MVP is the acronym you hear in every conversation about startups, and yet it hides more misunderstanding than almost anything else. Some people picture an unfinished product, some a pitch deck for investors, some a complete application with ten features. In this article I explain what an MVP really means, what belongs in it and what doesn't, what it costs and how to build one even if you don't have a team yet.
Quick answer: what is an MVP
An MVP (minimum viable product) is the smallest version of a product that someone will actually use or pay for. It is not a demo prototype and not half of the final application. It is a sharp cut: one core process that works end to end, sign-in, basic admin and measurement of what users do.
| What belongs in an MVP | What does not |
|---|---|
| one key process (say "I submit a request, I get a quote") | a second and third process "just in case" |
| sign-in and basic roles | advanced permissions and team accounts |
| simple admin | reports, exports, dashboards |
| measurement (what people click, where they leave) | a mobile app next to the web app |
| a landing page that explains the product | integrations with systems nobody uses yet |
The purpose of an MVP is single: find out as cheaply as possible whether anyone wants the product. Everything that does not help with that can wait.
How much does an MVP cost
The price of an MVP is driven by scope, not technology. What matters is the number of screens, the number of roles and the number of connections to external systems.
| Scope | Price | When it makes sense |
|---|---|---|
| Fixed-price MVP | from €3,200 | a clear idea, one process, you need to be live within two months |
| Ongoing development | €40/h | you have your own product owner and the scope changes every week |
| Production version | from €7,200 | the MVP proved demand, you are adding roles, payments, integrations |
The prices match my pricing on the MVP for startups page. A more detailed breakdown by feature is in How much does custom app development cost.
For comparison: a team of a designer, two developers and someone for operations costs several times more per month even part-time, and that is before they write the first line. Most startups can't afford that at the MVP stage and don't need it.
How an MVP is built: five steps
1. One sentence
Write what the first version should do and for whom, in one sentence. "A fitness coach sends a client a training plan and sees whether it was completed." If the sentence needs the words "and also", cut.
2. Brief and wireframe
The sentence becomes a list of screens and one main flow. A wireframe is a black-and-white skeleton that verifies the flow makes sense before anything is programmed. This is where the price is decided: every extra screen is days of work.
3. Build
A proven stack anyone can take over later: Next.js and React on the frontend, PostgreSQL via Supabase as the database with sign-in, Stripe for payments, cloud deployment. No experiments a new developer would have to rewrite.
4. Security from day one
The most common startup mistake is "we'll finish security once we have users". But the first users arrive on day one. The minimum that belongs in every MVP: verified sign-in, row-level data permissions, limits against form abuse, database backups, keys outside the repository and GDPR (account deletion, data processing).
5. Launch and measurement
An MVP without measurement is just an app. From the first day you need to know how many people signed up, how many completed the main process and where they leave. Only with those numbers does it make sense to decide what to build next.
How long does an MVP take
With a clear brief, 4 to 8 weeks from workshop to launch. Longer timelines usually don't mean more care, but an unclear brief or too many features. When an MVP takes more than a quarter, it usually isn't an MVP anymore.
Does a startup need a team for an MVP?
No. For an MVP scope one person who can handle design, development, security and deployment is enough, and it is an advantage: no coordination of four people, one person responsible, one invoice. A team makes sense once the MVP has proven demand and the product grows faster than one person can keep up with. Until then a team is the most expensive way to find out nobody wants the product.
That is exactly how I build MVPs for startups: the whole process from brief to operations, at a fixed price and without equity. The code, accounts and domain are yours from the start.
The most common MVP mistakes
- Too many features. Every extra feature postpones the answer to whether anyone wants the product.
- No measurement. Without numbers you decide by feeling, and a founder's feeling is always optimistic.
- Postponed security. A data leak in the first month is something a startup does not survive.
- Code nobody else can take over. An unusual stack or a repository not in your name gets expensive when the team arrives.
- Building without a first customer. The best MVPs are built for a specific person who is waiting for them.
Summary
An MVP is the smallest version of a product someone will use or pay for: one process, sign-in, admin, measurement. It costs from €3,200, takes 4 to 8 weeks and does not need a team. The most important decision is not made during the build but in the brief: what to cut.
If you have an idea and want an estimate of scope and price, describe it to me in three sentences. You will get an answer within 24 hours.