An MVP is not a smaller version of your product. It is the cheapest thing that answers your riskiest assumption — and that assumption is almost never technical.
The one that needs testing first is usually this: will somebody use it, and will they pay for it. Everything else waits.
Want us to scope your MVP with you?
Tell us the idea and what worries you most and we will tell you what is enough to answer it.
What goes in and what gets cut
The rule: one user journey, start to finish, with no detours.
| Goes in | Gets cut |
|---|---|
| The core action that solves the problem | Settings and preferences |
| Sign-in, as simple as possible | Roles and permissions |
| One payment method, if you are selling | Discounts, vouchers, bundles |
| One results screen | Reports and exports |
| Manual support from you | Automated support |
The last line surprises people. In an MVP, anything that can be done by hand is done by hand — and it shows you what is worth automating later.
Three questions for every feature you consider adding:
- Without it, does the core journey work? If yes, cut it.
- Do we learn anything from it? If not, cut it.
- Can it be done manually for 20 users? If yes, do not build it.
The four-week plan
| Week | What happens | Deliverable |
|---|---|---|
| 1 | Scoping, user journey, screens | Clarity on what gets built |
| 2 | The core function | The main action works |
| 3 | Sign-in, payment, data | The complete journey |
| 4 | Testing, fixes, first 10 users | Live with real people |
The first week decides the other three. If you cannot write on one page what the user does and in what order, the timeline will not hold — the format for that document is in technical requirements.
And one rule that saves week four: freeze the scope at the end of week one. Every new idea goes on a list for later, not into the build.
What it costs
| Approach | Cost range | When it fits |
|---|---|---|
| No-code / low-code | €0–3,000 | A simple flow, testing demand |
| An external team | €8,000–25,000 | Real usage, payments |
| An in-house developer | Salary cost | If a team already exists |
These ranges are indicative of the market and shift — confirm with quotes.
Start with the first row if your assumption is commercial. If the question is “will anybody want this?”, it does not take custom code to answer — see what no-code platforms do.
How to choose and brief an external team: outsourcing software development.
Want an estimate for your own scope?
Send us your user journey on one page and we will tell you what it takes in time and cost.
How you know it worked
The wrong criterion is “it works”. The right one is what you learned.
| Metric | What it means |
|---|---|
| Completion of the core action | They understand what it does |
| Return within seven days | It solves a real problem |
| Payment, even from a few | There is value, not just interest |
| What they ask for unprompted | Where the next step is |
The third is the only unarguable one. Ten people paying tell you more than a thousand waiting-list sign-ups.
And one thing that gets overlooked: the first twenty users are found by hand. If you need an advertising budget to find twenty people, the demand assumption has not been tested yet.
The mistakes that cost time
- Building whatever the first user asked for. One user is not a pattern.
- Perfectionism about appearance. Clean is enough; visual polish comes after validation.
- No measurement. Without basic usage tracking, you will not know what happened.
- Delaying for “one more feature”. An MVP that never shipped learned nothing.
- Technical decisions for scale that does not exist. Infrastructure for a hundred users, not a hundred thousand.
The fourth is the commonest. Every week of delay is a week without an answer to the question you actually care about.
How the findings connect to the numbers in your plan: business plan.
Frequently asked questions
Do I need a technical co-founder?
Not for an MVP. For sustained development after validation, it helps considerably.
What if the MVP fails?
It answered the question for a few thousand rather than tens of thousands. That is the method succeeding.
Do I throw it away afterwards?
It usually gets substantially rewritten. That is why you do not invest in perfect architecture at the outset.
How do I protect the idea?
Execution protects you more than secrecy does. For contracts and code ownership, speak to a solicitor.
Do I need a legal entity from the start?
For testing, not always. From the moment you take money, yes — ask an accountant.
The first step
Write your riskiest assumption in one sentence: the one that, if wrong, invalidates the rest.
Then ask: what is the cheapest thing that answers it? Often it is not software — it is ten conversations with prospective customers.
Next step
Book a free 30-minute call with the team at The Dev Alley. We will scope your MVP and work out what can ship in four weeks.