Types of MVP: The 9 MVP Variants Explained
There is no single MVP. The term covers a spectrum of nine experiment formats — from a landing page to a single-feature build — that trade build cost for evidence strength. Each one tests a different assumption: whether people want your product, whether you can deliver its value, or whether the thing even works.
Quick Answer: The main MVP types are the landing-page, smoke-test, fake-door, video, and pre-order MVPs (which test demand) plus the concierge, Wizard-of-Oz, piecemeal, and single-feature MVPs (which test whether you can deliver value). Pick the cheapest one that challenges your riskiest assumption.
Why MVPs Come in Types: Cost Versus Evidence
MVPs come in types because different questions demand different experiments. An MVP is not a smaller version of your product — it is a test. Eric Ries defines it in The Lean Startup as the version of a product that lets a team collect "the maximum amount of validated learning about customers with the least effort."
The word "least effort" is why types exist. If demand is your open question, a web page answers it faster and cheaper than a real build. If value delivery is the question, you need customers actually using something. Match the test to the assumption and you stop wasting months building the wrong evidence — which starts with understanding the underlying MVP definition.
Cheap and strong pull in opposite directions. A landing page costs little but only signals interest; a pre-order or a working feature costs more but proves intent or value. Testing Business Ideas by David Bland and Alex Osterwalder frames this as an experiment library — dozens of tests ranked by cost and evidence strength, run cheap-to-strong as confidence grows.
Demand-Signal MVPs: Do People Actually Want It?
Demand-signal MVPs test desirability — whether real demand exists — before you build anything shippable. They are the cheapest tier, and their currency is a click, a sign-up, or a pre-order rather than a working product.
Landing-Page MVP
A landing-page MVP is a single web page that pitches your value proposition and asks visitors to act, usually by entering an email. It tests demand and messaging: do people care enough to sign up? Build cost is low, and because no one has paid, treat sign-ups as a soft yes rather than proof of a business.
Smoke-Test MVP
A smoke-test MVP drives targeted traffic — often paid ads — to that page or offer and measures click-through and conversion before the product exists. Where a landing page tests your words, a smoke test buys a cold, unbiased audience to see whether strangers respond. Use it to compare demand across segments, channels, or positioning.
Fake-Door MVP
A fake-door MVP places a button or menu item for a feature that does not exist yet, then counts how many people click it. Also called a 404 test, it measures demand for a specific feature inside a live product. Handle it honestly with a "coming soon" message and, ideally, a way to capture emails or feedback.
Video MVP
A video MVP demonstrates the product in a short walkthrough to gauge interest when the concept is hard to prototype or explain in text. Dropbox famously used a demo video to build a waitlist before its syncing engine was finished. It tests whether the promised experience is compelling enough to make people want in.
Pre-Order and Crowdfunding MVP
A pre-order or crowdfunding MVP asks people to pay — or pledge — before the product ships. It is the strongest demand test on this list because money is a far harder commitment than an email. Reach for it when you need to prove willingness to pay, validate a price point, or fund production before you build.
Delivery MVPs: Can You Create the Value?
Delivery MVPs test whether you can actually create value for a user — by delivering it manually, faking the automation, stitching existing tools, or shipping only the core feature. They demand more effort than a demand test but return far stronger evidence, because real people are getting a real outcome.
Concierge MVP
A concierge MVP delivers the service manually, by hand, to a small set of real customers with no automation behind the scenes. Eric Ries describes this pattern in The Lean Startup: you personally do what the software eventually will, so you learn the workflow and whether it creates value. It tests value delivery, not just interest — see our deeper guide to the concierge MVP for how to run one.
Wizard-of-Oz MVP
A Wizard-of-Oz MVP puts a fully automated-looking front end in front of users while humans do the work behind the curtain. Customers believe they are using finished software; in reality you fulfill each request by hand. It tests whether people want the automated experience — and how it should behave — before you invest in building the automation itself.
Piecemeal MVP
A piecemeal MVP stitches together existing tools — form builders, spreadsheets, payment links, automation apps — to deliver a working product without custom code. It tests the end-to-end solution quickly and cheaply, and often doubles as your real operating stack for the first customers. Reach for it when the parts already exist and integration is the only open question.
