Back
Pricing Survey Questions: Van Westendorp and More

Matthieu Saussaye

Pricing survey questions ask target customers how they perceive and react to a price, so you can find an acceptable price range, a likely demand curve and the packaging that fits. The main methods are Van Westendorp's four price questions, Gabor-Granger purchase-likelihood questions, conjoint or choice-based trade-offs, and simple value-for-money and willingness-to-pay questions.
The short version:
Van Westendorp finds an acceptable price range from four questions. Good early, when you have no price anchor.
Gabor-Granger asks purchase likelihood at specific prices. Good when you have a shortlist of price points.
Conjoint and choice tasks show how price trades off against features. The most realistic, and the most work.
Direct willingness-to-pay questions ("What is the most you would pay?") are easy and unreliable on their own.
Always describe the product precisely first. Respondents price what they imagine; make sure it is what you will sell.
Survey the people who actually buy, screened by budget responsibility, not a general audience.
Pricing survey methods compared
Method | Questions asked | What you get | Best for | Main limitation |
|---|---|---|---|---|
Van Westendorp PSM | Four open price questions: too cheap, cheap, expensive, too expensive | Acceptable price range and indifference point | New products with no price reference | Ignores competitors; no demand estimate unless extended |
Gabor-Granger | Purchase likelihood at a sequence of prices | Demand curve and revenue-maximizing price | Choosing between known price points | Respondents see price in isolation, without alternatives |
Conjoint / choice-based | Choices between product profiles with different features and prices | Value of each feature, price sensitivity, simulated market share | Packaging, tiers, feature-price trade-offs | Needs larger samples and careful design |
MaxDiff on features | Most and least important feature from sets | Ranked feature importance | Deciding what goes in which tier | Does not measure price directly |
Direct WTP | What is the most you would pay? | Rough upper bound | Quick exploration, qualitative context | Stated answers tend to differ from real purchase behavior |
Value perception | How would you rate the value for money? | Satisfaction with current price | Existing customers, price changes | Does not tell you the right price |
25 pricing survey questions, ready to use
Screening and context
Who is responsible for purchasing [category] in your household or organization? (Me / Shared / Someone else)
How much do you currently spend on [category] per month? (bands)
Which products or services do you currently use for [need]? (multi-select)
How satisfied are you with what you pay today for [category]? (5-point)
Van Westendorp Price Sensitivity Meter
Show the product description first, then ask the four questions in this order. The wording below follows the common form described in the Van Westendorp reference:
At what price would you consider [product] to be so expensive that you would not consider buying it? (Too expensive)
At what price would you consider [product] to be priced so low that you would feel the quality could not be very good? (Too cheap)
At what price would you consider [product] to be starting to get expensive, so that it is not out of the question, but you would have to give some thought to buying it? (Expensive)
At what price would you consider [product] to be a bargain, a great buy for the money? (Cheap / good value)
To estimate demand as well as a range, add the Newton-Miller-Smith extension: ask purchase likelihood at the respondent's own "expensive" and "cheap" prices.
At [your expensive price], how likely would you be to buy [product] in the next six months? (1 Very unlikely to 5 Very likely)
At [your cheap price], how likely would you be to buy [product] in the next six months? (1 to 5)
Gabor-Granger
If [product] cost [price A], how likely would you be to buy it? (Definitely would / Probably would / Might or might not / Probably would not / Definitely would not)
And if it cost [price B]? (asked higher or lower depending on the previous answer)
And at [price C]?
Package and model preference
Which of these plans would you choose? (show 3 tiers with features and prices, plus None)
How would you prefer to pay for [product]? (Monthly / Annually / One-off / Pay per use)
Which of these features would you be willing to pay extra for? (multi-select)
Which feature matters most and which matters least to you in this set? (MaxDiff, repeat with rotating sets)
Would you prefer [Option A: lower price, fewer features] or [Option B: higher price, more features]? (choice task)
Value and price change
How would you rate the value for money of [product]? (Very poor to Excellent)
Compared with alternatives, is the price of [product]... (Much lower / Lower / About the same / Higher / Much higher)
If the price increased by [amount], what would you most likely do? (Keep using it / Switch to a cheaper plan / Look for an alternative / Cancel)
What would make [product] worth a higher price to you? (open)
What is the main reason you think the price is fair or unfair? (open)
What is the most you would pay for [product] per [period]? (open number, use as context only)
How to analyze Van Westendorp results
Plot the cumulative share of respondents for each of the four questions against price. Conventional practice inverts the "too cheap" and "cheap" curves so that the lines cross. According to the Van Westendorp reference, the method was introduced in 1976 by Dutch economist Peter van Westendorp, and the intersections are commonly read as follows:
Intersection | Name | Interpretation |
|---|---|---|
Too cheap x Expensive | Point of marginal cheapness (PMC) | Lower bound of the acceptable range |
Too expensive x Cheap | Point of marginal expensiveness (PME) | Upper bound of the acceptable range |
Expensive x Cheap | Indifference price point (IPP) | Equal shares call the price cheap and expensive |
Too cheap x Too expensive | Optimal price point (OPP) | Equal shares say the price exceeds their lower or upper limit |
Read the range, not a single number. The method does not account for competitor prices on its own and does not estimate volume without the purchase-likelihood extension, so pair it with Gabor-Granger or a choice task before you set a final price.
