PoC vs Prototype vs MVP: What's the Difference, and Which Do You Need?

A PoC, a prototype and an MVP retire different risks. A proof of concept answers “can this be built?” (technical risk, 1–3 weeks, usually no UI). A prototype answers “is this the right shape?” (usability and investor risk, 1–3 weeks, clickable but not real). An MVP answers “will people use and pay for it?” (market risk, 6–12 weeks, real software in production).

Those numbers are how Kinetico, a product design and engineering studio, scopes them. Most founders don't need all three; some need only the last one. The mistake is picking by how impressive the artifact looks rather than by which unknown is actually threatening the business. This guide is the decision procedure, with the numbers attached.

What is a proof of concept (PoC)?

A proof of concept is the smallest thing you can build to find out whether a technical approach works at all. It usually has no user interface — a script that calls the payment rail, a notebook that runs the model on your data, a spike that renders the thing at the frame rate you need. It is seen by your team and maybe a technical advisor. It is disposable by design: you keep the answer, not the code.

When you need one: there is a real, specific unknown — an integration that may not expose what you need, a data source that may not be clean enough, a model that may not be accurate enough, a browser capability that may not be fast enough. When you don't: the stack is standard (auth, CRUD, payments via Stripe, a dashboard). Building a PoC for something a thousand teams have already built is procrastination with a technical vocabulary.

What is a prototype?

A prototype is a simulation of the product: clickable screens, real copy, real flows, and nothing behind them. It exists to be shown to people — five to ten target users, an investor, a co-founder — to find out whether the shape of the thing makes sense before anyone writes production code. A good prototype takes one to three weeks of design work and answers questions like “do they understand what this does?”, “where do they get stuck?”, “would they click pay?”.

The trap: a prototype is only cheap if it survives into the build. When a design agency prototypes and a separate dev shop builds, the screens get re-specified, re-interpreted and re-drawn — you pay for the same thinking twice. When the people who drew the prototype write the code, the prototype simply becomes weeks one and two of the MVP. That is the whole case for one team from first screen to production, and it is why we scope prototype and MVP as one engagement.

What is a minimum viable product (MVP)?

An MVP is the product — the first real version, deployed, with real accounts and real data, that real users can sign into and (ideally) pay for. It contains one core workflow done properly, and as little else as possible. It is not a prototype with a database bolted on, and it is not a rough draft; the code is the code you will keep. In Kinetico's scoping an MVP is 6–12 weeks and $15,000–$50,000 depending on roles, screens and integrations — the full cost breakdown is here.

What it retires: market risk — will anyone use it, will they come back, will they pay, can you operate it. None of those can be answered by a PoC or a prototype, because none of them put real software in front of real people for weeks.

PoC vs prototype vs MVP: side by side

Proof of concept vs prototype: which comes first?

The PoC comes first only if there is a technical unknown. The two answer unrelated questions — a PoC asks whether the engine can be built, a prototype asks whether anyone wants to drive the car — so building a beautiful prototype of something that turns out to be infeasible is the expensive order of operations. If feasibility is obvious, skip the PoC entirely and prototype.

MVP vs prototype: when is a prototype enough?

A prototype is enough when the decision you need to make is a design decision or a fundraising decision. It is never enough when the decision is a market decision. People say yes to prototypes; they churn from MVPs. If your next milestone is “do users come back in week two” or “does anyone pay”, only an MVP answers it — and the prototype should be treated as the first two weeks of that MVP, not as a separate deliverable.

MVP vs PoC: do you need both?

PoC vs MVP is the pairing founders ask about most, and the answer is: you need both when there is a genuine technical unknown and a market unknown — in that order, and only that order. A PoC that proves the integration works costs one to three weeks and can save you from spending $30k on an MVP that cannot be built. A PoC for a standard stack costs the same one to three weeks and saves you nothing. Ask one question: “if this fails technically, would it surprise a senior engineer?” If not, go straight to the MVP.

MVP vs pilot: what is the difference?

A pilot is not a fourth artifact. It is a way of deploying a product — usually an MVP — with one customer or a limited group, under agreed conditions and success criteria, before a wider rollout. In B2B the healthy sequence is prototype → MVP → pilot with a design partner → general availability. Founders who call their MVP a “pilot” are usually describing the same software with a more comfortable name for the customer.

Which one do you need? Pick by the risk

Write down the sentence that starts “I don't know if…” and is keeping you up at night. That sentence tells you the artifact:

Most funded founders we talk to are in row three: standard technology, standard flows, unproven demand. For them, the PoC and the standalone prototype are delay. The right move is a Lean MVP — one role, one core workflow, payments switched on — which is the smallest thing in our MVP development service.

What does this look like on real products?

Three from our own work, one per artifact:

Frequently asked questions

What is the difference between a PoC, a prototype and an MVP?

They retire different risks. A proof of concept (PoC) answers “can this be built?” — technical feasibility, usually no UI, 1–3 weeks. A prototype answers “is this the right shape?” — usability and investor buy-in, clickable but not functional, 1–3 weeks. An MVP answers “will people use and pay for it?” — real, deployed software with one core workflow, 6–12 weeks.

What is the difference between a proof of concept and a prototype?

A proof of concept tests whether something can work technically — an API integration, a model, a rendering approach — and is often a script or backend experiment with no interface. A prototype tests whether the product makes sense to a person — flows, screens, and interactions — and usually has no working backend at all. PoC = feasibility; prototype = form.

What is the difference between an MVP and a prototype?

A prototype is a simulation: it looks like the product but nothing behind it is real, and it is shown to people. An MVP is the product: real accounts, real data, deployed, and used by people. You can test a prototype in an afternoon; an MVP has to survive weeks of real usage.

What is the difference between a PoC and an MVP?

A PoC reduces technical risk (can it be built?); an MVP reduces market risk (does anyone want it?). A PoC is disposable and internal; an MVP is the first version of the real product. Many products need both, in that order — but only if there is a genuine technical unknown.

How much do a PoC, a prototype and an MVP cost?

In Kinetico's scoping: a PoC is 1–3 weeks of engineering time; a clickable prototype is 1–3 weeks of design; an MVP is $15,000–$50,000 over 6–12 weeks depending on scope. Prototype and MVP are usually one engagement — the prototype is weeks one to two of the build.

What is the difference between an MVP and a pilot?

An MVP is a product; a pilot is a deployment. A pilot runs a product (often an MVP) with one customer or a limited group under agreed conditions to test operations, adoption and value before a wider rollout. In B2B the sequence is often prototype → MVP → pilot with a design partner → general release.