MEASURED IMPROVEMENT · Tile Match
29% more starters reached Level 20 in Tile Match.
A few levels were stopping players for no good reason. After balancing, more players reached Level 20, the win rate improved, and more players earned Token 3.
Two 30-day telemetry periods compared.
THE MAIN MOVEMENTS
Before and after, shown side by side.
Win rate
+9.35 pointsLoss rate
-7.85 pointsToken 3
+10.74 pointsReach to Level 20
+29%A CLOSER LOOK AT THE DATA
The level-by-level progression view.
The chart follows the change beyond the headline checkpoint.
Player reach before and after balancing
Each period starts at 100% of its Level 1 players. The lines show how much of that starting group reached each later level.
Level 20 reach increased by 29%. The later period stayed above the earlier period through most of the game.
THE NUMBERS BEHIND THE CHART
Reach at selected checkpoints
The percentage is normalised against the Level 1 count inside each reporting period.
| Level | Before reach | After reach | Change |
|---|---|---|---|
| Level 1 | 100% | 100% | Starting base |
| Level 5 | 40.43% | 48.26% | +19.4% |
| Level 10 | 20.74% | 25.10% | +21.0% |
| Level 20 | 9.31% | 11.97% | +28.6% |
| Level 40 | 2.66% | 5.02% | +88.7% |
| Level 100 | 1.86% | 2.32% | +24.7% |
Two 30-day telemetry periods: 16 May–16 June and 16 June–16 July 2026. Level 20 is the headline checkpoint; the full curve shows how the change carried through the rest of the game.
What was happening
Some levels were losing players too often, while the token targets did not match the kind of clear we wanted to reward. We checked how comfortably players cleared, how often they earned Token 3, whether they played again, and how far they progressed.
What the data showed
The later period had a higher win rate and lower loss rate. More players earned Token 2 and Token 3, replay after a win rose by 5.28 points, and normalised reach to Level 20 moved from 9.3% to 12.0%.
What we did
We checked all 100 levels, found the walls that needed attention, and specified the difficulty and token-target changes. We also told the team what to measure after release so an increase in easy wins would not be mistaken for better progression.
A REAL PART OF THE WORK
What was delivered
| Level | Objective total | Time limit | Token 2 | Token 3 | Check after release |
|---|---|---|---|---|---|
| Level 2 | 42 | 70s → 85s | 60% of total | 85% of total | Win rate, time left, and timeout pressure |
| Level 10 | 93 | 100s → 110s | 50% of total | 70% of total | Win rate, time left, and replay after a win |
| Level 20 | 90 | 80s → 85s | 65% of total | 90% of total | Completion, token attainment, and Level 21 reach |
| Level 38 | 105 | 90s → 100s | 50% of total | 75% of total | Win rate, failure pressure, and time left |