Reading Gabor-Granger results
For each price, count the share of respondents who said they definitely or probably would buy. Multiply each share by its price to get a revenue index, and look for the price where that index peaks. Treat the curve as relative: it shows how demand falls as price rises among your respondents, not the exact volume you will sell.
How to run a pricing survey that you can trust
Describe the product exactly
Respondents price what they picture. Show a concrete description, with what is included, what is not, and the billing unit (per month, per user, per pack). A vague concept produces a vague price.
Survey real buyers
A general population sample will price a B2B tool or a premium product unrealistically. Screen for people who buy the category and control the budget. Use quota sampling to keep segments such as company size or income in proportion, and check your base per segment with the sample size calculator; pricing results are often reported by segment, which divides your sample.
Watch for hypothetical bias
People answer price questions without spending money, and stated willingness to pay tends to differ from actual behavior. Use relative methods (comparing prices, choosing between packages) rather than trusting any single stated amount, and validate with real-world tests such as a pricing page experiment when you can.
Control the order and anchors
Do not show your intended price before the Van Westendorp questions; it anchors every answer.
In Gabor-Granger, randomize the starting price or use a structured ladder, so the first price does not set the frame.
Ask value and open "why" questions after the price questions, not before.
These are forms of the anchoring and order effects covered in our survey bias guide.
Ask why
A price range tells you where; it does not tell you what drives perceived value. An open follow-up such as "What would make it worth more to you?" often reveals the feature or proof point that moves the acceptable range upward.
Which pricing method to use when
Your situation | Start with | Then |
|---|---|---|
New product, no price reference | Van Westendorp | Gabor-Granger on 3-5 prices inside the range |
Choosing between 3 price points | Gabor-Granger | Open "why" question on value |
Designing tiers or bundles | MaxDiff on features | Choice-based conjoint with price |
Planning a price increase | Value-for-money and reaction questions with customers | Segment by usage and plan |
Very early idea | Qualitative interviews about current spend and alternatives | Van Westendorp once the concept is defined |
Pricing questions also fit inside a broader concept test, where you measure appeal and price perception of the same concept.
Run pricing research with real explanations using SmartInterview
SmartInterview runs pricing surveys by voice or text, with quotas to keep buyer segments balanced. After a respondent gives a price or rates value for money, the AI asks a follow-up, in their language, about what drives that judgment. Open answers are coded into themes automatically, so you see the reasons behind price sensitivity alongside your price curves.
Start free and build your first pricing survey.
Frequently Asked Questions
What questions should be asked in a pricing survey?
Ask screening questions to confirm the respondent buys the category, then the four Van Westendorp questions (too expensive, too cheap, expensive, bargain) or purchase likelihood at set prices, a package-preference question, and an open question on what would make the product worth more.
What is the Van Westendorp pricing model?
The Van Westendorp Price Sensitivity Meter is a survey method introduced in 1976 by Dutch economist Peter van Westendorp. Respondents give four prices: too expensive, too cheap, starting to get expensive, and a bargain. The intersections of the cumulative curves define an acceptable price range.
What is the Gabor-Granger method?
Gabor-Granger asks respondents how likely they are to buy a product at a series of prices, moving up or down based on each answer. The results form a demand curve, which you multiply by price to find the price that maximizes revenue among respondents.
How do you ask about willingness to pay in a survey?
Prefer indirect methods: Van Westendorp ranges, purchase likelihood at specific prices, or choices between packages at different prices. A direct "What is the most you would pay?" question is easy to ask but stated amounts often differ from real purchase behavior, so use it only as context.
How many respondents do you need for a pricing survey?
It depends on the method and on how many segments you will report. Van Westendorp and Gabor-Granger work with a few hundred qualified respondents in total; conjoint studies usually need more. Plan for enough respondents in each segment you will analyze separately.
What is the difference between Van Westendorp and conjoint analysis?
Van Westendorp asks about price alone and returns an acceptable price range. Conjoint shows respondents complete product profiles that vary in features and price, then estimates how much each feature is worth and how price affects choice. Conjoint is more realistic but needs more design work and larger samples.


