Common mistake 1 with Price sensitivity (Van Westendorp)

Treating the optimal price point as a recommendation. There is no quantity in this method, so no revenue and no margin - it cannot tell you what to charge.

Common mistake 2 with Price sensitivity (Van Westendorp)

Reading the acceptable range as demand. Nobody was asked to buy anything.

Common mistake 3 with Price sensitivity (Van Westendorp)

Keeping the inconsistent respondents to preserve sample size. The option is there and it is off by default, because their answers cannot be interpreted in either direction.

Common mistake 4 with Price sensitivity (Van Westendorp)

Using it where a competitor set exists. Conjoint measures trade-offs against real alternatives and is the better tool then - this app has one.

EconStatsCraft › Price sensitivity (Van Westendorp)

Price sensitivity (Van Westendorp)

The four price questions, turned into four cumulative curves whose intersections give the indifference price, the optimal price point and the range of acceptable prices. Measures perceived value - not demand.

What it needs

So cheap you would doubt it
A bargain
Getting expensive
Too expensive to consider

The intuition

Van Westendorp asks four questions about price - at what price is it too cheap to trust, cheap, expensive, and too expensive to consider - and plots the cumulative answers. Where the curves cross gives a range of defensible prices. It is cheap, fast and widely used, and it measures stated attitudes to a price rather than willingness to pay, which is a real limitation and not a fixable one.

When to use it

Early pricing work on something new, where there is no sales history and no competitor to anchor on. It is quick, it needs no design, and it gives a range to start from.

Where the estimator comes from

Four cumulative curves are built from the four questions: the shares finding the product too cheap and cheap are plotted DECREASING in price, and expensive and too expensive INCREASING. Four intersections follow. The point of marginal cheapness is where "too cheap" meets "expensive"; the point of marginal expensiveness is where "cheap" meets "too expensive"; between them lies the range of acceptable prices. The OPTIMAL PRICE POINT is where "too cheap" meets "too expensive" - the price minimising the share who reject it outright in either direction - and the INDIFFERENCE PRICE is where "cheap" meets "expensive", often read as the price the median respondent considers normal.

The model

\text{OPP}:\ F_{\text{too cheap}}(p)=1-F_{\text{too expensive}}(p),\qquad \text{IDP}:\ F_{\text{cheap}}(p)=1-F_{\text{expensive}}(p)

What it assumes

That a respondent’s four prices are in order: too cheap < a bargain < getting expensive < too expensive. Anyone who answers otherwise has misread the question, and the module counts and excludes them - 600 respondents in the sample here, 64 of them inconsistent.
That people can report what they would pay. Stated price acceptability is systematically more generous than behaviour, which is why the acceptable range always looks wide.
That the product is understood from a description alone.
Respondents can price a product they have not seen in a market they do not face. The four questions ask for judgements people are poor at making, and the method inherits every optimism in the answers.
The product is described identically to everyone. Van Westendorp is highly sensitive to the description, so two framings produce two acceptable ranges and neither is the truth.

Statistical properties

It measures price PERCEPTION, not demand. Nobody is asked whether they would buy, so no quantity is implied and no revenue can be computed from it.
There is no theory behind the intersections. They are conventions that have proved useful, not equilibria, and the method has never had a formal justification.
Respondents must understand the product. Asked about something unfamiliar, they anchor on the first number they think of, and the four answers become internally consistent but meaningless.
A respondent whose four answers are not monotonically ordered has misunderstood the question. Those cases should be screened out and the share reported.
The optimal price point is often near the middle of the acceptable range, which makes the method feel more decisive than it is. The RANGE is the finding; the single point is a convention.
Gabor-Granger and choice-based conjoint both ask about purchase at a price and can produce a demand curve. Van Westendorp cannot, and should be treated as a first pass.

