To scope an MVP so the first version costs less, define one user, one problem and the single path that solves it, list every feature, then cut everything that is not needed to test that path. Write what is excluded, agree it before coding, and plan version two separately.
An MVP is the smallest product that can test your idea with real users or investors. Most overspending on a first version comes from building features before anyone knows which ones matter. Scoping is where you save the money, long before a developer writes any code.
Step 1: What is your one sentence?
Write it in this form: a [type of user] can [do one thing] so that [result]. If you need the word "and" to finish it, you probably have two products. Pick one.
Step 2: Which features survive the cut?
List every feature you want. Then ask one question of each: do we need this to test the sentence? Here is an example for a made up product, Kiteline, a marketplace where parents find home tutors.
| Feature | Needed to test the idea? | Verdict |
|---|---|---|
| Parent sign up | Yes | First version |
| Tutor profile page | Yes | First version |
| Search by subject | Yes | First version |
| Request a booking | Yes | First version |
| Online payments | No, payment links can be sent by hand | Later |
| Reviews | No, can be collected by message | Later |
| In app chat | No, WhatsApp works | Later |
| Mobile app | No, a mobile friendly website tests the idea | Later |
| Simple admin screen | Yes, you need to see requests | First version |
Step 3: What can you do by hand?
Do manually what you can, for now. Send payment links yourself. Onboard the first twenty users over WhatsApp. If people want the product, you will find out without building anything automatic. If they do not, you will have saved the cost of the automation.
Step 4: Is the scope in writing?
Write down what version one includes, what it excludes and when it will be ready, and agree it before any coding starts. The exclusions matter as much as the inclusions. They stop a feature quietly growing from a small ask into a month of work.
On our projects, the scope, exclusions and date are agreed and signed before coding begins.
Step 5: Have you seen it before it is built?
Ask for a clickable prototype and tap through it on your own phone before building starts. Changing a screen in a prototype usually costs far less than changing it once it is coded. This is also the cheapest moment to learn that a feature nobody needs is still on the list.
Step 6: Is the build in short rounds?
Building in two week rounds, with a working version you can log into at the end of each one, lets you see real progress and cut scope if the budget tightens. You are never waiting months to find out whether it works.
What does an MVP cost?
Our typical first version costs $22,000 to $60,000 and takes 10 to 16 weeks, quoted in writing after a call. Where you land in the range depends on scope: the number of user types, payments, connections to other systems, the number of platforms and how custom the design is. A tight scope pushes it toward the lower end.
If you would rather add a team to an existing effort, a dedicated design and development team costs $3,000 to $7,000 a month, with a three month minimum.
Which scoping mistakes cost the most?
- Designing for several kinds of user at once.
- Building an admin system before there are users.
- Releasing on web and mobile together. Our post on whether to build an app or a website first helps you choose.
- Adding features for rare cases.
- Writing a feature list with no exclusions.
- Perfecting custom design on every screen before the idea is tested.
Frequently asked questions
What is the difference between a prototype and an MVP?
A prototype is a clickable mockup. An MVP is a working product that real people can use.
Who owns the code?
You do. The repository moves to your account when the final invoice is paid, and you can see changes as they are made.
Can you take over a product someone else started?
Yes. We begin with a paid code review and a written summary of what we find, before any further quote.
Will you sign an NDA?
Yes, before the first detailed call if you ask.


