What An MVP Actually Is
The smallest thing you can build that delivers genuine value to a real user and teaches you whether the rest is worth building.
Both halves matter. Something small nobody can use is an unfinished product. Something usable that answers no open question is just phase one. The word doing the work is viable: a thing somebody would use in preference to what they do now.
Why Most Things Called An MVP Are Not
Two different reasons produce the label, and they make very different products.
Genuine uncertainty. Nobody knows which parts of the idea matter, so building a small honest slice and putting it in front of real users is cheaper than guessing at the whole thing and being wrong at scale. This is the version that works.
Budget. The budget does not cover what was described, and MVP is the polite way of saying so. Not dishonest, and worth recognising, because a scope cut driven by budget produces a different shape from one driven by learning. Budget cuts remove depth from every feature. Learning cuts remove whole features and keep the survivors complete.
The second shape is almost always better and you frequently have to ask for it.
The Failure Mode To Avoid
Every feature at twenty percent, rather than twenty percent of the features at full quality.
A product with eleven half-working areas teaches you nothing, because when people do not adopt it you cannot tell whether the idea was wrong or the execution was thin. It also cannot be shown to a customer with any confidence, which removes the point.
The test when you are shown a proposed scope: would you be comfortable putting each item in front of a paying customer as it stands? Anything failing that should be cut rather than shrunk.
Prototype Or Foundation
Decide this before anything is built, because they are built differently and the difference is expensive to discover later.
A prototype answers a question and is then thrown away. Built fast, roughly, with no expectation of surviving.
An MVP is the first real version of something you intend to keep. Small, and with proper foundations.
The worst conversation in software is discovering after a successful launch that the thing everybody likes has to be rebuilt before it can grow. That happens when a prototype was funded and an MVP was assumed. Prototypes built quickly with AI tools raise the same question, and an AI Sign-Off has a developer read one and say whether it can carry real users or needs work first.
What Moves The Price
How clearly the question is defined. “We want to know whether accountants will pay for this” scopes tightly. “We want to test the market” does not.
Whether it needs to handle money. Payments, subscriptions and refunds bring compliance, failure paths and reconciliation. A genuine step up in scope.
Whether it needs accounts. Anonymous use is dramatically simpler than authentication, permissions and account recovery.
How real the data has to be. Seeded demo data is cheap. Importing a customer’s actual records is a project.
How many user types. One is an MVP. Three usually means three MVPs.
Where These Go Wrong
No definition of what would make you stop. Deciding in advance what result would change your mind is the difference between a test and an expensive act of faith.
Launching to nobody. An MVP with no user waiting is a smaller build, not a learning exercise. Line up the first users before it is finished.
Adding “just one thing” repeatedly. Each addition is reasonable; the fifteenth is what makes it late. This is scope creep in its purest form and MVPs attract it badly.
Measuring the wrong thing. Signups are easy to get and mean little. Whether anybody came back is the number.
Questions To Ask Whoever Builds It
- What question will this version answer? There should be one, stated plainly.
- Is this a prototype or a foundation? And is the price consistent with that answer.
- What are we deliberately not building? Ask for the list.
- What happens if it works? The path from MVP to real product should be described, not assumed.
- How quickly can we change something after launch? Learning is worthless if acting on it takes six weeks.
A Brief You Can Send Anyone
The idea, in two sentences: [what it does and for whom]
The question we need answered: [the actual uncertainty]
Who the first users are: [named, or a specific group, and how you reach them]
What they do today instead: [the incumbent, including “nothing”]
What would count as success: [a number and a date]
What would make us stop: [decide this now]
Whether it handles money: [yes / no / later]
Budget and deadline: [both, honestly]
How We Approach It
Weeks rather than quarters, with the smallest complete slice first and real users on it early. Built properly, on foundations that survive if it works, because the cost of doing that at MVP scale is small and the cost of discovering you did not is not.
If the honest answer is that you need a prototype and not an MVP, we will say so, and it will be cheaper.
Get in touch with the question you need answered and we will scope the smallest thing that answers it. If it grows into a product, SaaS development covers what comes next.