. Currency symbols are ignored. No thousands separators. A header row is skipped. Rows whose prices do not rise in order are dropped and counted.
Paste responses, then click Analyze.
Result
| Price point | Estimate | 5th percentile | 95th percentile |
|---|
How it is calculated
For every price that appears in your data the tool counts, as a share of respondents, how many called that price too cheap or cheap (at or above it) and how many called it expensive or too expensive (at or below it). The curves cross where two groups are equal: marginal cheapness is too cheap against not cheap, marginal expensiveness is too expensive against not expensive, the optimal price point is too cheap against too expensive, and the indifference price is cheap against expensive. Crossings are linearly interpolated between the two nearest survey prices, and the first crossing from the left is used. Bands come from 300 resamples of your respondents with replacement, with a fixed random seed.
Not modelled: purchase intent, demand, revenue, competitor prices, and segments. Unverified: checked by hand calculation and unit tests only. Stated prices in a survey often differ from what people pay, so test any price before committing.
Common questions
What is the Van Westendorp price sensitivity meter?
A survey method from 1976. You ask each respondent four questions: at what price is this so cheap you would doubt its quality, a bargain, getting expensive, and too expensive to consider. Plotting the cumulative answers gives four curves, and where they cross marks a range of acceptable prices.
What are the four price points?
The point of marginal cheapness (where too-cheap answers overtake bargain answers), the point of marginal expensiveness (where too-expensive answers overtake not-expensive answers), the optimal price point (where as many people call it too cheap as too expensive), and the indifference price (where as many call it a bargain as expensive). The range between the first two is the range of acceptable prices.
How many survey responses do I need?
Aim for at least 100 in the segment you sell to. This tool runs from 5 coherent responses and warns below 30, because the curves get noisy. It also shows a bootstrap band around each price point so you can see how much a small sample moves them.
What input format does it need?
Four prices per respondent, one respondent per line, in the order too cheap, cheap, expensive, too expensive. Commas, tabs, semicolons or spaces separate them, currency symbols are ignored, an optional header row is skipped, and thousands separators are not supported. Rows where the prices do not rise in order are dropped and counted.
Does it tell me what price maximises revenue?
No. It measures stated price perception, not willingness to pay or how many people would buy. Use the acceptable range to bound a test price, then run a real price test.
What is the bootstrap band?
The tool resamples your respondents with replacement 300 times, recomputes each price point every time, and reports the 5th to 95th percentile. A wide band means the point moves a lot with who happened to answer. The resampling uses a fixed seed, so the same data gives the same bands.
Is my survey data uploaded anywhere?
No. Everything you paste stays in this browser tab. There are no accounts, cookies or analytics vendors.