On this page
To research willingness to pay, don't ask "How much would you pay?" Ask participants about their last real purchase decision: what they compared, what worried them, and why they bought or didn't. A price named out of thin air is heavily swayed by context, anchors and the wish to please, while past decisions leave a record you can check. After reading this guide, you'll be able to design a purchase decision interview guide, know when to pair it with Van Westendorp or Gabor-Granger for quantitative research, and sort price concerns separately from value concerns.
Why asking "How much would you pay?" is unreliable
- Naming a price costs nothing: participants don't actually have to pay, so their answers may run high, or they may deliberately lowball to push the price down.
- No point of comparison: in a real purchase, people weigh alternatives, competitors and not buying at all. When you ask about price in the abstract during an interview, none of that is in front of them.
- Anchoring: any number the interviewer says, or any price the participant just saw, affects the answer.
- Price is often the scapegoat: "too expensive" is the easiest way to say no, and behind it there is usually a value or trust problem.
So the job of the interview is to reconstruct the decision process. Leave the price itself to quantitative methods.
When to use it, research goals and participants
- Few trial or free users convert to paid, and you want to know where they get stuck.
- Before pricing a new product or plan, to learn what users compare it with and what they care about.
- Before or after a price change, to find out which users might leave because of it.
- "I'll think about it" is what sales hears most in follow-ups, and you want to know the concerns behind it.
Write the research goal as a decision question, for example "Are users who didn't pay after the trial mainly held back by price or by value?" or "What price range for the new plan won't scare off core users?" The first relies mainly on interviews; the second needs interviews plus quantitative research.
Who to interview: select by behavior
- Trial users who didn't convert: users whose trial ended 1 to 3 weeks ago, who actually used the product but didn't pay. For those who never used it at all, the problem is activation, not payment.
- Left after viewing pricing: users who recently visited the pricing or checkout page and didn't buy.
- Recent buyers: users who paid within the last month. As a comparison group, they tell you how they talked themselves into it.
- Affected by a price change: users who downgraded or canceled after a price adjustment.
Ideally interview both non-paying and paying users, 10 to 15 per group. If you only talk to people who didn't buy, you'll learn what their concerns are but not which concerns can be overcome.
Interview guide
- [Opening] Thanks for making the time. Today I'd like to understand how you thought about (product). Whether you bought it or not, there's no right or wrong answer, and we won't try to sell you anything.
- [Opening] What first got you looking for this kind of product?
- [Core topic: last purchase decision] Think back to the most recent time you spent money on this kind of need. What did you buy? What steps did you go through between first thinking about it and placing the order?
- [Core topic: last purchase decision] What other options did you look at? Why did you end up choosing that one?
- [Core topic: this decision] During the trial, or when you saw the price, what were you thinking?
- [Core topic: this decision] At the moment you finally decided to pay or not, what made up your mind?
- [Core topic: what they compare it to] Without (product), how would you handle this? How much money and time would it take?
- [Core topic: value] During the trial, was there a moment when it felt genuinely useful? If not, what were you hoping to see?
- [Core topic: concerns] What else made you uneasy? For example, results, the contract, payment options, or needing to check with someone.
- [Probes] If you hear "too expensive": Expensive compared with what? What price would feel right, and why that number?
- [Probes] If you hear "I'll think about it": What do you still need to confirm? Whose opinion do you need?
- [Probes] Separate price from value: If the price were cut in half, would you buy it right away? What else would be missing?
- [Probes] Ask who decides: Is this your call, or does someone else need to approve it?
- [Closing] For you to pay today, what's the one thing that would most need to change?
- [Closing] Anything else you'd like to add? Thank you.
"If the price were cut in half, would you buy it?" is the key question for separating price concerns from value concerns. If the answer is still no, price isn't the problem.
Separate price concerns from value concerns
| Type of concern | Example quote | How to respond |
|---|---|---|
| Price concerns | It's a good product. It's just a lot of money to pay all at once. | Payment options, installments, smaller plans, entry-level pricing |
| Value concerns | I used it for two weeks and didn't feel it was much better than what I do now | Help users see results sooner and explain clearly what they'll get |
| Trust concerns | I'm not sure the results will last. I'm worried I'll buy it and not use it. | Refund guarantees, case studies from similar users, monthly billing |
| Decision concerns | I need to talk it over with my family or my manager | Materials for the decision maker, trial extensions |
Each of the four types of concern calls for completely different changes. If you log them all as "too expensive," the team will likely cut the price, conversion won't move, and revenue will drop.
Why trial users don't convert to paid
Common reasons trial users don't pay: they never reached the step where they could see results during the trial; they happened not to need it when the trial ended; the checkout flow or payment options were inconvenient; they didn't get a reminder before it expired. In interviews, first ask what they did during the trial and which time was most useful, then ask about the day it ended. Don't open by asking why they didn't pay.
