A wizard-of-oz MVP looks like working software to the user, but humans operate it behind the curtain. It tests whether people want the automated experience before you pay to automate it — especially useful for AI products.
Use it when the expensive part is the automation itself and the open question is demand for the experience.
The wizard tests desirability, not feasibility — users wanting it doesn't prove the automation is buildable at acceptable cost. AI products especially need a feasibility check alongside the demand test.
Practical answers
It's standard practice when the gap is disclosed appropriately and no harm rides on it. You're testing the experience, then building the machinery demand justifies.
Until the demand signal is clear — typically a few weeks. Past that, manual operation costs compound and the learning flattens.
Related pages
MVP Development
MVP development agency for founders: product scope, UX, engineering, and launch support from €7,000 — without overbuilding the first version.
Read pageGlossary
Concierge MVP explained: validating a product by delivering the service manually before building software — when it works and how to run one.
Read pageAI MVPs
MVP development for AI startups, including product scope, model integration, UX, evaluation loops, and launch priorities.
Read pageGlossary
Riskiest assumption test (RAT) explained: finding the belief most likely to kill your product and testing it before building anything else.
Read pageNext step
Send your idea through the intake form and we'll turn it into scope, priorities, and next steps.
