Questions this report answers
- Which price range does the target group accept, and where does "too cheap" start to hurt credibility?
- At which price point does expected revenue peak?
- How much extra will buyers pay for a given feature, pack size or service level?
- How price-sensitive are different segments, channels and countries?
- How far can we raise the price before losing a meaningful share of buyers?
- Which tier or bundle structure moves most customers to the premium option?
When you need it
The typical trigger is a decision that is hard to reverse: a launch with no obvious reference price, a price increase after input costs have risen, a move from one-off sales to a tiered subscription, or an existing range entering a new country.
That last case is where copying a home-market price is most dangerous. Eurostat's 2024 price level indices (EU-27 = 100) put restaurants and hotels at 53.2 in Bulgaria and 124.6 in the Netherlands, while food and non-alcoholic beverages sit far closer, at 88.8 and 98.9. Tolerance varies by category as well as country, so measure it locally. Sentiment shifts too: Eurostat's consumer confidence indicator for the EU-27 was -17.3 in October 2026, against -12.7 a year earlier, which is a good reason to re-test thresholds measured in a more optimistic year.
Pricing research is the wrong tool when the price is set by regulation or a tender, when a handful of B2B buyers negotiate individually, or when you have enough traffic to run live price experiments. If the concept itself is unproven, test the idea first and the price second.
Sample table of contents
- Executive summary
recommended price range and key trade-offs on one page - Objectives and hypotheses
the pricing decisions the study informs - Methodology and sample
methods, quotas, fieldwork dates and cleaning rules - Competitive price context
shelf and online price points against positioning - Acceptable price range
Van Westendorp curves with optimal and indifference points - Demand and revenue curves
Gabor-Granger purchase likelihood at each price - Value of features
conjoint utilities and willingness to pay per attribute - Scenario simulations
share of preference under competitor moves - Segment and country differences
where sensitivity is highest - Recommendations and risks
price architecture, roll-out and post-launch monitoring
Methodology and sources
Choosing the method
The Van Westendorp Price Sensitivity Meter asks four questions, from "so cheap you would doubt the quality" to "so expensive you would not consider it"; the points where the curves intersect mark an acceptable range. It suits early stages but shows no competitors and does not predict volume. Gabor-Granger asks about purchase likelihood at a sequence of prices and returns a demand curve and a revenue-maximising point for one defined product.
Conjoint analysis has respondents choose between realistic offers that vary in price, features and brand, with a "none of these" option. The model yields utilities per attribute level, willingness to pay per feature and a scenario simulator. Price elasticity can be read from those simulations or modelled from actual sales history, ideally both.
Sample and quality control
Fieldwork is usually an online survey of screened category buyers with quotas by age, region and channel. Conjoint needs larger samples, as every segment you want to read needs a robust base, so we size it with you before fieldwork. We remove speeders and straightliners, check that Van Westendorp answers run in logical order and keep holdout tasks to validate the conjoint.
What you receive
- Report in PDF with the recommended price range and the evidence behind it
- Interactive Excel simulator for testing price and feature scenarios
- Van Westendorp and Gabor-Granger charts by segment and country
- Cross-tabulations in Excel for every question
- Anonymised respondent-level data in CSV or SPSS format
- Results workshop with your pricing, sales and finance teams
How the project runs
- Briefing and hypotheses — 2–4 days to agree decisions, products and markets
- Price context and design — about a week to review competitor prices, or an existing competitor analysis, and set attributes, levels and the price ladder
- Questionnaire, scripting and pilot — 3–5 days, including a soft launch
- Fieldwork — 1–2 weeks, longer for low-incidence buyers
- Analysis and simulator build — 1–2 weeks
- Report and workshop — 3–5 days
Worked example
Example: a hypothetical online speciality-coffee retailer selling in Germany, Poland and Bulgaria wants to launch a subscription. Management needs to know which delivery frequencies and bag sizes to offer, whether free shipping justifies a higher monthly price and whether one price list can work in all three markets.
A choice-based conjoint with regular coffee buyers in each country tests subscription offers against buying single bags online and against a supermarket alternative. The simulator shows that free shipping is valued more highly than a larger bag in every market, that an extra grinding option adds almost nothing to willingness to pay, and that Bulgarian buyers react much more sharply to the highest price level than German ones.
The retailer drops the planned third tier, keeps the same two-tier structure everywhere, sets country-specific price points and schedules a follow-up once three months of real cancellation data are available.
Common mistakes and how to avoid them
- Asking "How much would you pay?" outright. Answers anchor on familiar prices; structured methods produce usable curves.
- Testing price without context. A bare price invites abstract answers, so conjoint tasks need realistic alternatives.
- Treating stated intent as a sales forecast. Surveys overstate buying, so read the shape of the curve and calibrate against sales data.
- Setting a poor price ladder. Levels should span the plausible range around your own and competitor prices, not the extremes.
- Reporting one average. It can hide a price-sensitive group next to a premium one, so read results by segment.
Who it suits
| Suitable for | Why |
|---|---|
| FMCG brands planning a price increase | Shows where buyers start trading down to own label |
| Subscription and software businesses | Conjoint values tiers, usage limits and add-ons side by side |
| Hotel and restaurant groups designing packages | Tests whether extras such as breakfast justify a higher rate |
| Pharmacy and health brands entering a new country | Price tolerance is measured locally, not copied |
| E-commerce retailers launching own-brand ranges | Positions the new range against marketplace prices |
Frequently asked questions
How long does a pricing study take?
Most projects take four to seven weeks from briefing to workshop. A Van Westendorp or Gabor-Granger study in a single market sits at the shorter end; a multi-country conjoint with a simulator needs longer, because design, translation and the pilot all come before fieldwork.
What affects the cost of a pricing research project?
Mainly the method, since conjoint needs design work and larger samples than a price sensitivity meter. The number of countries and languages, how hard the buyers are to reach, how many segments you need and whether you want a simulator also count. Send a short brief and we will prepare a tailored quote.
Should I use Van Westendorp or Gabor-Granger?
Van Westendorp suits early stages, when you need a sensible range for an unfamiliar category. Gabor-Granger suits a defined product with a shortlist of candidate prices and a revenue question. Many studies use both, and our price sensitivity calculator lets you explore Gabor-Granger data yourself.
Can you test the price of a product that does not exist yet?
Yes, if the offer is described well enough for buyers to judge its value: what it does, how it compares and how it is sold. We often pair the price module with a concept test, so you learn whether the idea appeals before learning what it can charge.
Is survey-based pricing as reliable as a live price test?
They answer different questions. Live tests measure real behaviour, but only for prices you dare to try, and they need traffic and time. Surveys can explore prices and competitor moves you would never risk in the market. Use the survey to narrow the options and live data to confirm them.
How do you keep our pricing plans confidential?
We sign a non-disclosure agreement before the briefing, use neutral or disguised branding where needed and never tell respondents who commissioned the study. Personal data is handled under the GDPR and the ICC/ESOMAR International Code, and the file you receive is anonymised.
Related reports
Concept & product testing
Concept testing measures how target buyers react to a new product idea, recipe, pack design or brand name before you…
Read →Market segmentation
Market segmentation research divides a market into groups of buyers who share needs, attitudes and behaviour, then…
Read →Competitor analysis
A competitor analysis is a structured comparison of the companies your customers could choose instead of you — direct…
Read →Consumer research
Consumer research is the systematic study of how people buy and use products in a category: who buys, how often, where,…
Read →Sources
Updated: 5 Nov 2026