Prototype vs MVP vs pilot: which do you need first?

Hephaestus Technology7 min read

In the prototype vs MVP vs pilot choice, start with whichever answers your riskiest open question most cheaply. A prototype tests whether something can be built or how it should work, usually with a few users in a controlled setting; an MVP tests whether real customers will use or pay for it; a pilot tests whether it delivers a measurable result inside a specific customer's business. Most software founders need a prototype or MVP first, while B2B and hardware founders often reach a pilot sooner than they expect.

Key takeaways

  • A prototype answers 'can it work and how should it work?', an MVP answers 'will customers use or pay?', a pilot answers 'does it deliver results in this customer's business?'.
  • Pick the first step by your riskiest assumption: technical, usability, demand or operational value.
  • Jakob Nielsen's research suggests testing a prototype with about five similar users finds most usability problems.
  • An MVP is defined by validated learning, not by being a smaller version of the final product.
  • Irish supports reflect the sequence: LEO feasibility grants can cover prototype development, and the Pre-Seed Start Fund ideally looks for an MVP or live beta.

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

They differ in the question they answer, who sees them and what counts as success.

  • A prototype is a model of the product, from paper sketches and clickable mock-ups to a working technical proof of concept, used to test feasibility or usability. It is often not usable on real work.
  • An MVP (minimum viable product) is the smallest version real customers can use, built to learn whether they want it. Eric Ries defines it as the version "which allows a team to collect the maximum amount of validated learning about customers with the least effort."
  • A pilot is a time-limited deployment with a specific customer, with agreed success criteria, designed to prove value in their operations and lead to a contract.

How do they compare side by side?

Prototype MVP Pilot
Main question Can it work? How should it work? Will customers use or pay for it? Does it deliver a measurable result for this customer?
Who uses it Team, testers, a few target users Early adopters in the target market One or a few named customers
Typical form Sketches, mock-ups, proof of concept, test rig Landing page, manual service, no-code app or basic software MVP or near-product, often with manual support
Success signal Feasibility shown; usability issues found and fixed Usage, sign-ups, retention, payments Agreed success criteria met; contract signed
Typical cost Lowest to moderate Low to moderate Moderate, partly funded by the customer if paid
Main risk addressed Technical and usability risk Demand risk Operational and commercial risk

When should you build a prototype first?

Build a prototype first when you are not sure the product can be built, or not sure how users will interact with it. Typical cases:

  • Technical uncertainty: a new algorithm, an AI feature with unproven accuracy, or electronics that must hit size, power or cost targets.
  • Usability uncertainty: a workflow users must learn quickly, or an interface with many steps.

Prototype testing does not need large numbers. Jakob Nielsen of Nielsen Norman Group argues that testing with about five users from a similar group finds around 85% of usability problems, and recommends several small rounds with redesigns in between rather than one large study. For hardware, a sequence of prototypes is normal; our guide on how to build an IoT product covers those stages.

In Ireland, early prototype costs may be partly fundable: the Limerick Local Enterprise Office feasibility grant, for example, lists prototype development among eligible costs, up to 50% of the investment to a maximum of €15,000 excluding VAT, for businesses with fewer than 10 employees. Check your own county's LEO for current terms.

When should you build an MVP first?

Build an MVP first when the product is clearly feasible and the real question is whether anyone wants it. For many software ideas, feasibility is not in doubt; demand is. In that case a prototype can be a costly detour.

Steve Blank stresses that an MVP is "not always a smaller/cheaper version of your final product" and urges founders to "think about cheap hacks to test the goal". Buffer is a well-documented example: its founder launched a two-page site and then a pricing page to see whether people would sign up and click on paid plans, and built the product only after they did. Our explainer on what an MVP is covers MVP types and scoping.

When should you go straight to a pilot?

Go to a pilot when you sell to businesses, the value only shows up in the customer's real operations, and you have at least one prospect who has confirmed the problem. B2B buyers often cannot judge value from a demo; they need to see it on their own data, processes and people.

A pilot does not need a polished product. Many pilots combine a basic MVP with manual work behind the scenes, as long as you can reliably deliver the agreed outcome. What a pilot does need is written success criteria, a fixed scope and duration, and a pre-agreed next step if it succeeds. Our worked example of a paying B2B pilot shows how to structure one.

How do you decide which comes first?

Decide by naming your riskiest assumption and choosing the cheapest test that addresses it:

  1. "We don't know if it can be built." Prototype (technical proof of concept).
  2. "We don't know how users will use it." Prototype (clickable mock-up, tested with about five users per round).
  3. "We don't know if anyone wants it." MVP (landing page, concierge service or simple product).
  4. "We know buyers want it, but not whether it delivers value in their operations." Pilot.

In Strategyzer's terms, prototypes and early MVP tests help establish problem-solution fit, evidence that "customers care about certain jobs, pains, and gains", while MVPs in real use and pilots build evidence towards product-market fit.

Do Irish funders expect a prototype, MVP or pilot?

It depends on the programme. Local Enterprise Office feasibility support, as in the Limerick example above, is aimed at early research and prototyping. For Enterprise Ireland's Pre-Seed Start Fund, a Donegal County Council notice on the fund says "ideally the start-up would have a Minimal Viable Product (MVP) or be live in Beta," together with a validated customer proposition. A paying pilot customer is strong evidence for almost any funder, because it shows a business has committed money.

What are the common mistakes?

  • Calling a prototype an MVP. If customers cannot use it on real work, it cannot test demand.
  • Polishing a prototype for months. A prototype's job is to answer a question, then be thrown away or rebuilt.
  • Running free, open-ended pilots. Without success criteria and a next step, pilots drift and rarely convert.
  • Skipping straight to full product. Each step exists to stop you spending heavily on the wrong thing.

Frequently asked questions

Is a proof of concept the same as a prototype?

A proof of concept is a type of prototype focused narrowly on technical feasibility: can this core technology work at all? Other prototypes focus on usability or appearance. Neither is usually ready for customers to use on real work.

Can an MVP and a pilot be the same thing?

Sometimes, in B2B. If your MVP is deployed with a named customer under agreed success criteria and a defined duration, it is functioning as a pilot. The difference is the commercial framing: a pilot is designed to prove value to that customer and lead to a contract.

How many users should test a prototype?

Jakob Nielsen's guidance is about five users from a similar group per round, which he estimates finds around 85% of usability problems. If you have very different user groups, test a few people from each.

Do hardware startups need an MVP?

Yes, but it often looks different. Hardware founders usually build several prototypes first to prove feasibility, then use a small batch of units, or a manual service that mimics the device, as an MVP or pilot with early customers.

Should I skip the prototype if I'm short of money?

If feasibility and usability are not in doubt, yes: go straight to the cheapest MVP that tests demand. If either is uncertain, a small prototype is usually cheaper than discovering the problem after building an MVP.

Not sure which step your idea needs next? Book a free 30-minute call and we will talk it through.

Cover photo by ready made on Pexels.

Hephaestus Technology

We help founders and businesses validate, build and launch products across software, AI and hardware, from offices in Ho Chi Minh City, Dublin and Delaware. See how we work.

Tell us what
you're building.

A free 30-minute call. You'll get an honest view of the idea, the risks we see and what we'd do first, whether or not we end up working together.