Skip to main content

Retention

Track how players behave after their first deposit — whether they keep coming back, how many stay, and how much revenue each cohort brings in over time. Retention groups players by the day they made their first-ever deposit, then follows each cohort forward day by day so you can spot exactly where players drop off or where value builds.

Use Cases

Spot your drop-off cliff. You launch a welcome offer and want to know if it actually keeps players around. Open the Rate view and look at the D1 and D2 columns — if retention collapses right after D0, the offer isn't creating a reason to come back.

Compare cohorts week over week. You changed your onboarding flow last Monday. Group cohorts by Week and compare the new cohorts against the ones before the change — same shape of curve, or did retention improve?

Find your payback point. You spend to acquire players and need to know when they earn that spend back. Switch to Cumulative ARPPU and watch the per-user revenue climb across days — that's how fast each cohort recovers its acquisition cost.

Check the quality of a traffic source. Add a Player UTM Source filter and isolate one channel. A channel that brings volume but flat retention is very different from one that brings fewer but stickier players.

How It Works

A cohort is a group of players bucketed by the date of their first-ever deposit. Every player in a cohort starts at D0 — the day they deposited. From there, each following day (D1, D2, and so on) measures what that same group did relative to where they started.

Only successful real-money deposits are counted. D0 is always the full cohort, so in the Rate view D0 is always 100%.

Each row in the table is one cohort. The Average row at the bottom is the simple mean across all cohorts that have a value in that column — so later columns average over fewer cohorts, since the most recent cohorts haven't lived long enough to have data there. The chart draws each cohort as a faded line and the Average as the bold line.

There are three views of the same cohorts:

  • Rate — the share of the cohort still active on each day, as a percentage of the original cohort. D0 is 100%, and each later day shows what fraction came back.

  • Users — the raw count of players from the cohort active on each day. No percentages, just headcount.

  • Cumulative ARPPU — total revenue the cohort has brought in up to and including that day, divided by the cohort size. This is per-user payback at each point in time: how much one player from that cohort has returned by D0, D1, D2, and so on.

Step-by-Step

  1. Go to Analytics → Retention.

  2. In Cohort Period, set the Cohort Period date range. This is the window of first-deposit dates that defines which cohorts appear.

  3. Under Group cohorts by, choose how players are bucketed:

    • Day — one cohort per calendar day.

    • Week — one cohort per week, starting Monday.

    • Month — one cohort per month, starting the 1st.

  4. Under Periods to show, choose how many columns the report displays:

    • Auto — matches your selected range, capped at 30 columns (D0–D29). A 10-day range shows 10 columns; a 90-day range grouped by month shows 3.

    • 7, 14, or 30 — a fixed number of columns.

  5. In Scope, pick your Project.

  6. (Optional) Click Add filter to narrow the player set. Available filters:

    • Player ID — IN or NOT IN a list of IDs.

    • Country — IN a selected list of countries.

    • Wallet Currency — = or != a value.

    • Player Language — = or != a language.

    • Refcode — =, !=, ILIKE, or NOT LIKE a value.

    • Player UTM Source — = or != a value.

    • Player UTM Medium — = or != a value.

    • Player UTM Campaign — = or ILIKE a value.

    • Player Browser — = or != a value.

    • Player OS — = or != a value.

    • Player Device — = or != a value.

    • Player Platform Type — = or != one of Browser, Facebook, Instagram, Mobile App, or WebView.

  7. Switch between the Rate, Users, and Cumulative ARPPU tabs above the chart to change what each cell measures.

  8. Click Apply to run the report.

  9. (Optional) Click Export to download the results.

Tips / Things to Know

  • D0 is always the full cohort: 100% in Rate, and the full headcount in Users.

  • The Average row and the bold chart line are simple means across cohorts. Recent cohorts contribute to early columns only, so the far-right columns are averaged over fewer cohorts and can move sharply.

  • A small cohort (low SIZE) makes its percentages jumpy — one player is a big swing. Check the Users view alongside Rate to see how many players are actually behind a number.

  • Filters here match the ones in Registration Cohort Metrics, so you can slice both reports the same way.

Related Articles
Did this answer your question?