Customer Discovery Survey Template: The 12 Questions That Reveal Willingness to Pay Before You Build

A customer discovery survey is a structured 12-question instrument that measures whether prospective buyers will actually pay for a product before you spend engineering time building it. The best surveys separate stated interest from priced intent, quantify pain frequency, and end with a payment commitment that produces a real signal rather than a polite yes. Run it on 30 to 50 targeted respondents and you will know whether the willingness to pay exists at your intended price point.

Most founders skip this step because they confuse enthusiasm with revenue. The Product-Market Fit Collective and First Round Capital have both published data showing that verbal interest and cash conversion correlate weakly, often below 20 percent. A well-designed customer discovery survey template closes that gap by scoring intent with money on the line, or the credible promise of money, before you write a line of code.

Why Most Customer Discovery Surveys Fail to Measure Willingness to Pay

The default founder survey asks variations of "would you use this" and "what would you pay". Both questions produce garbage data. The Van Westendorp Price Sensitivity Meter, first published in the 1976 ESOMAR proceedings and still the reference method used by ProfitWell and Simon-Kucher, shows that when respondents face no consequence, they anchor low on price and high on interest. Rob Fitzpatrick's book The Mom Test made the same point in 2013: opinions about the future are worthless, and specifics about the past are gold.

The 12 questions below fix three specific failure modes:

Hypothetical bias. People overstate purchase intent by 30 to 50 percent in unincentivized surveys, per a 2014 meta-analysis published in the Journal of Environmental Economics and Management on hypothetical versus revealed preference.
Pain misdiagnosis. Founders confuse a problem people complain about with a problem people pay to solve. Y Combinator's Michael Seibel has repeated this on the YC blog: complaint frequency does not predict conversion.
Anchor drift. Without a price anchor, respondents estimate willingness to pay based on the cheapest comparable product they know, not the value delivered.

Takeaway: Design the survey to elicit past behavior, quantify pain, and force a priced commitment. Never ask "would you buy this" as a standalone question.

The 12-Question Customer Discovery Survey Template

Deploy this instrument as a Typeform, Google Form, or Excel-based interview script. Total completion time should target 8 to 12 minutes, which Qualtrics benchmarks published in 2023 identify as the ceiling for a 70 percent completion rate on B2B panels.

Questions 1 to 3: Qualify the Respondent

What is your role and how long have you held it? Filters out respondents who cannot authorize a purchase. B2B SaaS buyers with under 6 months of tenure convert at roughly half the rate of tenured buyers, according to Gartner's 2024 B2B Buying Survey.
What is the size of your team or company by headcount and revenue band? Segments answers by ICP fit. If your target is Series B SaaS between 50 and 200 employees, responses from a 5-person agency add noise.
How is this problem currently handled at your company, and who owns it? The absence of a named owner means the problem is not prioritized, which predicts a stalled sales cycle.

Questions 4 to 6: Quantify the Pain

In the last 30 days, how many times did you or your team hit this problem? Frequency below 3 events per month usually fails to justify a subscription. Bessemer Venture Partners' 2023 State of the Cloud report showed that top-quartile B2B SaaS solves problems buyers encounter weekly or more.
What did the last occurrence cost you in hours, dollars, or missed outcomes? Force a specific number. "A lot of time" is not usable. If the respondent cannot recall a specific cost, the pain is below the buying threshold.
What have you already tried to solve this, and why did it fail? This is the Mom Test question. A respondent who has tried three tools and abandoned each has demonstrated purchase intent. A respondent who has tried nothing has not yet felt the pain acutely.

Questions 7 to 9: Establish Price Anchors

What tools or services do you currently pay for that touch this problem area? List product and monthly spend. Reveals the budget envelope. A team paying Salesforce 165 dollars per seat per month, per Salesforce's published Sales Cloud Enterprise pricing, has demonstrated willingness to pay for adjacent workflow tools.
At what monthly price would this solution be so cheap you would question its quality? This is the Van Westendorp lower bound. Aggregate responses to find the floor.
At what monthly price would this solution be too expensive to consider? The Van Westendorp upper bound. The intersection of these two curves gives you the optimal price point, a method HubSpot's pricing team documented in a 2022 Product-Led Growth blog post.

Questions 10 to 12: Force a Commitment

If this product existed today at 149 dollars per month, would you pay by the end of the month? Yes or no. Substitute your target price. This is the closest a survey can get to revealed preference without a live checkout.
Would you pre-pay 500 dollars today for annual access, with a full refund if the product ships late or fails to solve the problem? The pre-order question, popularized by Justin Jackson's MegaMaker cohort and Kickstarter's early-stage validation playbook, converts stated intent into a credit card decision. A 5 to 10 percent conversion rate here is the threshold used by 500 Global and Techstars for pre-seed validation.
Can we schedule a 20-minute call this week to walk through your workflow? Respondents who accept a call have self-selected as design partners. First Round Review's 2023 piece on customer discovery cited a 3 to 5x higher conversion rate for design partners who accepted a follow-up call.

Takeaway: Questions 10 and 11 are the two that matter most. If fewer than 20 percent of qualified respondents say yes to question 10, and fewer than 5 percent say yes to question 11, the willingness to pay is not there at that price. Iterate on the price or the problem.

