The weights
- Participation 35%
- Consistency 25%
- Improvement 25%
- Mutual support 15%
JESS MOVEChallenges
Inclusive competition
A leaderboard that rewards capability is always won by the same person, and everybody else stops opening the app in week two. So capability is not weighted down in the scoring function — it is absent from it. A ten-year-old, a wheelchair user and an eighty-eight-year-old contribute on exactly the same four terms as a marathon runner, and the runner has no lever the others lack.
Team Score
| Term | What it measures | Why it is there |
|---|---|---|
| Participation 35% | The share of the team who did anything at all this week. | It is the only term a large team cannot win by carrying one strong member. |
| Consistency 25% | How evenly effort is spread across the days, per person. | Rewards the person who does two minutes on five days over the one who does a burst. |
| Improvement 25% | Movement against each person’s own previous fortnight. | Everyone starts from where they are, so everyone has the same headroom. |
| Mutual support 15% | Encouragement given, sessions joined, teammates brought back after a lapse. | Makes helping somebody else the highest-leverage thing you can do. |
The guard
teamScore() inspects the keys it is handed and throws if any of them resembles raw output. Fourteen field names are refused outright, so a well-meaning integration cannot quietly reintroduce a fitness leaderboard.
Substring matching, so weeklySteps and avgHeartRate are caught as readily as steps. Asserted in continuous integration against every name on the list.
This is the whole design in one chart. The team with three genuinely exceptional athletes scores 32.8. The team where everybody manages two minutes most days scores 83.5.
No individual may supply more than this share of a team's total. Our simulated superstar on a two-person team was capped at 18%.
Against a target of 60. A challenge a team of entirely median people cannot win is not shipped — this one passes.
Leaderboards
Loss framing works, and it works by making people feel worse. Six rules govern every ranked surface in the product, and they are asserted as a build gate rather than left to a design review.
Nothing. Participation is measured against active members rather than the roster, so stepping out during illness, a flare-up, a bereavement or a new caring responsibility cannot drag your colleagues down. The engine withdraws you automatically on any of the triggers above — you do not have to ask, and nobody is told why.
The library
Workplaces with a back-to-back culture
Any team, any size
A class, with the teacher triggering breaks
A household across up to four decades
Residents and carers together
Councils and public-health programmes
Anyone; a sponsor converts participation to a donation
Teams with people who lapsed
Create an account and a challenge takes about a minute to set up. For a whole workforce or a school, a pilot is the better route.
Free to create. 150 AI credits a month for 2 months, then AI features need a plan — everything that is not AI carries on either way. Premium is £5.99 a month and cancels in one click. Under 18 a guardian confirms before the account opens.