Prompt and Use Case
You own a B2B SaaS product for small and mid-sized businesses. It has a prototype and a few trial users but no stable willingness-to-pay curve. The product lead wants a price range and a next-step validation plan in two weeks. Explain whom to recruit, what to ask first, how to avoid anchoring, how to segment the analysis, when to use Van Westendorp or conjoint, and which results trigger a paid pilot or a return to positioning.
The question tests research design, not the ability to guess a “correct price.” Assume the product is not yet generally paid for. Recruit users, budget owners, and prospects who evaluated but did not buy; treat sample sizes and prices as hypotheses to validate.
What the Interviewer Evaluates
A strong answer decomposes “what would you pay?” into value references, the buying process, alternatives, budget ownership, and price objections instead of asking for a number in a vacuum. The interviewer also wants you to distinguish qualitative interviews that answer “why,” quantitative methods that estimate a distribution, and real transactions that test “will they actually pay?”
Weak answers interview only current users, show a price first, treat polite approval as an order, or call a Van Westendorp intersection the optimal price. A strong answer states the anti-anchoring order, sample strata, evidence thresholds, and stopping rules.
Clarifying Questions Before Answering
Which decision must the research support?
Launch pricing, packaging changes, entering a segment, and diagnosing churn require different evidence. If the decision is whether to run a paid pilot, focus on buying conditions. If it is seat versus usage pricing, compare the value unit with budget predictability.
Who owns the budget and who uses the product daily?
Users know the pain, budget owners approve purchases, and procurement or security may control launchability. Combining them into one average hides the actual constraint.
What is the alternative?
Respondents may compare spreadsheets, manual work, a competitor, internal development, or doing nothing. A price without a reference cannot explain what “expensive” means.
Is the product and price concept concrete enough?
Quantitative research must describe a specific tier, usage, service level, and time period. A vague concept produces vague numbers that should not be pooled.
30-Second Answer Framework
“I would define the pricing decision and target segments first, then recruit users, budget owners, and evaluated-but-lost prospects across different outcomes. Interviews would start with real workflows, alternatives, budget, and evidence of value, and discuss prices only after that context. I would probe every number for its reference point and buying condition, code value and objections by decision role, and turn the themes into testable hypotheses. Once qualitative themes stabilize, I would use segmented price sensitivity or conjoint, then validate behavior through quotes, preorders, or a paid pilot. If objections point to positioning, trust, or delivery cost, I would change that hypothesis rather than simply cut price.”
Step-by-Step Deep Dive
1. Write a decision contract
State the decision due in two weeks, such as “choose a testable range for one buyer segment and decide whether to run a paid pilot.” Record constraints that cannot be violated: a margin floor, implementation capacity, predictable billing, compliance, or sales-cycle limits. Without this contract, interviews turn into a collection of attractive quotes.
2. Recruit by buying role and outcome
Cover at least three groups: prototype users, budget owners involved in procurement, and prospects who evaluated but did not buy. Segment further by company size, pain severity, and existing alternative. Do not claim that “12–15 people” fits every situation; a public B2B SaaS framework uses 12–15 semi-structured interviews per buyer segment as a starting point, while saturation still depends on repeated themes and segment differences.
3. Ask facts, then value, then price
Start with the last real occurrence: who did what, how long it took, what failed, and how it is handled today. Ask who drives approval, which budget pays, and what would replace the product tomorrow. Discuss price only after the value reference; otherwise the first number becomes an anchor.
1. Walk me through the last time you handled this problem.
2. Which step consumed the most time or created the most loss? How do you know?
3. If you could not use this product tomorrow, what would you do instead and what would it cost?
4. Who participates in the purchase decision, and what evidence or approval is needed?
5. Which outcome or capability is worth protecting? Why?
6. Given this specific scope, usage, and service, what price feels too cheap, a bargain, expensive but acceptable, or too expensive?
7. What is each price compared with, and what condition would change your judgment?4. Probe “expensive” and “willing”
“Too expensive” may mean insufficient budget, unproven value, low trust, switching cost, or the wrong scope. Ask “Compared with what?”, “What result would make it reasonable?”, “Who would object?”, and “What if the feature or service changed?” Convert emotional conclusions into hypotheses. Do not treat the highest stated number as forecast revenue.
5. Bound a Van Westendorp study
The Van Westendorp Price Sensitivity Meter asks four price questions: so cheap that quality is suspect, a bargain, expensive but still considered, and too expensive to consider. It provides exploratory signals for an acceptable corridor, not an automatic optimal price, and cannot replace segments, competitive references, or behavior. In an interview, its value comes from probing the logic behind each answer rather than collecting four numbers alone.
