D1, D7, and D30 retention measure the share of players who come back to a game one day, seven days, and 30 days after installing it. If 1,000 people install on Monday and 400 open the game on Tuesday, D1 retention is 40%. If 150 of the original 1,000 open it the following Monday, D7 retention is 15%.

Teams often report these numbers against benchmarks and work to push them up. Read side by side, they also work as a diagnostic, because each window covers a different stage of the player's relationship with the game, and a drop at each stage usually points to a different kind of problem. Below is what tends to break at each stage, and which research can answer it.

What retention means

Player retention means the share of players who return after a set amount of time. It is usually measured by cohort: everyone who installed on a given day is one group, and the team tracks what share of that group comes back on each day after.

Industry data puts mobile game churn at 97% of users by Day 30. Our studies have produced an average 7% increase in D1 retention, and our D1 ROI calculator estimates what a lift like that is worth for your game.

D1 retention: did the first session work?

D1 measures whether players came back the day after installing. Most of what affects it happens in the first session, and much of that in the first 15 minutes.

What usually breaks at D1:

• The player did not understand what to do, or how.

• The tutorial taught one way to play and the game then asked for another.

• The player understood the game and decided it was not for them.

• Passive screens, interruptions, or long dialogue stretched out the time between moments of play.

• The game asked for money, an ad view, or a wait before the player had a reason to want anything.

What research answers it: An FTUE playtest, where players who have never seen the game play it for the first time while the session is recorded. Six to eight participants per segment is usually enough to see the same problem repeat. FTUE: what the first-time user experience means in games covers this window in full.

Peaksel's analytics showed players leaving The Vault before level 3. In a six-player FTUE study, engagement dropped within the first 45 seconds of live gameplay. The tutorial had taught players to work on a single vault face, and the live game asked them to explore several faces, with no prompts. Peaksel redesigned levels 1 to 3, and its CEO, Marko Petkovic, reported 9.6% more users completing level 3, 7.3% more reaching level 10, and a 14.5% LTV uplift. The Peaksel case study covers the full study, and our FTUE Benchmarks Report analyzes 7,600 clips of first-session play across 22 games.

D7 retention: is there a reason to come back?

D7 measures whether players were still returning a week later. Players still around on Day 7 usually understand the game, so this window shows whether they still want to play it.

What usually breaks at D7:

• The core loop is clear but stops feeling new after a few sessions.

• Progression slows down, and the next goal feels too far away.

• Rewards stop feeling worth the effort.

• Difficulty rises in a way that feels unfair or random.

• The player never found the systems that would have kept them playing, such as events, collections, or social features.

What research answers it: A longitudinal study, where the same participants play across several sessions over a week or more, often combined with a diary study where they record short entries after each session. Interviews with players who stopped, and with players who stayed, help explain the difference between them. Level telemetry shows where progression slows.

Difficulty comes up often at D7. In our Royal Match study, a single failure usually raised emotional intensity, and players retried with more focus. Engagement dropped after repeated failures. A level that players fail once and a level they fail five times can both end in a completion, while affecting players in opposite ways. The full study is in our Royal Match deconstruction.

D30 retention: has the game become a habit?

D30 measures whether players were still returning a month later. Players at D30 have usually formed a routine around the game. They know its systems, and many have already decided whether to spend.

What usually breaks at D30:

• The player has seen most of what the game offers, and new content arrives too slowly.

• The metagame stops giving reasons to come back, because finished goals do not open new ones.

• Too many simultaneous events or systems make the game feel like work.

• Monetization pressure increases as the player progresses.

• Social ties are weak, so leaving costs the player nothing.

What research answers it: Playtests and interviews with experienced players, surveys of the live player base, and telemetry on long-term behavior such as event participation and spending. Studies at this stage usually recruit players who already play the game, instead of new ones.

In our Royal Match study, experienced players found the number of simultaneous events overwhelming. One player counted them out loud: "there are so many of them... one two three four five six seven different things going on right now." In our Merge Mansion study, each finished task revealed another unfinished area, object, or story beat, which gave players the next thing to want as soon as they completed the last one. Deconstructing Merge Mansion's retention engine covers how that loop holds players five years after launch.

The three windows side by side

What analytics can and cannot tell you

Retention numbers come from analytics, which can show which cohort dropped, on which day, after which level, and how that compares with last month. It cannot show what players experienced before they left.

In our Merge Mansion study, the event logs recorded idle time the same way whether a player was planning a move or stuck. The session recordings, with facial expressions and concentration matched to what was on screen, told the two apart. Analytics or playtesting for Day-1 churn? goes through an example.

For the wider context, see Game user research: what it is and what teams use it for.

Emhance playtests record players' emotional intensity and concentration second by second, so you can see what happened in the minutes before they left. See how Emhance playtesting works.