Progression

Progression that cannot be farmed

Gamification usually fails in one direction: the points become the goal. This one is built to make that unprofitable.

How does Crescendo gamify habits?

Crescendo has a progression system with experience points, seasons and ranks. The daily score rises linearly to a knee at 70 points and then saturates toward a hard cap of 130, so an unusually intense day cannot dominate a consistent week. Scores are recomputed from the ledger rather than incremented, which makes them idempotent and correct when data arrives late.

How the scoring works

Each day produces a score from what you logged, and that score is deliberately not linear. It rises normally up to a knee at 70 points, then saturates as it approaches a hard cap of 130. The effect is that the tenth habit completed in one day is worth much less than the second, which is the correct incentive: consistency across a week beats a single heroic Sunday.

Partial credit is given for partial quantity, capped at the target. Exceeding a target does not earn more, for the same reason.

Why the score is recomputed rather than incremented

Days get rescored constantly in normal operation — a watch syncs last night's sleep at noon, a completion is logged the following morning. A system that adds points on write has no correct answer for any of that; it has to guess whether it already counted something, and it will eventually guess wrong in a direction people notice.

So scoring a day is idempotent: running it five times leaves the same ledger row and the same total. That is what makes late-arriving data harmless rather than a source of silent double-counting.

Seasons and ranks

Progression runs in seasons rather than accumulating forever. An unbounded lifetime total rewards the people who arrived earliest and tells a new user that the interesting part of the system happened before they got there.

The ledger behind a rank is append-only, and every entry records the reason it exists — which is what makes a rank auditable rather than merely displayed.

How this differs from other gamified trackers

The usual failure mode of habit gamification is that optimising the score and improving your life come apart, and the score is easier. The saturating curve, the cap on exceeding targets and the integrity checks all exist to keep those two things pointed the same way.

It is a weaker dopamine loop than a tracker that awards points for everything. That is the trade, and it is deliberate.

At a glance

Daily score knee
70 points
Daily hard cap
130 points
Exceeding a target
No extra credit
Rescoring
Idempotent — recomputed, never incremented
Ledger
Append-only, with a reason on every row
Progression
Seasonal, not lifetime-cumulative

Common questions

Can I farm points by adding lots of easy habits?
Not usefully. The daily score saturates above a knee at 70 points toward a hard cap of 130, so each additional completion in the same day is worth progressively less. Ten trivial habits score close to what a few meaningful ones do, and take far longer to log.
Do points reset?
Progression runs in seasons, so ranks are seasonal rather than a lifetime total. Your underlying habit and goal history is never reset — only the competitive progression layer restarts.
What happens if my watch syncs yesterday's sleep today?
The affected day is rescored and the total recomputed from the ledger. Scoring is idempotent by design, so late data corrects the record rather than double-counting against it.
Does gamification actually help build habits?
It helps with starting and hurts with sustaining, in most implementations, because the reward becomes the point. The mitigation here is to make score maximisation and actual progress hard to separate: saturating returns, no credit for exceeding a target, and integrity checks on abnormal patterns.
Is there a leaderboard?
Progression is seasonal and rank-based. The design intent is comparison against your own prior seasons first; anything social is opt-in rather than a default that broadcasts your habits.

How this works in detail

Related

Give it a month and see what it finds

Twenty paired days is the minimum before the statistics will say anything at all. That is roughly three weeks of ordinary use.

Start free
Want it on your phone?

On iPhone: Safari, then Share, then Add to Home Screen.

On Android: Chrome, then the menu, then Install app.

Prefer to wait for native apps? Leave an email and we'll tell you the day they land. Nothing else.