6. Use conjoint only when trade-offs are real
If buyers must trade features, brand, service, and price, conjoint can approximate choice behavior better than a single price question. It also costs more to design, sample, and analyze. Use it when the pricing stakes justify the investment and the attributes are concrete. If the concept is unstable, qualitative work should come first.
7. Code the analysis and search for counterexamples
Record buyer role, alternative, value outcome, budget source, price reference, objection, and next action for every interview. Separate “willing to pay” from “willing to try under conditions,” and current-customer satisfaction from lost-prospect rejection. Search for counterexamples: high pain but no purchase, low pain but renewal, user approval but budget-owner rejection. Counterexamples often change packaging or positioning more than an average number does.
8. Set gates from qualitative to quantitative to behavior
The qualitative gate is repeated value themes and objection reasons within each segment; only then estimate a distribution quantitatively. A public B2B SaaS framework suggests roughly 100 responses per segment as a starting point for Van Westendorp-style methods, but that is not a universal statistical guarantee; state the sample, concept, and uncertainty. Behavioral validation uses real quotes, preorders, refundable pilots, or sales opportunities and observes conversion, discounting, activation, retention, and support cost.
9. Map results to actions
If users value the outcome, budget owners can approve, and quote behavior is stable, run a small paid pilot. If value is clear but the unit is hard to budget, change packaging or billing. If themes are scattered, return to positioning or segment definition. If objections center on trust or implementation risk, add evidence, trial, and service boundaries. A price cut is one action, not a substitute for fixing value, channel, or cost.
High-Quality Sample Answer
“I would turn the research into a decision: choose a testable range for one buyer segment and decide whether to run a paid pilot. I would recruit users, budget owners, and prospects who evaluated but did not buy, so satisfied users do not dominate the evidence.
I would begin with the last real workflow, alternatives, time or loss, approval process, and valued outcome, and show a concrete product and price concept only at the end. For every ‘too expensive’ response I would probe the reference point, buying condition, and objector. Van Westendorp can explore a price corridor, but an intersection is not the optimal price; I would use conjoint only when feature-price trade-offs are clear.
I would code roles, alternatives, value, and objections, and look for high-pain nonbuyers or user approval blocked by the budget owner. After themes stabilize, I would validate by segment quantitatively, then observe quotes or a refundable pilot for conversion, discounts, activation, retention, and support cost. If value is clear but billing is unpredictable, I would change the unit; if value is unclear, I would return to positioning instead of cutting price.”
Common Mistakes
Ask “What would you pay?” first → no reference and polite anchoring → reconstruct the workflow and alternative first
An abstract number does not explain value or budget. Put price after facts and value, then probe what the number is compared with.
Interview only current users → miss buying and rejection constraints → stratify by role and purchase outcome
Satisfied users explain usage value but not procurement approval, loss reasons, or competitive alternatives.
Call a Van Westendorp intersection the optimal price → ignore segments, cost, and behavior → treat it as an exploratory corridor and validate transactions
The four questions describe perceived price boundaries, not guaranteed conversion or margin.
Run a large survey before the product concept is clear → measure the wrong object precisely → clarify attributes qualitatively first
Tier, usage, service, and scope must be concrete before responses are comparable.
Hide segment differences behind an average → the price fits nobody → analyze by role and segment and retain counterexamples
Users, budget owners, trials, and lost prospects use different reference points and constraints.
Treat polite approval as willingness to pay → overstate demand → set quote, preorder, or pilot behavior gates
Behavior exposes budget, trust, procurement, and switching cost; verbal approval only creates a hypothesis.
Follow-Ups and Responses
The budget owner says the price is fine, but every sales opportunity asks for a discount. How do you interpret it?
Check whether it is truly a price issue. Value proof, contract risk, procurement authority, or competitor packaging may be the cause. Code discount requests by segment, stage, and objection, compare list quotes with conditional discounts and pilots, then decide whether to change price, packaging, or evidence.
After 12 interviews, themes still do not repeat. Recruit more or run the survey?
First check whether recruitment mixed segments or the question order produced shallow answers. If the core buyer segment still has new themes, do not rush to quantification; complete the strata and record a stopping rule. Quantitative answers become comparable only when the concept and major value themes are stable.
Customers will pay to “save time” but cannot provide reliable hours. What do you do?
Treat time saved as a direction, not a monetary fact. Ask about frequency, workflow steps, people involved, and alternative cost, then find observable behavior or outcome metrics. Use a pilot to estimate a range and carry uncertainty into the price hypothesis; do not multiply a verbal hour estimate by salary and call it a willingness-to-pay ceiling.
When should you stop pricing research and return to positioning?
When segments disagree about the problem, outcome, and alternative, or objections concern trust and usability rather than money, more numbers add noise. Narrow the audience, clarify the outcome and concept, then restart the research.