Designing a Progression System

I looked at how other games handled progression, Clash Royale most of all. You win and lose trophies per match, you climb through tiers, and the season resets and pulls the top of the table back down. Ours ended up much the same. Then I sketched, and wireframed every flow, including the awkward ones: a reward sitting unclaimed while you carry on playing, or a tier unlocking while you’re somewhere else in the game.

You earn trophies by placing in a run: 30 for first, 15 for second, 7 for third. Finish fourth to eighth and you lose 15. Losing them matters as much as winning them, because without that everyone ends up at the top eventually and just stays there. There’s a floor on the loss, though. It stops at a set tier, so a bad streak costs you position without wiping out what you’ve already earned.

There are 14 regular Tiers, each unlocking clothing items, lootbags and a 24-hour booster that halves every bag timer you’ve got running. Above Tier 14, at 5,000 trophies, 10 Ultimate Tiers work differently: at the start of each 28-day season you can drop from a higher Ultimate Tier to a lower one. Staying up there takes as much work as getting there did.

I prototyped it in Adobe XD first, because a tier map is mostly transitions: bars filling, rewards flying into the wallet, a tier unlocking underneath you. That’s hard to judge from a static wireframe. Then I wrote the design document for Art, Design, Code and QA, covering tier contents, run rewards by finishing position, the tier information popup and every tooltip state. My Lead Designer reviewed it, then the wider stakeholder group, and the flow, the prototype and the document all moved together after each round.

Art produced the assets and their own UI Supporting Document, Code built from both, and I stayed with them through implementation so what shipped matched the document. Then I tested alongside QA, raised what I found, and QA ran a final pass.

After soft launch the analysts reported the feature was performing well. For players in the regular tiers, average trophies climb steadily across the first 30 play days, from nothing to around 2,500. Early on, every session moves you up the map, which is what we were after.

Filter the same chart to Tier 14 and above and the climb disappears. The line sits flat between roughly 6,000 and 6,500 for the whole 30 days. That’s what we designed the Ultimate Tiers to do: once you’re up there you’re defending a position rather than climbing, so the numbers stop going up.

The distribution charts show fewer players in each Ultimate Tier than in the one below it. That’s expected. Getting up there takes skill and a lot of time, so the numbers thin out as the requirement goes up.

The map ends at 8,000 trophies. If you got there before the 28-day season closed, you’d carry on winning runs and earn nothing for them, so the best players had the least reason to keep playing.

So anything earned past the cap goes into a Trophy Vault. Hitting 8,000 triggers an Ultimate Champion screen and auto-claims the final reward, and after that wins bank into the vault and losses stop deducting, so you can’t drop back off the end of the ladder. I mapped it as a user flow first, because nearly all of it is edge cases: winning at the cap, losing at the cap, opening the map with rewards still unclaimed, and the season ending underneath any of those.

At the end of the season the vault unlocks, converts the trophies to coins and auto-claims them before the reset, so nothing earned in that window is lost. Then I wireframed it, wrote it up for the same four disciplines, and took it through the same review and iteration as the first half.

A player on day three climbs every session. A player at the top defends a position. A player who finishes keeps earning into the vault. Three different players, one ladder, and none of them run out of reasons to play.