BuildInSeven
← All articles

Fake Door Test: Validate Demand Without Building

7 min read

Learn how to run a fake door test to measure real purchase intent before writing a line of code. Step-by-step guide for non-technical founders.

A fake door test presents a product or feature to real potential buyers before that product exists. When someone clicks "Sign Up" or "Buy Now," they hit a page explaining the product is coming soon, or a waitlist form. The click rate and sign-up rate tell you whether demand is real, without spending anything on development.

That is the whole mechanism. The rest of this article covers how to set one up properly, what numbers to trust, and when the results mean you should build.

Why Clicks Reveal More Than Surveys

Asking someone "Would you pay for this?" is nearly useless. People are polite. They say yes to avoid awkward conversations, and they have no skin in the game when answering a hypothetical.

A fake door test asks people to take an action, not form an opinion. Clicking a button or entering a credit card number costs the person something, even if only three seconds and a small dose of intent. That cost filters out the noise. The signal that remains is far closer to actual purchase behavior.

This is why fake door tests consistently outperform surveys as a validation tool. The friction of acting reveals preference more honestly than the ease of answering.

What a Fake Door Test Actually Looks Like

The format varies, but the core structure is the same: a real-looking offer page, a call-to-action button, and a transparent follow-up when someone clicks.

Common formats:

FormatWhat You ShowWhat Happens on Click
Landing page + CTAProduct description, pricing, benefits"Coming Soon" page or waitlist form
Fake pricing pageTiered plans with a "Choose Plan" buttonWaitlist capture with honest explanation
Ad-driven testPaid or organic ad leading to offer pageSame landing page as above
In-product feature buttonButton for an unbuilt feature inside existing softwareModal explaining the feature is in development

The "coming soon" disclosure matters. Being transparent that the product does not yet exist is both ethically required and practically smart. People who sign up after seeing that disclosure are warm leads. They know exactly what they are waiting for.

Running a Fake Door Test in Five Steps

1. Define the one thing you are testing

Pick a single value proposition, not a feature list. "Automated invoice chasing for freelancers" is testable. "An all-in-one business management platform" is not, because you cannot tell which part of the pitch drove the click.

Write one sentence that describes the outcome the customer gets. That sentence becomes the headline on your test page.

2. Build a minimal landing page

You do not need a developer for this. Tools like Carrd, Notion, or even a simple Mailchimp landing page are enough. The page needs:

  • A headline stating the specific benefit
  • Two or three bullet points on what the product does
  • A price point or price range (leaving it vague weakens the test significantly)
  • One CTA button

Price is the most important element most founders leave out. A click on a free product tells you very little. A click on a $49/month product, followed by a waitlist sign-up, tells you someone found the value worth that number.

3. Drive targeted traffic

A fake door test with twelve visitors proves nothing. You need at least 200 to 500 visitors to draw any conclusion, and they need to be the right people.

For finding and reaching potential early customers, How to Find Early Users Before You Build covers specific sourcing approaches in detail. In brief: Reddit communities, LinkedIn posts, niche newsletters, and small paid social campaigns targeting a narrow job title or interest are all viable sources. Do not use general audiences.

4. Set your success threshold before you look at the data

Decide in advance what conversion rate would signal genuine demand. A common benchmark for a B2C product at a modest price point is 5% clicking through to the CTA and 2% completing a waitlist sign-up. For a B2B product at higher price points, even 1% click-through with five qualified sign-ups in a week can be meaningful.

Setting the threshold before you see results prevents motivated reasoning. If you decide 3% is the bar and you get 3.1%, you build. If you get 0.8%, you either pivot the framing and retest, or you reconsider the idea.

5. Follow up with everyone who clicked

The quantitative result (conversion rate) tells you whether demand exists. The qualitative follow-up tells you why. Email every person who signed up and ask two questions:

  • What problem were you hoping this would solve?
  • What have you tried before?

Those answers shape your actual product scope better than any brainstorming session.

What the Numbers Are Actually Telling You

A high click rate with a low sign-up rate usually means your headline is compelling but your explanation or price creates doubt. Test the explanation or the price.

A low click rate from a relevant audience usually means the value proposition itself is not landing. The audience understood the offer and chose not to engage. That is important information.

A high sign-up rate with zero email responses to your follow-up questions suggests people signed up out of politeness or curiosity, not real intent. This is less common when you include a price on the page, which is another reason not to skip pricing.

Fake Door Tests and the Cost Decision

A fake door test should happen before any conversation about development cost or structure. Once you have real click and sign-up data, those conversations become much more productive.

If you are weighing how to structure a development engagement after your test shows positive results, Fixed Price vs Hourly: Which Is Safer for Founders? covers the trade-offs in practical terms.

Validation data also changes the scoping conversation. A product with 40 waitlist sign-ups and clear answers to "what problem does this solve" is far easier to scope correctly than an idea described from memory. Founders who skip validation often scope based on what they imagine customers want, then discover the mismatch after money has been spent.

For a fuller view of how fake door tests fit within a broader validation sequence, How to Validate a Software Idea Before You Build covers the complete process.

Common Mistakes That Corrupt the Results

Testing with friends and colleagues. People who know you click out of support. Use strangers from relevant communities.

Omitting price. A price creates real friction. Without it, you are measuring curiosity, not purchase intent.

Sending traffic to a generic homepage. The test page must match the exact framing of whatever drove the click. If your ad says "invoice chasing for freelancers," the landing page must say the same thing.

Running the test for one day. Weekly patterns matter. Run the test for at least five to seven days to capture behavior across the week.

Treating the test as pass/fail. A low conversion rate is not a failure if it tells you that the framing was wrong, the price was too high, or the audience was off. Reframe and retest before abandoning the idea.

When a Fake Door Test Is Not Enough

A fake door test measures intent, not capability. It cannot tell you whether you can build the product at a cost that allows a viable business. It cannot tell you whether a regulation blocks your approach. It cannot replace conversations with the people who signed up.

For complex B2B products where the buyer and the user are different people, a fake door test may undercount real demand because budget holders rarely click ads. In those cases, direct outreach to specific companies often produces better signal.

Think of the fake door test as one instrument in a small panel, not the only reading you need before committing to build.

Frequently asked questions

Is a fake door test ethical?
Yes, provided you are transparent. The test page or the follow-up screen should tell people the product is not built yet and invite them to join a waitlist. Charging someone's card for a product that does not exist is not a fake door test, it is fraud. Transparency also improves your data quality, because only genuinely interested people complete the sign-up once they know the product is not live.
How much traffic do I need for a fake door test to mean anything?
A minimum of 200 targeted visitors gives you enough data to read a trend, though 400 to 500 is more reliable. The critical word is targeted. Two hundred visitors from a Reddit community of your exact target customer is worth more than 2,000 visitors from a broad Facebook audience.
Do I need to include a price on the test page?
Including a price produces much more useful data. Clicks on a free offer tell you people found the headline interesting. Clicks on a specific price point tell you people found the value worth that number. If you are unsure of pricing, run two versions of the page with different price points and compare conversion rates.
How long should a fake door test run?
Run it for at least five to seven days. A single day can catch a spike from one traffic source that does not represent typical behavior. A full week captures variation across different days and gives you a stable conversion rate to evaluate.
What should I do if my fake door test shows low conversion?
Before abandoning the idea, change one variable and retest. Common culprits are a headline that does not match what the audience cares about, a price that is too high without enough explanation of value, or traffic from an audience that is not actually the buyer. Treat a low result as a diagnostic, not a verdict, until you have tested at least two or three meaningful variations.