Retention

See who comes back, and what brings them back

New players are only worth something if they return. Retention shows how many do, how that is changing, and which valuable players are drifting away.

What Retention does for you

Day 1, 7 and 30 return

The share of new players who played again exactly one, seven or thirty days after their first day, so weeks and games compare fairly.

Cohort heatmap

Players grouped by when they first joined, and what share came back in each later period. Switch between daily, weekly and monthly.

Retention curve

Weekly retention over 8 weeks, with your last 4 weekly cohorts against the 4 before. A lower recent line means something changed for new players.

First session length

How long new players stay in their very first session. A big share leaving within 5 minutes points at your spawn or tutorial.

Lifecycle and at risk

Every player is new, active, at risk, lapsed or churned. At-risk players are listed by spend, so you reach the valuable ones first.

Return rate by source

Retention split by where players came from, so you can tell a source that brings many players from one that brings players who stay. On Growth and up.

How you'd use it

A new spawn that scared people off

The situation

You rebuilt spawn and the tutorial, and a week later player counts feel flat even though joins are steady.

With Analyse

First session length shows more new players leaving within 5 minutes, and day 1 return for the latest cohorts sits below the ones before, so you know it is the first minutes, not your marketing.

A voting site that brings tourists

The situation

One voting site sends more new players than anything else, and you are about to pay for a featured slot.

With Analyse

Return rate by source shows those players rarely come back on day 7 compared with players from your YouTube domain, so you put the money behind the source whose players stay.

Catching regulars before they lapse

The situation

Some long-time supporters have not joined for over a week.

With Analyse

The at-risk table lists them sorted by spend, so you reach out to the top few with a personal message before they drift to another server.

Questions

How do I measure player retention on a Minecraft server?

Install the Analyse plugin. Retention is worked out from sessions automatically: day 1, 7 and 30 return rates, cohorts and lifecycle stages, with days in your game's timezone.

What is a good day 1 retention for a game?

It depends heavily on genre and where players come from, so the most useful comparison is your own game over time. The retention curve compares your latest weekly cohorts with the ones before, and an alert can tell you when day 1 return drops.

What counts as an at-risk player?

A player who has had at least 3 sessions and has been quiet for 7 days or more. After 30 days they are lapsed, and after 90 days churned.

Which plan do I need for retention?

Free shows day 1 and day 7 return. Starter unlocks the full page with cohorts, curve, lifecycle and at-risk players. Return rate by source needs Growth or higher.

Connect one game and see it for yourself.

Free to start, 14 days of Pro, and about four minutes of setup.