Pricing Survey Template
A faithful Van Westendorp price sensitivity meter — four price fences from real buyers that reveal your acceptable range and optimal price point.
Four questions about price, answered with numbers, not opinions. There are no right answers — your gut reaction is exactly the data we need.
Asking "would you pay $29?" is the worst way to price a product: the number anchors the answer, and people say yes to hypothetical prices they would never pay. The Van Westendorp Price Sensitivity Meter — the method this template implements exactly — avoids both traps by asking respondents to generate four price fences themselves: too cheap to trust, a bargain, getting expensive, and prohibitively expensive.
Why four numbers beat one opinion. Each question probes a different psychological boundary. "So cheap you'd question the quality" finds the floor where low price destroys credibility — a real effect that surprises founders every time. "A bargain" and "getting expensive" bracket the comfortable middle, and "too expensive to consider" marks the ceiling. Plot the four as cumulative curves across your responses and the intersections give you the acceptable price range and the classic optimal price point where "too cheap" and "too expensive" curves cross. The CSV export hands you the four columns ready for exactly that spreadsheet exercise.
Why the familiarity screen matters. Price answers from people who use the product daily mean something different from guesses by someone who heard of it this morning. The screen lets you weight or segment: many teams run the analysis twice — all respondents, then active users only — and price toward the second. The billing-preference question is a bonus with teeth, since "annually at a discount" versus "usage-based" changes packaging, not just the number. The final question — which capability justifies the price — tells you what to put at the top of the pricing page.
Sanity-check the fences before plotting. A coherent response rises monotonically — too-cheap below bargain, bargain below getting-expensive, that below too-expensive. Rows that break the order are speed-runs or misreads; strip them out before you build the curves. The validation bounds on each number field already stop stray zeros and negative prices at the door, which keeps the cleanup short.
What we left out. Feature checklists, competitor price comparisons, and income questions. Willingness-to-pay research works best clean: numbers in, curves out, no distractions that invite respondents to negotiate.
Who runs this. Founders pricing a first product, SaaS teams testing a plan restructure before announcing it, and consultants moving from hourly to packaged pricing who need evidence before the leap.
Make it yours. State the product plainly in the intro — one sentence of what it does — and set the currency in each question's placeholder. If your product is priced annually or one-time, change "monthly" consistently in all four questions. Aim for at least 30–50 qualified responses before trusting the curves; the close-after-N setting caps the study cleanly, and duplicate prevention keeps repeat submissions from bending the lines.
Frequently asked questions
How do I turn the four numbers into a price?
Export the CSV, build cumulative distributions for each of the four questions, and read the intersections — the range between "getting expensive" and "bargain" crossings is your acceptable window.
How many responses do I need for reliable curves?
Thirty qualified respondents is a working minimum; fifty-plus smooths the intersections. Use close-after-N responses in Settings to cap the study at your target.
Should I survey non-customers too?
Yes, but segment them — the familiarity question exists so you can split curves by exposure level in the export and price toward the segment you are actually selling to.
Can respondents enter nonsense numbers?
The number blocks accept digits only and are capped by validation bounds you can tighten per question. Obvious outliers are easy to trim in the CSV before plotting.