Testing

What Is a Fake Door Test and When Should You Use One?

Fake door testing measures interest in a feature or product before the underlying capability is fully built.

A fake door test presents a real-looking entry point to a product, feature, or offer that is not yet fully available. When a user clicks, they are told the capability is coming soon or invited to join a pilot.

Start with the uncertainty that matters

The method is useful because clicks reveal behavioral interest rather than stated preference. Someone saying they would use a feature is weaker evidence than that same person actively trying to access it.

A good fake door test should be honest at the point where the action would otherwise create a false expectation. The goal is to measure interest, not trick users into believing something exists when it does not.

What to measure

Use fake door tests for questions like: which of three proposed capabilities attracts the most interest, whether existing users care about a proposed workflow, or whether a new market responds to a specific outcome.

They work best when the decision being tested is narrow. A click on “export to Salesforce” tells you something about interest in that integration. It does not tell you whether the user would pay for the whole product.

How to run the test

Pair the result with a follow-up question or interview where possible. The click gives you behavioral evidence; the conversation helps explain motivation, urgency, and expected value.

Avoid fake door tests where the user would incur a meaningful cost or where trust is especially sensitive. They are a research technique, not a shortcut around customer communication.

Common questions

What makes this a useful validation method?

It looks for a concrete behavior from the intended buyer rather than relying only on opinions or hypothetical interest.

How much traffic or data is enough?

There is no universal number. The useful threshold depends on audience quality, conversion difficulty, and the decision you are trying to make. Pre-define the sample you consider large enough before interpreting the result.

Can a weak result still be useful?

Yes. A weak result can tell you that the audience, problem framing, offer, price, or channel needs to change before you invest further.