Single-Feature MVP
A single-feature MVP is a real, working product stripped to the one core feature that delivers your primary value, and nothing else. Unlike the faked or manual approaches above, users get a genuine — if narrow — product. It tests whether that core value resonates in daily use, and is the natural next step once cheaper experiments have confirmed demand.
MVP Types Compared: What Each One Tests
The table below maps each MVP type to the assumption it tests and the situation it fits. Read it as a menu, not a ranking.
| MVP type | What it primarily tests | When to use it |
|---|---|---|
| Landing-page | Demand and messaging | You need a fast read on interest and positioning |
| Smoke-test | Demand from a cold audience | You want to compare segments, channels, or offers |
| Fake-door | Demand for a specific feature | You are deciding what to build next |
| Video | Interest in a hard-to-prototype concept | The idea is complex or expensive to demo live |
| Pre-order / crowdfunding | Willingness to pay | You must prove purchase intent or fund production |
| Concierge | Value delivery and workflow | You are still learning what customers actually need |
| Wizard-of-Oz | Desire for the automated experience | Automation is costly and you want proof first |
| Piecemeal | Feasibility of the whole solution | The building blocks already exist as tools |
| Single-feature | Core value in a real product | Cheaper tests already confirmed demand |
Takeaway: No type is "best." The right MVP is the cheapest experiment that produces trustworthy evidence about your single riskiest assumption — which is why founders often run two or three of these in sequence rather than betting on one.
How to Pick the Right MVP Type for Your Risk
Pick the cheapest MVP that produces trustworthy evidence about your riskiest assumption. Name that assumption first, then match it to a test:
- If demand is the risk — start with a landing-page or smoke-test MVP, then escalate to pre-orders when you need proof of payment.
- If value delivery is the risk — run a concierge or Wizard-of-Oz MVP so you learn the workflow before you automate it.
- If feasibility is the risk — try a piecemeal MVP, then graduate to a single-feature build.
Sequence experiments; do not pick just one. Bland and Osterwalder advise pairing cheap, weak-signal tests early and progressing to stronger, costlier ones as evidence accumulates. At Edmired we treat MVP type as a decision you revisit at each stage, not once — it is a single step inside a broader startup idea validation process.
Key Takeaways
- An MVP is a learning tool, not a smaller product. Ries defines it as the version that earns the most validated learning for the least effort, so the type you choose follows from what you need to learn.
- The nine types split into two jobs. Demand-signal MVPs (landing-page, smoke-test, fake-door, video, pre-order) test whether people want it; delivery MVPs (concierge, Wizard-of-Oz, piecemeal, single-feature) test whether you can create the value.
- Cost and evidence strength trade off. Landing pages give fast but weak signals; pre-orders and real builds cost more but prove intent and value.
- Money is the strongest demand signal. A pre-order or paid pilot beats a pile of email sign-ups because payment is a far harder commitment.
- Manual beats automated while you are still learning. Concierge and Wizard-of-Oz MVPs let you deliver value by hand before investing in software.
- Match the MVP to your riskiest assumption. Name the one belief that would sink the idea if wrong, then pick the cheapest test that challenges it.
- Run experiments in sequence. Start cheap and weak, then escalate to costlier, stronger tests as confidence grows.
Frequently Asked Questions
What is the cheapest type of MVP to build?
The landing-page MVP is usually the cheapest, followed by the fake-door and smoke-test variants. All three test demand with little more than a web page, some copy, and optional ad spend — no product required. They give fast, low-cost signals, but because no one has paid, treat the results as interest rather than proof of a viable business.
Which MVP type gives the strongest evidence?
Pre-order and single-feature MVPs give the strongest evidence, because they involve real money or a real working product. A completed purchase proves willingness to pay, and a shipped feature shows whether the core value holds up in daily use. The trade-off is cost and time: strong evidence takes far more effort than a landing page does.
Can you combine multiple MVP types?
Yes, and most founders do. A common sequence is a landing-page or smoke-test MVP to confirm demand, then a concierge or Wizard-of-Oz MVP to deliver value manually, then a single-feature build. Each stage de-risks a different assumption, so combining types over time is usually smarter than betting everything on one large experiment.
Is an MVP the same as a prototype?
No. A prototype is a model used to explore and refine a design, often before any customer sees it. An MVP is a real experiment released to actual users to test a business assumption and collect validated learning. A prototype answers "can we design this?"; an MVP answers "do people want and value this?"