Product & MVP 6 min read
Launched an MVP but can't tell if it works? What to measure from day one
Many MVPs launch without a clear way to judge them. Here is how to choose a few useful numbers, track them from day one and decide what to do next.
By Digitice Admin
You launched your MVP three months ago. People have signed up. Some use it and some do not. When an investor or your board asks, "Is it working?", you are not sure what to say.
This is a common problem, and it is costly. You have spent time and money to learn something, but you cannot say what you learned. So the next big decision rests on opinion. That decision might be more funding, a new feature, or stopping. Weeks pass while the team argues about what the numbers mean.
An MVP (minimum viable product) is the smallest version of a product that real users can use. Its job is to prove one thing. If you cannot measure that one thing, the MVP has not done its job. This article explains how to fix that, ideally before you launch.
Why many MVPs cannot answer the question
The cause is rarely the product. It is usually one of these:
- No success measure was agreed. The team built the product, but nobody wrote down what "working" would look like.
- Tracking was added late, or not at all. By the time someone asks, the early usage data is gone.
- The wrong numbers get attention. Page views and sign-ups feel good. They rarely tell you if people value the product.
- Too many things were built at once. When ten features launch together, you cannot tell which one made a difference.
Each of these is a planning problem. That is good news, because planning is cheap compared with building.
Start with your riskiest assumption
Every new product rests on a few beliefs. One of them is the riskiest. If it is wrong, nothing else matters.
For example, you might believe that small landlords will pay monthly for a tool that tracks repairs. The riskiest part may not be whether they sign up. It may be whether they keep using it after the first month.
Write this assumption down in one sentence. Then write the number that would prove or disprove it. Then set a target and a date. For example: "Within eight weeks of launch, a clear share of new landlords log a repair in their second month." Agree the exact target with your team before launch, so nobody moves it later.
This also shapes what you build. You only need the one user journey that tests the assumption. A user journey is the path a person takes to get a job done, such as signing up and logging a first repair. Build that one journey properly, and leave the rest for later.
The few numbers that matter
You do not need a large dashboard. Most MVPs need four or five numbers. Here is a simple way to think about them.
| Stage | The question it answers | Example measure |
|---|---|---|
| Acquisition | Are the right people finding us? | Sign-ups from your target group |
| Activation | Do they get to the useful part? | Share of new users who complete the key action |
| Retention | Do they come back? | Share of users still active after four weeks |
| Revenue | Will they pay? | Share of active users who start a paid plan |
A few plain definitions:
- Activation means a new user reaches the moment where the product helps them. You decide what that moment is.
- Retention means users return over time. For most products, this is the strongest sign of real value.
- Conversion means the share of people who move from one stage to the next, for example from free to paid.
Pick the one number that tests your riskiest assumption. Treat it as the main measure. The others give context, but they should not distract you.
Set up tracking before launch, not after
To get these numbers, the product must record what users do. This is called event tracking. An "event" is a single action, such as "signed up", "created first project" or "started paid plan". Each event is saved with a time and a user, so you can count and compare them later.
Good tracking needs a little care:
- Name events clearly and consistently. "Repair logged" is better than "button 3 clicked".
- Track the key journey from start to finish. Then you can see where people drop out.
- Build a simple dashboard. A dashboard is one screen that shows your main numbers. Everyone should look at the same one.
- Respect privacy. Tell users what you collect and why, and only collect what you need.
- Test the tracking before launch. Broken tracking is very common, and you usually find out too late.
If you add tracking after launch, you lose the first weeks of data. Those weeks often matter most, because early users are the most curious and the most honest. If your data is already spread across several tools, our data and analytics service explains how to bring it into one view.
Read the results, then decide
Numbers tell you what users do. They do not tell you why. So pair your dashboard with short conversations. Speak to a handful of users who stayed and a handful who left. Ask what they were trying to do and what got in the way.
Set a review date before launch, for example six or eight weeks after. On that date, compare the main number with your target. Then choose one of three paths:
- Continue. The number met the target. Plan the next phase and the next assumption to test.
- Change. Some users found value, but the main number fell short. Change one thing, such as the onboarding or the price, and measure again.
- Stop or pivot. The number is far from the target and conversations confirm the problem. This is a useful result. It saves you from spending more on the wrong idea.
The trade-off here is speed against certainty. Waiting longer gives you more data, but costs time and money. A fixed review date protects you from waiting forever.
What this means for how you build
Measurement only works if the product underneath is reliable. If the MVP breaks under real use, or loses data, the numbers mean little. So the foundation matters: real user accounts, secure data, automated tests and safe updates. Tests and safe releases let you change one thing at a time after launch, which is how you learn.
This is the approach we take in MVP development. In the first week we agree the riskiest assumption, the one journey that tests it and the success measure. Analytics is part of the build, not an extra. For a validated idea, MVP45 delivers this in 45 days for a fixed price.
Your next step
Before you build anything more, write two sentences. The first names your riskiest assumption. The second names the one number that will test it, with a target and a date. If you cannot write them yet, that is the most useful thing to work on this week. If you would like a second opinion on them, get in touch.
Questions
Frequently asked questions
What is the most important metric for an MVP?
The most important metric is the one that tests your riskiest assumption. For many products this is retention, meaning whether users come back, because it shows real value. But you should choose it by asking which belief, if wrong, would sink the product.
How long should I wait before judging an MVP?
Set a fixed review date before launch, often several weeks after release. The right length depends on how often users would naturally use the product. A tool used weekly needs less time than one used monthly.
Are sign-ups a good measure of success?
Sign-ups alone are a weak measure. They show interest, not value. Look at what happens after sign-up, such as whether people complete the key action and return.
Can I add analytics after my MVP has launched?
Yes, you can, but you will lose the early data you never recorded. It is better to plan and test event tracking before launch, so the first weeks of real use are captured.