Description
Present an entry point, offer or landing page for something not yet built, count how many exposed people act on it, and compare with a preset threshold.
These tests estimate demand from behaviour rather than prediction. Forms vary by where the door sits: a landing page or crowdfunding page with a buy, pre-order or sign-up action reached by targeted ads (smoke test); an in-product button, menu item or field that leads to a 'coming soon' note (painted door or feature stub); or a static upgrade page with price and benefits, A/B tested across prices or messages. The metric is actions over exposures, judged against a threshold set before launch. Randomising exposure turns it into a controlled experiment that can also reveal harm: an inert coupon field at checkout reduced revenue by sending users off to hunt for codes. A click is intent, not purchase; asking for real commitment (pre-order, payment) is stronger. Ethics matter: keep exposure small, be honest with clickers and avoid harmful versions.
Attributes
- Discipline
- Experimentation & Causal Inference
- Loss type
- Active
- Stage
- Strategy, Discovery, Definition
- Data
- Quantitative
- Effort
- Days
Steps
- State the demand hypothesis and, before launch, the ratio of actions to exposures that will trigger building (or pivoting).
- Choose the door: a landing or pre-order page reached by targeted ads, an in-product button or menu item, or a static price page.
- Build the smallest credible entry point where users would expect it, plus an honest 'not available yet' message; for landing pages make the design credible.
- Expose a known, small number of target users, ideally randomised against a control that does not see the door so the overall evaluation criterion can be compared.
- Log exposures, clicks and flow progress; optionally ask a few questions at the start of the flow or A/B test prices and benefit wording.
- Thank, apologise to and if possible compensate people who clicked; call sign-ups to learn what they thought was promised.
- Compare the rate with the threshold and decide to build, pivot or drop, remembering that a click measures intent, not use or payment.
Sources
- UX for Lean Startups (Klein, 2013)
- Validating Product Ideas (Sharon, 2016)
- Designing with Data (King et al., 2017)
- Lean Analytics (Croll & Yoskovitz, 2013)
- Trustworthy Online Controlled Experiments (Kohavi et al., 2020)
- Product Management for UX People (Crumlish, 2022)
- UX Strategy (Levy, 2015)
Also known as
- Fake door test
- Painted door test
- Fake feature test
- Feature stub
- Fake button test
- Button to nowhere
- 404 testing
- Smoke test
- Landing-page test
- Landing page MVP
- Fake buy button landing page
- Pre-order test
- Teaser landing page
- Smokescreen test
- Broken glass test
- Hard test
Categories
- Analytics & Behaviour
- Experimentation & Causal Inference