If you have an InBody or similar body-composition scanner in your store or gym, a challenge is the obvious way to make it earn its footprint. Members love a leaderboard, staff get something to talk about at the counter, and the mechanic is naturally sticky — you cannot have a result without coming back in to be measured again.
The obvious way to score it is also the one that kills it. “Whoever loses the highest percentage of body fat wins” sounds fair, feels fair, and in a mixed field it hands the trophy to the heaviest person who joined on day one. Everybody else works out the standings by week two and stops showing up. This post is about the four design decisions that stop that happening.
Why raw fat-loss percentage isn't fair
Do the arithmetic with two plausible members.
Member A starts at 280 lb with 34% body fat — that's 95 lb of fat mass. Member B starts at 165 lb with 20% body fat — 33 lb of fat mass. Both do genuinely good work for eight weeks and each reduce their fat mass by the same relative amount, 10%.
- Member A loses 9.5 lb of fat. As a share of starting body weight, that's 3.4%.
- Member B loses 3.3 lb of fat. As a share of starting body weight, that's 2.0%.
Identical effort, identical relative result, and A sits nearly three times clear of B on any leaderboard keyed to pounds or to percent-of-body-weight. This isn't a quirk — it holds across the whole field. Heavier starters have more fat mass to move and more water to shed early, and in the first two weeks they post numbers a lean member physically cannot match.
Two structural fixes, and you can use both:
- Score relative to what the member started with, not in absolute pounds. “Percent of your own fat mass reduced” puts A and B at a dead tie in the example above, which is the correct answer.
- Run brackets. A challenge with a men's and women's board, or a board split by starting body-fat range, is fairer than any single formula and takes no extra machinery. It also gives you more winners to celebrate, which is the actual marketing asset.
Multipliers and a combined score
Relative scoring fixes the biggest unfairness, not all of it. Two more realities push on the standings:
- Body composition moves at different rates by age and gender. A 55-year-old woman and a 24-year-old man doing equally disciplined work will not post the same percentages, and pretending otherwise means the same demographic wins every quarter.
- Some members aren't trying to lose fat. Someone adding four pounds of lean mass at stable weight has done something excellent and scores zero on a fat-loss board.
Hence two knobs. Age and gender multipliers scale a member's raw improvement before it's ranked — a small handicap, in the golf sense. A combined score adds fat lost and muscle gained into one number, so a recomposition counts.
Our challenge engine supports both: pick the metric — fat-loss percentage, muscle gain, weight loss, scan count, or a combined score — and attach age and gender multipliers to it. Deliberately a configuration rather than a built-in truth: there is no scientifically blessed multiplier table, and a vendor who hands you one is making it up.
Which leads to the rule that matters more than the formula: publish the scoring before anyone joins. Write the metric and the multipliers on the sign by the scanner. Members forgive a formula they think is slightly off. They do not forgive discovering the formula after they've lost.
Snapshot the baseline at join, not at scoring time
Here's the failure that turns a challenge into an argument. If your leaderboard computes “first scan in the period versus latest scan,” a member who scans monthly can shop for a flattering starting point while a member who scanned on a bad day is stuck with it. Worse, someone who hears about the challenge on day three can take a deliberately unfavourable scan — dehydrated, post-workout — and start from an inflated baseline they'll beat trivially.
The fix is structural, not a policy you police at the counter: joining is an explicit opt-in that snapshots the baseline at that moment. The member taps join, the challenge records the numbers as they stand, and no later scan moves the starting line. That's how ours works, with three useful side effects:
- Consent is explicit — nobody lands on a public leaderboard because a machine measured them.
- The join event is a real signal: you know who is engaged and can talk to them differently.
- Late joiners are handled honestly. They compete over a shorter window from their own baseline — allow it (more participants) or cut it off with a join deadline (cleaner comparison), but decide and say so up front.
One practical note on scan conditions: hydration, food, and time of day all move a body-composition reading, sometimes by more than a week of real progress. Tell members to scan under consistent conditions — same time of day, similar hydration, before training rather than after. You won't control this perfectly and don't need to; you need the noise to be similar for everyone, which means staff saying the same sentence to every member. The mechanics of the scan program itself are in our post on turning a body-scan machine into a retention engine.
