An MVP (minimum viable product) is the simplest working version of a product that a real customer can pay for, subscribe to or come back to.
Unlike a prototype or a proof of concept, only an MVP tests the key question — will the market pay?
For an investor it is proof of demand: first payments, conversion, retention and a first estimate of customer acquisition cost instead of hypotheses on slides.
Investors often open a meeting with “Who has already paid for your product?” — and most early pitches fall apart right there, because demand has not been validated.
An MVP solves this: within a few weeks the team ships the simplest working version, secures first payments or pre-orders and backs the story with real numbers.
Below: how an MVP differs from a prototype and a PoC, the types of MVP, which investor risks it reduces, what data to collect, how to build one step by step, examples and a checklist.
MVP vs prototype vs PoC
| Format | What it is, in plain words | Why it exists | What the investor sees |
|---|---|---|---|
| Prototype | A demonstration model of the interface with nothing behind it | Show the idea to the team quickly and get internal feedback | “I see how it might look, but not whether anyone will pay” |
| PoC (proof of concept) | A technical experiment proving the core technology works at all | Remove the “impossible to build” risk | “The technology works, but market and money are out of the picture” |
| MVP (minimum viable product) | The smallest working product through which a real customer can pay, subscribe or return | Test demand, collect first revenue and metrics | “People are paying; I can see conversion and early economics” |
Why does an investor care about the MVP specifically? It answers two questions at once: will the market pay, and can the team deliver value quickly and cheaply.
A prototype answers the design question, a PoC the technology question — only an MVP confirms that customers will part with money.
An MVP is not a cut-down version of the future product but an experiment. “Minimum” refers to the feature set; “viable” means it solves the customer’s problem well enough that they pay for it.
Types of MVP
An MVP does not have to involve code. Often the fastest way to test demand is to write none at all.
| Type | How it works | When it fits |
|---|---|---|
| Landing page with pre-orders (smoke test) | A page describing the product with a payment or sign-up button before the product exists | Testing interest and willingness to pay before development |
| Video demo | A video showing how the product will work | Complex technical products that take long to build |
| Concierge | The team delivers the service by hand, and the customer knows it | Services and B2B: understanding the customer’s process before automating it |
| Wizard of Oz | The product looks automated to the customer, but people do the work behind the scenes | Testing demand for automation without paying for it |
| Single-feature product | One key feature that solves the main problem | Software where the value sits in one action |
| No-code and off-the-shelf tools | A product assembled from builders, spreadsheets, bots and payment forms | A fast launch without a development team |
Well-known examples: before building the product, Dropbox released a demo video and saw from the response that demand existed; Zappos’ founder photographed shoes in stores, posted them online and only bought them once ordered; Airbnb’s founders rented out air mattresses in their own flat.
None of them started with a “real” product — they had a way to test whether people would pay.
Four investor risks an MVP reduces
- Demand risk. Interviews and first sales through the MVP show that people have a pain and will pay.
- Solution risk. Retention on day 7 and day 30 shows the product is not only bought but used again.
- Channel risk. Test campaigns give a cost per click and a first estimate of customer acquisition cost (CAC).
- Economics risk. A rough payback period and LTV show potential profitability.
When the MVP turns these risks into numbers, the investor sees less uncertainty and is ready to talk about funding. How to calculate these metrics: Startup unit economics.
The data investors want to see
Quantitative signals:
- Conversion — the share of visitors who take the target action: sign-up, payment.
- D7 / D30 retention — the share of users who return after 7 and 30 days.
- Proxy CAC — a first estimate from cost per click and the funnel: impressions → clicks → payments.
- Average order value or average revenue per paying user (ARPPU).
- Proof of willingness to pay — as a guide, 10–20 real payments in B2C or signed letters of intent (LOIs) in B2B.
Qualitative signals: 10–15 problem–solution interviews confirming the pain and willingness to pay.
Artefacts: analytics screenshots, a hypothesis table with sprint results, a short demo video.
Together, this is early traction. How to present it at different stages: What is startup traction.
How to build an MVP in six steps
- Value hypothesis. State a specific problem for a specific segment. If you have not chosen a segment yet, start by testing your niche.
- Success criterion. One metric and a minimum threshold, e.g. “5 paying customers in 30 days”.
- Minimum feature set. Only what is needed to reach the threshold; everything else goes on the later list.
- Test channel. One traffic source with a measurable funnel; fix the budget and expected conversion.
- Unit economics check. Estimate CAC and payback; payback over 12 months is a signal to optimise.
