BYD TechThe short answer
After an MVP launches, measure whether people get value and come back, not how many arrive. The core metrics are activation (the share of new users who reach the first moment of value), retention by cohort (how many from each week’s sign-ups are still active later), one north star metric that reflects value delivered, and evidence of commitment such as payment or repeat use. A retention curve that flattens instead of falling to zero is the clearest early sign of something people want.
Key points
- Sign-ups and page views are vanity metrics on their own.
- Define activation: the first moment a user gets real value.
- Track retention by cohort, not as one average.
- A flattening retention curve is the signal to look for.
- Talk to users too: numbers say what, conversations say why.
Vanity metrics vs real ones
Page views, downloads and sign-ups go up with marketing spend and launch posts, whether or not the product works. They are useful for the top of the funnel, but they cannot tell you if anyone wants what you built. For that, look at what users do after they arrive.
The core metrics
| Metric | Question it answers | Example |
|---|---|---|
| Activation rate | Do new users reach value? | Share of sign-ups who create and share their first booking link |
| Retention by cohort | Do they come back? | Share of a week’s sign-ups still active four weeks later |
| North star metric | Is value growing overall? | Appointments booked through the product each week |
| Commitment | Will they invest in it? | Upgrades, pilots signed, teammates invited |
| Qualitative signal | Would they miss it? | Answers to "how would you feel if you could no longer use this?" |
Reading retention by cohort
A cohort is a group of users who started in the same period. Mixpanel’s retention report, for example, groups users by when they first did a starting event and shows how many return to do it again. Comparing cohorts shows whether product changes are working: if newer cohorts retain better than older ones, you are improving.
- A curve that drops to near zero: people try it and leave.
- A curve that drops, then flattens: a core group keeps getting value.
- Newer cohorts flattening higher: the product is getting better.
Setting it up
- Before launch, name the activation event and the return action that counts as active.
- Track those few events with consistent names and real user IDs.
- Use a product analytics tool rather than page-view analytics alone.
- Review cohorts on a regular rhythm, alongside notes from user conversations.
These numbers feed the question of fit. See What is product-market fit? and How to prioritize features. Analytics for the key actions belong in the first release of any MVP development project.
Sources
Checked when this guide was written.