Sample dialogue: the participant says it's expensive, but really means it isn't worth it
This is a fictional example. The participant says it's expensive, but probing reveals they use it rarely and the free tool is good enough, which is a value concern. Lowering the price won't do much for users like this; the more useful group to study is high-frequency users.
When to pair interviews with Van Westendorp and Gabor-Granger
Interviews tell you why people buy or don't; quantitative methods tell you where to set the price. Usually you interview first and quantify second: once interviews show what users compare against and where the value lies, you can write accurate product descriptions and price points for the survey.
| Method | How to ask | Good for |
|---|---|---|
| Van Westendorp Price Sensitivity Meter | Four questions: at what price would it be so cheap you'd doubt the quality; at what price is it a bargain; at what price does it feel expensive but you'd still consider it; at what price is it too expensive to consider | When the category is new or there is no price precedent, to find an acceptable price range |
| Gabor-Granger pricing method | Present several prices one at a time and ask whether they would buy at each price | You already have a few candidate prices and want to estimate purchase rates and revenue at each one |
- Both methods rely on self-reported answers. They give you relative comparisons, not actual purchase rates.
- Size the sample to quantitative standards. A dozen or so interviews isn't enough for either analysis.
- Van Westendorp doesn't account for competitor prices, and Gabor-Granger is very sensitive to how the product is described. Interviews fill exactly these two gaps.
- In interviews you can borrow the Van Westendorp questions as open-ended prompts, but only to hear the reasoning, not to calculate a price.
How to synthesize findings
- Map each participant's decision process: where the need came from, what they compared, the key moment, and the final decision.
- Sort the reasons for not paying into four types: price, value, trust and decision. Give each one a participant count, quotes and counterexamples. Example: of 16 participants who didn't pay, 9 said their main concern was not seeing results, and 3 said it was specifically the pressure of a one-time payment.
- Compare with the paid group: which concerns also came up among paying users, and how they got past them.
- Write down the change that addresses each type of concern, and mark which changes don't require lowering the price.
- If you ran quantitative pricing research, interpret what people compared against in interviews alongside the price ranges you measured.
Common mistakes
- Asking directly how much they'd pay: the numbers are unstable and don't tell you why. Ask about the last real purchase instead.
- Treating every "too expensive" as a price problem: first ask whether they'd buy at half the price, to separate price from value.
- The interviewer naming a price first: once a number is said out loud it becomes an anchor, and every answer after that revolves around it.
- Only interviewing people who didn't pay: with no comparison, you can't tell which concerns can be overcome.
- Using a handful of interviews to set prices: price ranges and demand curves need a quantitative sample.
- Ignoring the decision maker: in household or business settings, the person who uses the product isn't always the one who pays.
Validate findings with behavior data
- When selecting participants, use behavior data to tell how deeply they used the trial, for example whether they used core features during the trial, so you don't mistake an activation problem for a willingness-to-pay problem.
- Check interview findings against the data: if the main reason is not seeing results, see whether users who reached key features during the trial convert to paid at a higher rate.
- After changes ship, track conversion: if you change payment options, track the paid conversion rate and first-month renewal rate; if you show results earlier, track trial-to-paid conversion.
- Where possible, roll out price changes through an A/B test or in stages, and watch renewals over the following months, not just the first purchase.
How to do it in heari
With heari, you can select users in Agentic Engine by behavior, such as trial users who didn't pay, users who left after viewing the pricing page, or users who just paid, and AI drafts the study plan and interview guide. After review, AI runs many interviews at once, starting from the last purchase decision and probing what people compared and what worried them. In the insight report, each finding comes with quotes, counterexamples and participant counts, and the research lead can review each one. After release, you validate the findings with behavior data.
FAQ
Why can't you just ask users how much they'd pay?
Because naming a price costs nothing, and participants don't have the alternatives they'd weigh in a real purchase, their answers are easily swayed by anchors and the wish to please. A more reliable approach is to ask about their last real purchase and leave price ranges to quantitative methods such as Van Westendorp or Gabor-Granger.
Should you use Van Westendorp or Gabor-Granger?
Use the Van Westendorp price sensitivity meter when there's no precedent for the price and you first need to find an acceptable range. Use the Gabor-Granger method when you already have a few candidate prices and want to compare purchase rates and revenue at each one. Run both after the interviews, with a sample sized to quantitative standards.
Who should you interview for willingness-to-pay research?
Select by behavior: users whose trial ended recently and who used the product but didn't pay, users who recently viewed the pricing page without buying, and recent buyers as a comparison group. Run 10 to 15 interviews each with the non-paying and paying groups so you can see which concerns can be overcome.
How can you tell whether users think it's too expensive or not worth it?
Ask one follow-up question: If the price were cut in half, would you buy it right away? If the answer is yes, it's most likely a price concern, and you can start with payment options and plans. If the answer is still no, it's a value concern, and lowering the price won't help much; you need to help users see results sooner.
