Types of MVP for Startups: Which One to Build First
An MVP is the smallest experiment that tests your riskiest assumption with real users — not a half-built product. Choose the MVP type that matches what you need to learn: demand (landing page), willingness to pay for a service (concierge), whether a workflow feels magical before automation (Wizard of Oz), core product value (single-feature), or whether assembling existing tools works (piecemaker). Pick one learning goal per experiment.
Updated August 8, 2026.
Landing page MVP
A landing page describes the offer and asks for a concrete action: email, waitlist with qualification questions, booked call, or pre-order. It tests messaging and demand before you build. Pair it with traffic from a channel your ICP already uses, or with direct outreach, or page views alone will mislead you.
- Use when: you need signal on whether the problem/offer resonates before engineering.
- Strong signal: qualified signups, booked calls, or deposits — not bounce rate alone.
- Weak alone: vanity traffic without conversations or commitment.
Concierge MVP
You deliver the outcome manually as a high-touch service. Customers pay (or commit) for the result; you learn the workflow before you automate it. Classic for B2B processes where software would eventually replace founder labor.
- Use when: the value is an outcome, and you do not yet know every step to automate.
- Strong signal: paid engagements and clear steps you repeat for every customer.
- Watch for: unreplicable heroics that will not become product.
Wizard of Oz MVP
The product looks automated to the user, but humans perform critical steps behind the scenes. Useful when UX and perceived magic matter, and you need to validate the experience before investing in ML or complex backends.
- Use when: the front-end experience is the risk, not whether humans can do the work.
- Strong signal: users complete the flow and ask to continue without knowing it was manual.
- Ethics: be ready to disclose when asked; do not fake regulated or safety-critical automation.
Single-feature MVP
Ship one narrow workflow end-to-end that delivers the core value proposition. Cut adjacent features ruthlessly. Best when you already know the problem and need retention and usage signal on the primary job-to-be-done.
- Use when: discovery showed a sharp problem and you need product proof, not more interviews.
- Strong signal: repeat usage or retention on that one job.
- Avoid: a “thin” product that is actually five half-features.
Piecemaker (assembled) MVP
Also called a “piecemeal” or no-code stack MVP: stitch existing tools (forms, sheets, Zapier/Make, Stripe, Airtable, chat) into a working offer. Speed matters more than elegance. Ideal when distribution and offer testing beat custom engineering.
- Use when: you can deliver value by connecting tools your team already knows.
- Strong signal: customers complete the journey and pay despite rough edges.
- Plan the rewrite only after the offer and ICP are proven.
Put this into practice on Bowora
After an MVP produces real users or revenue, a clear Bowora profile helps later customers understand what you do — once the offer is stable enough to describe accurately.
Common questions
- What are the main types of MVP for startups?
- Common types include landing page, concierge, Wizard of Oz, single-feature product, and piecemaker (assembled tools) MVPs. Choose based on the assumption you need to test — demand, willingness to pay, UX, core value, or speed of delivery.
- Is a landing page a real MVP?
- Yes, if it tests a specific demand assumption with a real call to action. It is not enough alone to prove product-market fit, but it is a valid early experiment when paired with outreach or targeted traffic and follow-up conversations.
- Concierge vs Wizard of Oz — what is the difference?
- In a concierge MVP, the customer usually knows they are buying a high-touch service. In a Wizard of Oz MVP, the experience appears productized/automated while humans operate behind the scenes. Use concierge to learn the workflow; use Wizard of Oz to test the product feeling.
- When should I stop using an MVP and build the “real” product?
- When the learning goal is met: repeated willingness to pay, clear ICP, and a workflow you can specify for engineering. Do not rebuild for aesthetics while the offer and retention signals are still unclear.
Sources
Facts, frameworks, and program details were checked against these first-party references. Last content review: August 8, 2026.