---
title: "MVP checklist: plan your first release in 6 steps"
description: "Use this MVP checklist to plan your first release in six steps, from your riskiest assumption to the numbers that decide whether you keep building."
canonical: "https://benjaminschleier.com/en/guides/mvp-checklist/"
language: "en"
---

# MVP checklist: plan your first release in 6 steps

Start-ups

At Codestep we developed our own SaaS products and built software for start-ups. A first release should test exactly one assumption.

6 steps · 4 min read

## In short

Six steps for planning a first release before you put any money into development. In this order, you phrase your riskiest assumption, pick one user and one task, test demand, cut the scope, set a deadline and budget, and decide in advance which result counts as success.

## 1. Write down your riskiest assumption

Every business idea rests on several assumptions, and one of them can bring the whole thing down. Usually it is about willingness to pay. Phrase it as a sentence that could turn out to be false, for example “property management companies will pay monthly to have tenant inquiries sorted automatically.” Hang that sentence where you can see it, and check every later decision about scope against it. If a feature helps test the assumption, it stays. Everything else moves to the list for later.

### Questions to ask

- Which assumption would end your project if it turned out to be wrong?
- Could an experiment prove the sentence false?

**Warning sign** Your assumption reads “the market is huge.” That cannot be tested, and you cannot build a real plan on it either.

## 2. Pick one user and one task

A first release serves one kind of person with one problem. Describe that person so precisely that you could list ten of them by name. “Small businesses” is not enough. “Owners of physical therapy practices with a few employees” is. Then write down the one task this person gets done with your product, from the first click to the result. That path is your product. Roles, permissions and a second type of user get their turn once the first one is happy.

### Questions to ask

- Can you name ten people you could call tomorrow?
- Which path does the user take from sign-up to result?

**Warning sign** The description of your target group contains the word “everyone.”

## 3. Test demand before you build

Talk to the people on your list and ask about the last time the problem came up. After that, put a single page online that describes your offer and asks for a pre-order or an appointment. If your offer allows it, deliver the service by hand for your first customers. You learn in a short time what someone will actually pay for, and you still have not written a single line of code.

### Questions to ask

- Who has run into the problem themselves in the last four weeks?
- Has anyone committed to a pre-order or a fixed appointment?
- Which steps can you handle by hand for now?

**Warning sign** Everyone likes the idea, and nobody wants to commit.

## 4. Cut the scope down

Take your feature list and mark what the user absolutely needs for the one task. Everything else goes on a second list headed “later.” Experience says that list will hold the settings page, the admin area, the payment integration and the phone app. Invoices you write by hand at first. Two things are exempt from cutting, namely the core function and the way you handle your users' data. Whatever stays in has to run cleanly. Otherwise you are testing your implementation and learning nothing about the idea.

### Questions to ask

- Which features does the user really need for that one task?
- What can be handled by hand for the first few months?

**Warning sign** Your “later” list is empty.

## 5. Set a deadline and a budget

Give the first version a deadline and a budget, and write both down. If it gets tight, you cut scope. The deadline stays. Each time something slips, check whether your original assumption still holds or whether you are building features nobody asked for. How long it takes depends on the product. As soon as payments, health data or other regulated areas are involved, plan for considerably more time.

### Questions to ask

- How much money can be gone without ending your project?
- What will you see on screen after the first milestone?

**Warning sign** The date slips for the second time, and the scope has grown in the meantime.

## 6. Decide in advance what counts as success

Before launch, write down two numbers. One tells you the result at which you keep building, the other the point at which you change direction. That could be the number of paying customers after four weeks, or the share of testers who come back in week two. Build in a simple measurement that delivers exactly these values. Also set a date on which you look at the numbers and decide, and put it in the calendar today. Without that date, a half-successful product keeps running for months and burns through your money.

### Questions to ask

- Which number means “keep going” and which means “turn back”?
- On which day will you make the decision?

**Warning sign** After launch, the success criteria get adjusted to fit the results.

How small a first release can be, and why, is the subject of the blog article.

[How small should your MVP be?](https://benjaminschleier.com/en/blog/minimum-viable-product-scope/)

### Frequently asked questions

#### What is an MVP?

MVP stands for minimum viable product, the smallest version of a product that lets you test a business assumption with real users. It serves one target group with one task. Its purpose is to answer whether someone has the problem urgently enough to pay for a solution or keep coming back.

#### How long should it take to build an MVP?

Set a deadline before you start, and cut scope when things get tight. The deadline stays. Once payments or health data are involved, you need considerably more time.

#### Do I need developers to build an MVP?

Often not until later. Conversations with potential customers, a single page with a pre-order button and a service delivered by hand all work without code. You need developers once these tests show real demand and you know which steps software has to take over.

## Keep reading

- [How small should your MVP be?](https://benjaminschleier.com/en/blog/minimum-viable-product-scope/) · Start-ups · November 5, 2026 · 4 min
- [Choosing a SaaS development partner: 6 checks](https://benjaminschleier.com/en/guides/saas-development-partner-checklist/) · SaaS · 6 checks · 3 min
- [How to manage a remote team: 6 rules from my daily work](https://benjaminschleier.com/en/guides/how-to-manage-a-remote-team/) · Remote work · 6 rules · 3 min
