Hey Juno

Pricing research with qualitative interviews

By Hey Juno · Published · Updated

Qualitative pricing interviews tell you what price means to people: what "too expensive" refers to, how they judge value, and whether your pricing matches how they use the product. They do not tell you what price to set. Used with quantitative pricing methods such as Van Westendorp and conjoint analysis, they explain the numbers those methods produce. This guide covers what interviews can add to pricing research, how they pair with those methods, and the traps to avoid.

It is written for teams commissioning research, including with Juno, the AI researcher from Hey Juno.

What interviews can tell you that a pricing survey cannot

A pricing survey gives you a number from each respondent. An interview gives you the reasoning behind the number. Four things in particular are hard to get any other way.

What "too expensive" means. The phrase can mean the price is higher than a budget, higher than an alternative, higher than the value received or simply a surprise. Each leads to a different response. Asking someone to describe the last time the price came up, and what they compared it with, usually shows which one it is.

How people judge value. People rarely compute value. They use reference points: what they paid before, what a colleague pays, what the problem costs them, or what the cheapest alternative is. Interviews show which reference points your customers use, and who they believe they are comparing you with.

Whether the pricing metric matches how they feel value. Many products charge per seat, per use, per month or per feature. Customers may feel the value in something else, such as a project completed or a problem avoided. When the metric and the value point in different directions, people describe the bill as unfair even when the total is reasonable. You hear that in the words they use about it.

Where tier boundaries feel arbitrary. If a feature sits just above the line someone wants to cross, they notice. They tell you which plan they chose, which one they nearly chose and what made them stop. That is the detail that tells you whether a tier is working.

For related work on reasons behind a decision, see churn interviews and voice of customer research.

How qualitative research pairs with quantitative pricing methods

Quantitative pricing methods measure. Interviews explain. Each does something the other cannot, so the strongest studies often use both.

Van Westendorp's price sensitivity meter asks respondents four questions about a product: at what price it would be so cheap they doubt its quality, a bargain, getting expensive, and too expensive to consider. Plotting the answers gives a range of acceptable prices. It is quick and easy to field. It is also a stated-price method, so it shows the range people say they accept, not what they will do. The method comes from Peter van Westendorp's 1976 ESOMAR Congress paper, "NSS Price Sensitivity Meter (PSM): a new approach to study consumer perception of price". Chrzan and Orme give a plain overview alongside other survey-based pricing methods (Quirk's, 2022).

Conjoint analysis shows respondents products built from combinations of features and prices and asks them to choose. From the pattern of choices it estimates how much each feature and price level matters. Green and Srinivasan reviewed it for consumer research in their 1978 paper "Conjoint Analysis in Consumer Research: Issues and Outlook" in the Journal of Consumer Research. It handles trade-offs better than asking about features one at a time. It depends on you having already chosen the right features and price levels to test.

That dependency is where interviews help most:

  • Before a quantitative study, interviews tell you which features, price points and wording to include, and how customers describe the product, so the survey asks in their language.

  • After a quantitative study, interviews explain surprises: why a price band looks acceptable, why one feature outweighs another, why one segment responds differently.

An illustrative sequence: a team runs interviews and learns that customers see the product as paid by outcome, not by user, then tests a per-outcome tier in a conjoint study, then follows up with a few interviews on why one segment rejected it. This is an invented example of how the pieces fit, not a result.

Neither method is a substitute for watching what people do when they pay. Where you can, check what you learn against real purchase and usage data.

Pitfalls

Asking people what they would pay. This is the main one. Hypothetical answers are shaky. People anchor on the first number they hear, round to a comfortable figure and guess at a future they have not lived. Nielsen Norman Group makes a similar point in Attitudinal vs. Behavioral Research in UX (Laubheimer, 2024): what people say and what they do often differ, and you need both kinds of evidence. Instead, ask about decisions they have really made: the last time they paid for something similar, what they weighed and what changed their mind.

Leading with your price. Naming a price early makes the conversation about that price. Let people describe the problem and the alternatives first. See leading questions in research.

Talking only to people who bought. They have already decided the price is acceptable. Include people who looked and left, people who downgraded and people who chose a competitor.

Treating reasons as counts. Interviews can show that a concern exists and how it works. They cannot say how many customers hold it. Say so in the report.

Mistaking a stated reason for the full reason. People tend to give reasons that sound reasonable. A price complaint may sit on top of a doubt about value. Ask for the specific story.

Running pricing interviews with Juno

You describe the decision you face, such as whether to change a tier boundary, and Juno turns that into research objectives, not a list of questions. Juno interviews real people by voice or text, in their own language, and follows up on what each person says.

Findings come back as claims, each linked to the participant, the quote and the transcript moment. Hey Juno describes pricing objections as one use. You can ask Juno what it found and what did not hold up. Simulated interviews are for rehearsing a study, never for findings. Recruit customers through invite links or reach strangers through a panel. See participant recruitment for AI interviews and checking AI research findings.

Where this doesn't apply

Interviews will not set a price. If you need a demand curve or a revenue forecast, you need quantitative methods and, ideally, live experiments. They also cannot replace testing a price in market. With few customers, or a product too new for anyone to have formed a view, you will hear speculation more than experience. Price changes can carry legal and commercial constraints, so check those with your own legal team.

Frequently asked questions

Can interviews replace Van Westendorp or conjoint studies?

No. They answer a different question. Interviews explain how people think about price and value. Van Westendorp and conjoint estimate ranges and trade-offs across many respondents. Use interviews to design and interpret the quantitative work.

What should I ask if I can't ask what people would pay?

Ask about real decisions. "Tell me about the last time you chose between two options like this" or "What did you compare it with?" work better than any hypothetical price. Ask what they were doing before the product, and what it cost them.

How many people should I interview?

It depends on how varied your customers are. A small number can show recurring themes in a uniform group. If you want to compare segments, such as small and large accounts, plan for more in each. See qualitative interviews at scale.

Is willingness to pay the same as value perception?

No. Willingness to pay is what someone says they would spend. Value perception is how they judge what they get. The second explains the first, and interviews are the better way to find it.