Willingness to Pay: How to Test It Before You Build
What willingness to pay means, why what people say they would pay is a weak guide, and seven ways a startup can test it before building, from interview questions and Van Westendorp to fake doors, pre-sales and paid pilots.
By Suren, Founder & Growth Strategist at Gufy · Published
Willingness to pay (WTP) is the highest price a customer will pay for a product before they walk away. For a startup the useful question is narrower: will a specific group of people pay a specific price for this, and how do you find out before you spend months building it? The honest answer is that only a payment proves it. Everything else is a clue.
This guide covers why what people say is a weak guide, the methods from cheapest to strongest, and how to read what comes back.
Why stated prices mislead
Ask someone whether they would pay $20 a month for a product that solves their problem and many will say yes. It costs them nothing to say it. Researchers call the gap between what people say and what they do hypothetical bias, and it runs in one direction: people overstate what they will pay.
Three things make it worse for founders:
- Friends and family. They want you to succeed and will tell you so. Talk to strangers, other founders or investors, who will tell you to your face if the idea is weak.
- Surveys that prove the problem, not the payment. A founder we spoke with had surveyed people in her target market. Every answer confirmed that the problem was real and that having everything in one place would help. The survey never asked whether they would pay for it. It validated the problem, which is useful, and said nothing about the business.
- Free alternatives. People already solve most problems somehow: a search, a general AI assistant, a spreadsheet, a group chat, a friend. Saying "I'd use that" is easy. Paying for it when the current way is free is a different decision.
Treat any price someone states as an upper limit to test, not a forecast.
Three things to validate, in order
| Question | Evidence | |
|---|---|---|
| Problem | Do people have it, and does it matter to them? | Interviews, surveys, what people already do about it |
| Solution | Would they use your way of solving it? | Sign-ups, usage of a prototype or manual version |
| Price | Will they pay enough for it to be a business? | Payments, deposits, signed paid pilots |
Most early validation stops at the first row. The money is in the third. Our guide to product-market fit covers the first two in more detail.
The evidence ladder
Each signal below costs the customer more than the one above it, which is why it tells you more.
| Signal | What it costs them | What it proves |
|---|---|---|
| "Great idea" | Nothing | Nothing |
| "I'd pay about $X" in an interview | Nothing | A rough ceiling, inflated |
| They already pay for a tool, a contractor or staff time to deal with it | Money they spend today | The problem has a budget |
| A price-range survey (Van Westendorp, Gabor-Granger) | A few minutes | A range people find acceptable on paper |
| Clicking "start" on a page that shows the price | Attention and intent | Interest at that price, from people you did not know |
| A refundable deposit or pre-order | Money, for a while | Willingness to pay, close to the real thing |
| A paid pilot, or paying for a manual version of the service | Money and time | Willingness to pay, proven |
Aim to get as far down the table as you can before you build the full product.
Seven ways to test it
1. Ask what they spend today
Start with the past, not the future. People are poor at predicting what they would do and fairly accurate about what they did.
- How do you deal with this today?
- What does that cost you, in money and in hours?
- What have you paid for to fix it before? Why did you stop?
- Who approved that spend?
If nobody has ever spent money or real time on the problem, a price will be a hard sell however good the product is. If they already pay a contractor or lose hours of paid staff time to it, you have something to price against. How to Price a SaaS Product shows how to turn that into a first price.
2. Ask the price question, in steps
At the end of an interview, after you have described what the product does, ask in this order:
- Would you use this?
- Would you pay for it?
- Would you pay [your intended price]?
- If not, what would feel fair?
The answer to the last question is the most useful. If several strangers in the same segment land on a similar figure, you have a starting range. If they all say they would use it and none would pay, you have learned the most important thing early and cheaply.
Five conversations with people who do not know you will tell you more than fifty with people who do. Our guide to jobs to be done has interview questions that keep the conversation on what people actually do.
3. Run a price-range survey
Once people understand the product, a survey can show where the acceptable range sits.
Van Westendorp's price sensitivity meter, proposed by the Dutch economist Peter van Westendorp in the 1970s, asks four questions:
- At what price would this be so expensive that you would not consider it?
- At what price would it start to feel expensive, so you would have to think about it?
- At what price would it be a bargain?
- At what price would it be so cheap that you would doubt the quality?
Plot the share of people giving each answer at each price. Where the lines cross gives a range of prices that most respondents find acceptable.
Gabor-Granger is simpler: show a price, ask whether they would buy at it, then raise or lower it and ask again. It gives a rough demand curve.
Both are good at narrowing a range and poor at proving anyone will pay. They are also weak before a product exists, because respondents are pricing something they have to imagine. Use them to choose which prices to test with real offers, not to set the price.
4. Put the price on a landing page
Write a page for one customer type with the outcome, the price and a button to start. Send traffic to it from ads aimed at that group, and measure how many people click the button with the price in front of them.
