How to score a team competition at work fairly

Any contest between teams, branches or departments needs a way to turn activity into a score. The way you score decides who feels the contest was fair, and whether people come back for the next one.

Most of the advice below comes from vendors who sell contest or leaderboard tools, so read it as practical suggestions and not tested findings. Where research exists, it is a 2026 review of 18 studies, which did not appraise the quality of the studies it included. No independent, tested guidance turned up on how to adjust for teams of different sizes, and the article says so where it matters.

Decide what the contest is for

A contest can bring people together, or it can rank them. Choose before you design the scoring. Two vendor templates from MangoApps advise keeping a contest cooperative and time-bound so it does not become a permanent leaderboard that publicly shames lower performers, and using department-level standings, not individual rankings, because public ranking of individuals can create resentment.

Make it optional

A 2026 review notes that the effect of competition depends on design: optional competition and positive feedback support motivation, while required competition and a climate that centres on beating others undermine it. One workplace contest tool enrols individuals only if they opt in, so that nobody is publicly ranked without enrolling.

Make sure many people can win

A 2019 SHRM news article describes a 2008 hotel laundry leaderboard that staff called "the electronic whip". It also quotes a gamification vendor saying that a top-10-only board across 10,000 employees gives the average worker almost no chance and demotivates them. Another consultant in the same article says a design works when it creates many winners, and one company in the piece changed its weekly metric so that different people could win. These are anecdotes from the article, not measured results.

The 2026 review adds that leaderboards invite upward comparison, which can lower self-esteem for those who fall behind. Some options:

Keep the board at team level.
Run several short rounds with different measures.
Use separate leagues, so people compete with similar groups.
Compare each team's improvement against its own recent baseline, as a "most improved" format does. One vendor says to pick a baseline of equal length that avoids holiday weeks or outages.

Fix the teams and rules at the start

Lock the roster and the scoring when the contest begins. One contest tool snapshots team membership from the organisation structure at the start and freezes it, along with the scoring mode, the target and the time window. Changes can still be made, such as disqualifying someone, but with a recorded reason. A fixed roster stops people moving teams to chase points, and a fixed rulebook stops disputes about what changed.

Publish exactly what counts

Tell people before the contest starts what actions score, how many points each is worth and how standings are calculated. Vendor templates advise publishing the exact actions that count and not changing them after the start, and warn against managers sharing the prize but not the scoring rules. A recognition vendor makes the same point: share how scoring works with everyone.

Count verified actions where you can

Tie scoring to actions that can be checked, such as shifts completed or tasks closed in a system, not self-reported activity. One vendor says this stops a contest being inflated by self-reporting, and warns that missing or delayed logs cause disputes. If you cannot verify an action, say who will check it and how.

Make teams of different sizes comparable

If one team has 4 people and another has 12, a raw total favours the big team. Several sources suggest scoring per person. A workplace contest vendor says a six-person store can compete with a forty-person one under per-capita scoring, though it gives no formula.

An illustrative example, with invented numbers: Team A has 4 members and scores 120 points. Team B has 12 members and scores 240. Raw totals put Team B ahead. Per member, Team A has 30 points each and Team B has 20, so Team A leads.

Another approach scales each team's points by average team size divided by its own size. LiveSchool, a school house-points platform and a competitor of Arena, describes it that way in its balanced scoring help page: smaller teams get a boost and larger teams a modest penalty. A personal blog on gamification suggests the same idea of boosting smaller teams, though its written formula and its own table disagree on direction, so use LiveSchool's description. With teams of 4 and 12, the average size is 8, so Team A's points would be multiplied by 2 and Team B's by about 0.67.

Two cautions:

These are vendor and blog suggestions, not tested methods. A trivia-night blog models team size differently, assuming each person has a chance of knowing an answer and that the team picks it if any member does, and it notes this is unrealistic. The right adjustment depends on how individual effort adds up to a team result.
Whatever you choose, publish the formula before the contest. One vendor tells organisers to explain it to larger groups so they do not assume an automatic advantage.

You can also reduce the problem by building teams of similar size, which is our suggestion and not from a source, and by mixing departments deliberately, as one office-contest guide advises.

Vary the events, and avoid ones that exclude

If the contest has several events, a simple scheme is points by rank in each event: one guide suggests 3, 2, 1 and 0 for first to fourth place, and four teams as a good number for an office. The same guide warns that a self-selected sales team can leave you with "one dominant team and three resentful ones", and advises avoiding events that heavily favour physical fitness.

Have a tie-break ready

Decide before you start how ties are settled. One office-contest guide gives a quick head-to-head challenge as an example. A hackathon platform suggests deciding in advance whether judges vote or a specific criterion breaks the tie. Whatever you pick, say so in the published rules.

Keep rewards modest

A recognition vendor warns that if rewards are the only motivator, motivation could fade quickly, and that big rewards could encourage people to cheat to win. A small prize and public thanks are enough for most contests.

Check employee-relations rules

If your workplace has union, works-council or employee-relations requirements, the vendor templates advise presenting the contest as a cooperative team contest with fair scoring and no punitive public ranking of individuals, and keeping it separate from disciplinary policy. One tool has a setting for regions where public ranking of employees needs works-council consultation. These vendors say their own notes are for orientation only and to check with counsel. Do the same with your HR or legal adviser.

A checklist before you launch

A stated purpose, and a team-level board
Taking part is optional, and individual ranking is opt-in or absent
Teams and roster fixed at the start
The exact scoring actions and points published, and not changed
Only actions you can verify count
A way to compare teams of different sizes, with the formula published
More than one way to score well, and events that most people can join
A tie-break agreed
A modest prize
A check with HR or a legal adviser on local rules

Where this fits

Fair scoring applies to any team contest. See the workplace wellness challenge guide and the first internal hackathon guide for two common kinds. If you want a place to run a team competition, see Arena for companies.