Thinking about running one this quarter? If you want to see a challenge configured end to end — join, baseline, multipliers, auto-close — against your own scanner and member list, book a demo.
The rescan is the retention mechanic
This is the part most operators miss, and it reframes the whole exercise. A member who joins and never rescans has no result. Not a bad result — no result. They are not on the board at all.
Read that as an operations instruction, not a limitation. The challenge does not motivate the training; it motivates the measurement, and measurement requires a trip to your store. A member who rescans three times in a quarter has visited three times they otherwise might not have, and each visit carries a conversation, a recommendation, and a basket.
So build the whole program around producing rescans:
- Make a midpoint scan a thing. A four-week check-in on an eight-week challenge doubles your visits and shakes the standings before anyone gives up.
- Chase non-scanners specifically. Members who joined and haven't scanned since are the most responsive list in the challenge — they already opted in. A reminder that they're not on the board yet beats any promotion of the prize.
- Put the result somewhere public. A fresh scan on a screen on the gym floor does what an emailed PDF never will: the person next to them asks what it is.
- Score on scan count if participation is the goal. A “most scans” board rewards showing up rather than genetics — often the right first challenge for members who have never been measured.
The honest corollary: if scanner results don't reach the member's phone and your portal automatically, none of this works. Someone will be transcribing printouts by Friday and the board will be three weeks stale. That's the plumbing our scan integration removes — results land on the member record and the leaderboard updates as scans arrive.
Closing the challenge automatically
Every abandoned challenge we've seen was abandoned at the same point: scoring day. The period ends, and someone has to export scans, apply the formula in a spreadsheet, decide about the member who scanned twice on the last day, and announce a prize. That two-hour job lands in a busy week, so it slips, and the members who did the work learn the board doesn't really matter. Run that twice and nobody enters the third one.
So the close has to be automatic, and it has to be idempotent — meaning it can run more than once without paying anyone twice. Ours closes due challenges on a schedule, freezes the winners, and awards points with a reference tied to that challenge and that member, so a retry finds the award already recorded and does nothing. Without that property the safe choice is manual awards, and manual is what gets abandoned.
A few closing details worth deciding in advance:
- Prize in points, not cash. A points award lands in the loyalty balance and gets spent in your store, so the prize funds a return visit instead of leaving the till.
- Reward everyone who finished. A small award for every member with a valid rescan turns the bottom half of the board from losers into participants — and they're who you need entering the next round.
- Freeze the standings. Once closed, the result is a stored record, not something recomputed later from live scan data. Members screenshot the board; it should still agree with itself in a month.
Cadence: quarterly, six-week, or rolling
A rolling quarterly leaderboard has run at our founding store for years, and quarterly holds up for a specific reason: long enough that body composition genuinely moves, short enough that someone with a bad three weeks can wait for the next one. Nothing in the design is quarter-specific — the same pieces work for a six-week New Year push, a summer challenge, or a board that resets each quarter.
Two choices change the feel more than the length does. A fixed period with a join deadline gives a clean comparison and a real launch moment; a rolling board gives continuous participation with no launch push. And with more than one location, decide early whether the board is per store or company-wide: company-wide is more exciting and slightly less fair, per store gives every location a winner.
What a challenge will and won't do
It will give your scanner a reason to be used, produce return visits with conversations attached, and generate local word of mouth that costs nothing — which is why a challenge is usually the highest-return thing you can do with a body-composition machine you already own.
It won't fix a store with no traffic, and it won't hold members whose first visit left no impression — a leaderboard gives an engaged member a reason to return, not an indifferent one. If your scanner is new and barely used, run a most-scans challenge first and forget the multipliers; people need to be comfortable stepping on the machine before fair scoring is the interesting problem. And with twelve active members, run it on a whiteboard. The challenge engine is worth its fee when the board has to be right without anyone maintaining it, when the close has to happen whether or not you're on shift, and when the whole thing has to survive a busy quarter unattended — not before.