A minimum viable product (MVP) is the simplest version of a product that can be released to real users in order to learn whether the core idea solves a problem they care about and will pay for. Start-ups, corporate innovation teams and investors use the concept to limit how much time and money is spent before there is evidence of demand.
The term is associated with the "lean start-up" movement. It was used by the consultant Frank Robinson around 2001 and popularised by the entrepreneur Steve Blank and the author Eric Ries, whose book The Lean Startup (2011) built a method around a build, measure and learn loop. An MVP is often confused with a prototype or a beta. A prototype tests whether something can be built; an MVP tests whether anyone wants it. In regulated sectors such as payments, the "viable" part also has to include what the law requires before the first customer is served.
Minimum viable product in plain terms
Suppose you think your neighbourhood wants fresh bread delivered at 7 a.m. Before buying vans and ovens, you bake twenty loaves at home, post a sign-up sheet in the local café, and deliver by bicycle for two weeks. If people sign up and keep paying, you have learned something real. If they do not, you lost very little.
The bicycle round is the MVP. It is not the final business, and it is not a toy either: real customers pay real money for a real loaf. The point is to learn quickly and cheaply whether the idea deserves the vans and ovens.
How an MVP works
The method starts with a hypothesis, for example "small importers will pay a flat fee to see live exchange rates and book transfers online." The team strips the idea down to the one feature that tests it, releases that feature to a small group, and measures behaviour: sign-ups, repeat use, payment, and reasons for leaving. Results feed the next version, or lead the team to change direction (a "pivot") or stop.
Common forms include:
- Landing-page tests. A page describes the service and measures how many visitors join a waiting list before anything is built.
- Concierge MVPs. Staff deliver the service by hand, for example arranging transfers manually, while the customer sees a simple interface.
- Single-feature products. Only the core action works, such as sending money along one corridor.
- Rapid builds. Teams increasingly assemble early versions with no-code tools or vibe coding, accepting that much of the code will be rewritten.
The measure of success is learning. Investors who track deal flow often ask for MVP data, such as retention and customer acquisition cost (CAC), before funding a company, because it shows whether product market fit is in sight.
MVPs in payments and financial services
Financial products complicate the "minimum." A fintech that holds or moves customer money usually needs a licence, such as a money transmitter license (MTL) in the United States or an electronic money institution (EMI) authorisation in Europe, before going live. The minimum therefore includes anti-money laundering (AML) controls, KYC checks, complaint handling and secure data storage, even if the customer interface is basic.
Teams use several routes to keep the first version small while staying legal:
- Partnering with a licensed firm. A start-up can operate under a banking-as-a-service (BaaS) provider or as an agent of a licensed institution.
- Testing in a sandbox. A regulatory sandbox lets some firms trial a service with limited customers under supervision.
- Using restricted licences. Some jurisdictions offer lighter regimes, such as a small payment institution registration or a start-up MVP licence, with caps on volume.
Pricing is another area to test early. A provider might trial a flat fee against tiered pricing to see which one wins and keeps customers.
Criticism and limitations
Critics say the term is stretched to excuse unfinished or unsafe products. A poorly built MVP can give a false negative: customers reject it because it is clumsy, not because the idea is wrong. In finance the cost of cutting corners is higher, since weak controls can cause a data breach, losses to customers or enforcement action. Teams are generally advised to keep the scope small but to apply security hardening and compliance controls in full from day one, and to plan for the MVP's code and processes to be rebuilt once the idea is proven. A written business plan still matters, because banks and regulators assess the whole operation, not just the product.