How to read the output

The indifference price is where as many people call it cheap as call it expensive - often the price of the market leader, or the price people are used to.
The optimal price point is where "too cheap" and "too expensive" balance: the fewest people put off on either count. Optimal for minimising objections, which is not the same as maximising profit.
An optimal point BELOW the indifference price suggests a category seen as expensive for what it delivers; above it suggests price is read as a quality signal.
The output is a range and two crossing points, not a price. It bounds where to experiment; the experiment is what establishes a price.

A worked reading

536 usable respondents of 600 give a point of marginal cheapness of 660.08 and a point of marginal expensiveness of 940.44, so the range of acceptable prices spans 660 to 940. The optimal price point falls at 742.88 and the indifference price at 830.62. Look at the shape of that rather than the digits: the acceptable range is 280 wide on a product priced around 750, so it is more than a third of the price itself, and the two headline numbers sit 88 apart inside it. Quoting 742.88 as "the optimal price" to two decimals implies a precision the method cannot support - these are intersections of four hand-drawn cumulative curves, and moving any one of them by a few respondents moves the crossing point by more than a currency unit. The honest output is the band, with the optimal point marked inside it.

Symptom → diagnosis → what to do

If you see this	It means	Do this
Curves not crossing	Inconsistent answers or too few respondents	Screen for monotonicity first
A very wide acceptable range	Little consensus, or a heterogeneous sample	Segment before pricing
Optimal price above the point of marginal expensiveness	Impossible ordering	A data or plotting error
Many non-monotonic respondents	The question was not understood	Report the screen-out rate
Range far from the current market price	Unfamiliarity or a hypothetical bias	Validate against actual prices
Being asked for a revenue forecast	The method cannot produce one	Use Gabor-Granger or conjoint

Common mistakes

Treating the optimal price point as a recommendation. There is no quantity in this method, so no revenue and no margin - it cannot tell you what to charge.
Reading the acceptable range as demand. Nobody was asked to buy anything.
Keeping the inconsistent respondents to preserve sample size. The option is there and it is off by default, because their answers cannot be interpreted in either direction.
Using it where a competitor set exists. Conjoint measures trade-offs against real alternatives and is the better tool then - this app has one.

Where to read this properly

Van Westendorp (1976), "NSS Price Sensitivity Meter", ESOMAR Congress.
Lipovetsky, Magnan and Zanetti-Polzi (2011), "Pricing Models in Marketing Research", Intelligent Information Management 3(5).
Nagle and Muller, The Strategy and Tactics of Pricing (2018), ch. 12.

Related Marketing research

Conjoint analysis · MaxDiff (best–worst scaling) · TURF - reach of a product line · Market basket analysis · Perceptual map · Scale reliability (Cronbach’s alpha) · Key driver analysis

All analyses

Data & variables

Each example is simulated from a known model, so you can check whether the estimator recovers the truth - and nothing here is anyone's private data.

Filter variables

Drag a variable into a slot on the right - or just click it.

Choose an example dataset above to begin.

Choose a transformation - it opens in the data editor, with a live preview.

Apply to

Multiply by

Columns ctrl-click for several

Group by ctrl-click for several

Within groups

Statistic

Bands

How many

Cut points

Keep rows where

Value

and

Fill with

Value

Also mark which rows were filled

Rules

Base period = 100

Which part

Percent at each tail

Threshold

Survey weight optional

Direction

Names column

Values column

Largest first

Delete them, not just count them

Read this column as

Tidy the text

New name optional

EconStatsCraft rigorous, plain-spoken econometrics

Econometrics that explains itself.

Regression, panel data, time series and the whole battery of tests. Every result is followed by a verdict you can defend.

Load a dataset to see it here.

Price sensitivity (Van Westendorp)

What it needs

The intuition

When to use it

Where the estimator comes from

The model

What it assumes

Statistical properties

How to read the output

A worked reading

Symptom → diagnosis → what to do

Common mistakes

Where to read this properly

See also

Related Marketing research

All analyses

Charts

Maps

Explore

Regression

Time series - one series

Time series - several series

Panel data

Multivariate statistics

Marketing research

Causal

Finance & risk

Actuarial

Tests & breaks

Machine learning

Econometrics that explains itself.