-
1
Separate the leaderboard from the scorecard before you start
The first design mistake is treating the leaderboard like a mini scorecard with ten columns. A scorecard is private, multi-metric, and feeds a coaching conversation between a manager and a rep. A leaderboard is public, narrow, and feeds competitive energy across the whole floor. They do different jobs and they live in different places. Lock the distinction before you pick a single metric. Write it down for the sales leader: the scorecard stays in the one-on-one, the leaderboard goes on the TV. If you blur the two, you will end up with a public ranking on activity metrics that nobody can defend when a rep asks why the ranking says they are losing while they are hitting number.
- Define the scorecard in one sentence: private, multi-metric, coaching-driven
- Define the leaderboard in one sentence: public, one or two metrics, energy-driven
- Agree with the sales leader that neither replaces the other, they run in parallel
- Pick the owner for each: managers own scorecards, the sales leader owns the leaderboard
Tip: If someone asks to combine them into a single dashboard, say no. The leaderboard is a motivational artifact; the scorecard is a management artifact. Collapsing them produces a bad version of both.
-
2
Pick one outcome metric, or at most two
A leaderboard with five metrics is wallpaper. Reps cannot track five rankings at once, so they default to the top one and ignore the rest. Pick one headline outcome metric that maps directly to the quarterly plan: closed-won revenue for closers, pipeline sourced for prospectors, meetings booked for SDRs. If the quarter has a secondary push, add one supporting metric, and only one. The acid test is simple. Can a rep look at the TV, see their row, and know in two seconds whether they are winning or losing. If the answer requires reading a chart or doing math, the metric is wrong. Pure outcome metrics beat activity metrics on a public board because activity is easy to inflate and outcomes are not.
- Pick one outcome metric that the quarterly plan actually hinges on
- If a second metric is needed, make it a direct supporting signal, not a parallel vanity number
- Confirm every rep on the board can influence the metric this quarter, not just the top performers
- Write the exact definition and source table so the number cannot be argued about later
Tip: Avoid putting a leading indicator like dials or emails on the public board. Pay on outcomes, show outcomes. If you need a leading signal, put it on the scorecard where coaching happens.
-
3
Size the cohort so the ranking is fair
A single leaderboard that mixes SDRs with closers, or new reps with ten-year veterans, teaches reps that the game is rigged. Fairness is a design decision, not an aspiration. Split the board into cohorts that compete apples to apples: SDRs by segment, closers by segment, new hires in their own ramp cohort for the first two quarters. Keep each cohort large enough that the ranking is meaningful (eight to twenty rows is a good band) and small enough that every rep believes they can hit the top three on a good week. If a cohort is three people, there is no competition, just two losers. If a cohort is a hundred, every rep below rank twenty tunes out.
- List the natural cohorts on the floor: role, segment, tenure, territory
- Group into cohorts of roughly eight to twenty reps for a competitive ranking
- Carve out a ramp cohort for new hires in their first two quarters so they are not benchmarked against veterans
- Publish each cohort as its own tab or screen rotation so no rep sees themselves against the wrong peer group
-
4
Set the cadence: daily race, weekly ranking, quarterly story
Cadence is where most leaderboards drift. Daily rank changes keep competitive energy high and reward reps who respond to momentum, but they also whipsaw mid-cycle reps whose deals land in chunks. Weekly rank is calmer, more fair, and matches the rhythm of pipeline reviews. The best designs run both at once. Show today-in-progress live on the floor TV so reps can see the race, show week-to-date and month-to-date rank in Slack so the ranking feels earned, and run a quarterly recognition moment at the end so the top finishers get a real reward. Three cadences, one metric, zero confusion. Reset at known boundaries, never mid-period.
- Run a live daily race for in-the-moment energy on the floor display
- Lock a weekly ranking in Slack every Monday morning for the official standing
- Roll weekly ranks into a quarterly recognition moment with real stakes
- Only reset at calendar boundaries, never mid-quarter, so trust in the ranking compounds
Tip: If a rep closes a big deal on a Friday and the Monday ranking does not reflect it, the rep stops trusting the board. Automate the refresh, do not run it by hand.