This is advertising used to validate, not to collect leads. You are testing which message and which price get strangers to act. Two or three versions of the page, each changing one thing, will tell you more than one page left running. Our guide to ICE and RICE scoring covers choosing and running those tests, and Marketing Budget for a Tech Startup Without Revenue covers what a test like this realistically costs.
What to compare:
- The share of visitors who click to start, at each price.
- The same page with and without the price shown. If showing the price cuts clicks sharply, the price or the message is the problem.
- Which audience clicks. The segment that responds is often narrower than the one you aimed at.
5. Use a fake door
A fake door is a button for something that does not exist yet: a paid plan, an upgrade, a feature. When someone clicks it, tell them straight away that it is not ready and offer to let them know when it is.
It measures intent at a price with very little to build. Two rules keep it honest: say what is happening at the moment of the click, and never take payment for something you cannot deliver.
6. Pre-sell it
Ask for money before the product is finished: a refundable deposit, a pre-order, or a founding-member price with a date for delivery. For a business product, a paid pilot with a defined scope does the same job.
This is the closest you can get to the real decision without a finished product. A handful of strangers paying a deposit is stronger evidence than hundreds of survey answers. A signed letter of intent is weaker than a payment: it shows interest, but it costs the signer nothing.
Refund promptly if you do not go ahead.
7. Sell the manual version first
Deliver the result by hand, or with tools that already exist, and charge for it. A software idea can start as a service. An app can start as a paid group, a newsletter or a booking you run yourself.
We often suggest this to founders about to build an app to test whether people will pay for something. An app is an expensive way to test a hypothesis: it costs a lot to build and adds friction (people have to find it, install it and remember to open it). If people will pay for the manual version, you have a business model to build software around. If they will not, you have saved the build. Our guide to the minimum viable product covers the cheaper formats.
Who said yes?
The people who pay are often a narrower group than the people who have the problem. Look at them closely:
- What do they have in common? Industry, size, role, situation.
- Why is the free alternative not good enough for them? Often it is not available to them, or their time is worth more than the price.
- Who approved the spend? In a business that may not be the person who will use the product.
That group is your first market. What Is an ICP? and How to Choose a Niche cover turning it into a profile and a focus.
Check the price against your costs
A price people will pay is only half the answer. It also has to cover what it costs to win and serve each customer:
- Payment and app-store fees, hosting, support and anything charged per use.
- The cost of acquiring a customer, which is usually higher than founders expect.
- The customers who cancel early.
If the most people will pay is below what it costs to reach them, the idea needs a different customer, a different model or a different price, not more marketing. Unit Economics for Startups works through the sums.
Common mistakes
- Asking friends. They will say yes.
- Asking about the problem and assuming the price. Ask about money directly.
- Building first. A full product is the most expensive possible way to find out nobody will pay.
- Taking survey prices as forecasts. They are ceilings.
- Halving a competitor's price. You inherit their guess and their cost structure, and still have not tested anything.
- Testing everyone at once. Willingness to pay differs by segment. Test one group at a time.
- Not refunding. A pre-sale that turns into a dispute damages the trust you were testing.
Common questions
What is willingness to pay?
Willingness to pay (WTP) is the highest price a customer will pay for a product or service before deciding not to buy. It differs from one customer to the next, so in practice it is a range across a group of customers, not a single number.
How do you measure willingness to pay before you have a product?
Ask people what they spend on the problem today, ask a direct price question at the end of each interview, and then look for behaviour: clicks on a priced plan, deposits, pre-orders or a paid pilot. Survey methods such as Van Westendorp give a price range, but only a payment proves someone will pay.
What are the four Van Westendorp questions?
At what price would this be so expensive you would not consider it? At what price would it start to feel expensive, so you would have to think about it? At what price would it be a bargain? At what price would it be so cheap you would doubt its quality? Plotting the answers gives a range of acceptable prices.
Is a fake door test ethical?
It can be, if you are honest straight away. Show the plan or feature, and when someone clicks, tell them it is not ready yet and offer to let them know when it is. Do not take payment for something you cannot deliver, and refund any deposit promptly if you do not go ahead.
Why do surveys overstate willingness to pay?
Saying yes to a price costs nothing, people want to be helpful, and friends want to encourage you. A survey can also confirm that people have a problem without ever asking whether they would pay to solve it. Treat stated prices as a ceiling to test, not a forecast.
Where Gufy fits
Gufy helps SaaS and mobile-app startups test an idea before they spend on building or scaling it: go-to-market strategy to choose the customer and the hypothesis, then landing pages and paid campaigns built to validate a message and a price, not just to collect leads. A growth workshop is a quick way to turn "people say they'd use it" into a test plan with a clear pass mark.
If you are about to commission a build and have not yet asked anyone for money, talk to us first.