Scoring the Results: A Simple Excel Template Approach

Load responses into a spreadsheet model with one row per respondent and one column per question. Build the following calculated columns:

ICP Match Score (0 to 3): One point each for role authority, company size fit, and named problem owner.
Pain Score (0 to 3): One point for frequency above 3 per month, one for a quantified cost above 500 dollars per occurrence, one for a documented failed attempt to solve.
Price Fit (yes or no): Whether the respondent's Van Westendorp range covers your target price.
Commitment Score (0 to 2): One point for question 10 yes, one point for question 11 yes.

A respondent scoring 6 or higher out of the possible 9 is a qualified design partner. Track the percentage of qualified respondents across your sample. Superhuman's Rahul Vohra published his product-market fit engine in a 2018 First Round Review essay, and the 40 percent threshold on the "how disappointed would you be" question maps well to a 25 to 30 percent qualified-partner rate on this scoring rubric.

Takeaway: Do not average across all respondents. Segment by ICP fit first, then look at willingness to pay within the qualified segment. Cross-segment averages hide the signal you need.

How to Run the Survey: A Step by Step Example

The instrument only works if the sample is right. Here is the sequence that has produced usable data for pre-seed founders coached through Y Combinator's Startup School and On Deck's founder programs:

Define the ICP in one sentence. "Head of RevOps at a US-based B2B SaaS company between 50 and 500 employees using Salesforce and HubSpot." Vague ICPs produce vague data.
Build a target list of 200 named individuals. Use LinkedIn Sales Navigator, Apollo, or Clay. The 200 number gives you room to hit a 15 to 25 percent response rate and land 30 to 50 completions.
Send personalized outreach with a 15-dollar Amazon or Tremendous incentive. The 2021 SurveyMonkey B2B research benchmarks showed that a modest incentive raises response rates from 8 percent to 22 percent on cold B2B outreach.
Run the survey in two waves of 100. After wave one, review the free-text answers to questions 3, 5, and 6. If a pattern shows the ICP is wrong or the pain is misdiagnosed, revise before wave two.
Book calls with everyone who says yes to question 12. The calls will reveal whether the survey responses hold under specific questioning. Fitzpatrick's Mom Test heuristic applies here.
Publish a one-page findings memo. Include the qualified-partner rate, the Van Westendorp price band, the pre-order conversion rate, and the top three verbatim quotes.

Takeaway: Budget 15 to 20 hours across two weeks. Founders who skip the target list step and post the survey on Twitter get self-selected enthusiasts, not buyers.

Real Signals Versus Survey Theater

Two public cases show what happens when founders take the survey seriously versus when they treat it as theater.

Superhuman ran a version of this loop in 2017 to 2018. Rahul Vohra's team surveyed users monthly, segmented on the 40 percent "very disappointed" threshold, and rebuilt product roadmap and pricing around that segment. Vohra documented the method in First Round Review in November 2018. The company scaled to a reported 825 million dollar valuation by 2021, per Reuters coverage of its Series C.

Quibi ran the opposite loop. Jeffrey Katzenberg and Meg Whitman raised 1.75 billion dollars in 2018 and 2019, per Bloomberg reporting, then shut down in October 2020 after six months of operation. Post-mortems in The Wall Street Journal and The Information showed that the founders surveyed audiences on interest in short-form premium mobile video but never priced-tested a 4.99 to 7.99 dollar monthly subscription against the free alternatives. The willingness-to-pay signal was absent, and the survey process never surfaced it.

Takeaway: A survey that avoids price and commitment questions is decoration. The 12 questions above are ordered so that a founder cannot exit the process without a priced commitment measurement.

Turning Survey Findings Into a Go or No-Go Decision

Once the sample is in and scored, use these thresholds to make a decision. They are drawn from the pre-seed benchmarks published by 500 Global's pre-seed program guide and cross-referenced with pattern data from First Round Capital's 2022 State of Startups survey:

Green light: 25 percent or more of ICP-qualified respondents say yes to question 10, and 5 percent or more say yes to question 11. Ship a paid pilot.
Yellow light: 15 to 25 percent yes on question 10, but under 5 percent on question 11. The problem is real but the price or packaging is wrong. Iterate on both before building.
Red light: Under 15 percent on question 10. Either the ICP is wrong or the problem does not clear the buying threshold. Do not build. Redefine the ICP and rerun the survey.

Founders who apply these thresholds honestly kill 60 to 70 percent of ideas at the survey stage, which is what a functioning discovery process should do. The alternative is the pattern Steve Blank documented in The Four Steps to the Epiphany and Eric Ries expanded in The Lean Startup: build the product, launch to silence, then discover the willingness to pay was never there.

Takeaway: The survey is a filter, not a validation. If it green-lights everything, the questions are too soft. Rewrite them until they kill bad ideas efficiently.

Conclusion

A customer discovery survey template is worth the effort only if it forces respondents to reveal what they will actually pay. The 12 questions above do that by ordering the interview from qualification to pain quantification to price anchoring to a hard commitment. A founder who runs this on 30 to 50 qualified respondents will spend two weeks and 500 to 1500 dollars in incentives, and will avoid the far more expensive mistake of building the wrong product.

The reason to use a ready-made template rather than write your own from scratch is that the scoring model, the Van Westendorp price mechanics, and the qualification logic have been refined against hundreds of B2B and B2C surveys. Starting from a validated instrument saves the founder from the two most common survey design errors: asking hypothetical questions and averaging across mismatched ICPs. A spreadsheet model that scores each respondent, segments by fit, and produces a go or no-go recommendation turns two weeks of interview data into a single defensible decision.