- Iteration plan. 2–3 “launch → metrics → change” cycles with a hypothesis / metric / threshold / result table.
Timing depends on the product: a pre-order landing page takes days, a software MVP usually a few weeks. If your MVP is taking six months, it is probably no longer minimal.
Five MVP examples across industries
These examples are illustrative — they show how an MVP works and what numbers it produces, not the results of specific companies.
| Industry | What was built | Channel and result | Artefacts |
|---|---|---|---|
| B2B SaaS (marketing analytics) | A Google Sheets add-on calculating 3 metrics | A post in a niche Slack (~1,300 members), €90 → 3 paying customers; CAC ≈ €30, payback under a month | Dashboard screenshots, hypothesis table |
| Digital health | A no-code chatbot with SMS medication reminders | Medical communities on Facebook; in 2 weeks 50 active users, 12 pre-payments at $5; CAC $4, D30 40% | Demo video, analytics export |
| Fintech (micro-investing) | A widget with a single API integration into a mobile bank, built at a hackathon | Demo day → 2 LOIs and 300 waitlist sign-ups | API spec, LOIs, LTV and proxy CAC estimate |
| Used electronics marketplace | A landing page and Google Form, listings posted by hand | $350 of TikTok ads → 200 listings and $600 in commission; CAC ≈ $1.75 per listing, spend recovered immediately | Google Sheet, process video |
| EdTech subscription (language bot) | A GPT-based Telegram bot with content in Notion | A guest post reaching ~7,000 → 70 subscribers at $8; CAC $6, D30 32% | Stripe screenshot, active sessions chart |
Common MVP mistakes
- A vague hypothesis and five “key” metrics at once — it is unclear what is being tested.
- A “minimal” product nobody can buy — that is an extended prototype, not an MVP.
- Five channels at once — traffic gets mixed, distorting CAC and conversion.
- No success or failure threshold — any number can be called “encouraging”.
- Premature polish — a logo and brand guide instead of the feature people pay for.
- Testing on friends — friends and colleagues praise the product but do not pay; their feedback does not prove demand.
Keep the MVP’s goal in focus — showing investors a paying market and disciplined experimentation — and the deal moves faster.
MVPs and investors in Spain
A working MVP also helps with Spanish instruments.
ENISA assesses a project’s innovation and scalability, and the financial forecasts in a startup visa report are checked for realism — first payments and MVP metrics turn the forecast from a hypothesis into a calculation based on facts.
How it works: Business plan and financial model for Spain’s startup visa.
Business angels also generally expect an MVP and first customers rather than a bare idea — see How to prepare a project for investment.
MVP readiness checklist before talking to investors
- Segment and pain are stated in one sentence, with no abstractions.
- A success threshold (e.g. 10 sales in 30 days) and a deadline were set before launch.
- Conversion to payment and, where relevant, D7 / D30 retention are calculated and recorded.
- One channel was chosen with a transparent proxy CAC; spend and clicks are backed by screenshots.
- Every conclusion has artefacts: analytics screenshots, demo video, hypothesis table with results.
- The next two iterations are defined: what changes, which metric should move and to what level.
A 10-point self-assessment and a hypothesis template: MVP checklist for investors.
Frequently asked questions
What is an MVP in simple terms?
The simplest working version of a product that a real customer can pay for, subscribe to or come back to. Its job is to test demand at minimal cost, not to show a finished product.
How is an MVP different from a prototype?
A prototype shows what the product will look like, but nobody can use it or pay for it. An MVP works: customers solve their problem with it and pay, which is why only an MVP tests demand.
Can you build an MVP without code?
Yes. A pre-order landing page, a demo video, a concierge service, manual work behind an automated front or a product assembled from no-code tools often test demand faster and cheaper than development.
Which MVP metrics matter to investors?
Conversion to payment, day-7 and day-30 retention, proxy CAC, average order value and the number of real payments or letters of intent — plus customer interviews and artefacts that back up the numbers.
How long does an MVP take?
It depends on the product: a pre-order landing page takes days, a software MVP usually a few weeks. If an MVP is taking six months, it probably includes too much.
Key points about MVPs
- An MVP is a tool for managing uncertainty, not a cut-down product.
- Unlike a prototype or PoC, only an MVP tests whether customers will pay.
- An MVP need not involve code: landing pages, concierge services and demo videos work too.
- One hypothesis, one metric, one channel, a success threshold set before launch.
- Investors want numbers backed by artefacts: payments, conversion, retention, proxy CAC.