-
5
Build in guardrails against gaming and perverse incentives
Every public leaderboard will be gamed. Design the guardrails before launch, not after the first scandal. Three guardrails cover most cases. First, rank on booked revenue net of churn or chargeback inside the ranking period so reps cannot inflate the number with deals that fall out. Second, require a minimum stage-progression threshold before a deal counts toward pipeline-sourced rankings so SDRs do not stuff the pipeline with unqualified logos. Third, exclude renewals from new-business leaderboards so account managers do not look like heroes for routine work. Add a tiebreaker rule (usually average deal size or cycle time) so ties resolve without manager arbitration. Publish the guardrails. Reps will police each other once they see the rules are real.
- Net the ranking metric against churn, chargeback, or fallout in the ranking period
- Require a minimum stage threshold before pipeline counts toward prospector rankings
- Exclude renewals, expansions, or non-net-new revenue from new-business leaderboards
- Pick a published tiebreaker so no manager has to adjudicate close races
Tip: The clearest signal that your leaderboard is being gamed is a cluster of deals that close just before the Friday snapshot and reverse the next week. Watch the Monday reconciliation.
-
6
Design the recognition and the dignity floor
A leaderboard is a motivation design, not a punishment design. The top three slots do most of the motivational work, so put real recognition there: a visible call-out in the Monday meeting, a small prize that reads as status rather than cash, and a quarterly award that goes on the wall. Just as important is the dignity floor. The bottom third of the board is the demoralization risk. Design for it. Only publish the top ten rows of each cohort by name, show the rest as aggregate counts, and never publish an ordinal rank below the median. A rep who sees themselves ranked eighteenth out of twenty every Monday does not work harder, they look for another job. The goal is to celebrate the top, not to punish the bottom.
- Call out the top three in the Monday standup with a specific line about what they did
- Attach a quarterly award with real stakes to the number-one slot in each cohort
- Publish by name only above the median; show the rest as aggregate counts
- Keep ramp cohorts off the public board entirely until they graduate
Tip: If reps start trading tips about how to game the ranking, that is a healthy sign. If reps start hiding from the board or dreading Monday, the dignity floor is broken. Fix the floor before you tune anything else.
-
7
Ship it to Slack and the floor TV, not a dashboard
A leaderboard that lives inside a dashboard behind two clicks is not a leaderboard. The whole point is ambient visibility. Push the ranking to the surfaces where reps already spend time. Pin the weekly ranking to a dedicated Slack channel every Monday morning and auto-post deal-closed events into the channel with the rep name and the amount so the race feels alive. Put the live daily race on a floor TV and a team-room screen with large type, rotating cohort tabs, and no interactivity. Format for glanceability: rank, name, metric, delta from yesterday. If a rep has to open a tab to see the board, the board is dead. Instrument whether people are actually looking: Slack reaction counts and TV time-on-screen tell you more than you think.
- Auto-post the weekly ranking to a dedicated Slack channel every Monday morning
- Pipe deal-closed events into the channel in real time with rep, amount, and segment
- Mount a floor TV with the live race and rotating cohort tabs; no mouse, no clicks
- Format for glanceability: rank, name, metric, delta, and nothing else
-
8
Retire or refresh the leaderboard every quarter
Leaderboards decay. A board that runs on the same metric for a year stops moving behavior because reps have optimized around it. Treat the leaderboard as a quarterly instrument tied to the quarterly plan. At the end of each quarter, run a short retro: did the metric move the outcome, did the top finishers map to the real top performers, did any cohort check out, did any gaming behavior emerge. Based on the answer, keep the metric, rotate to a new one, or sunset the board for a quarter and run something different. Rotation is not failure, it is design. The sales leader owns the retire-or-refresh decision and signs it off before the new quarter starts.
- Run a quarter-end retro on the leaderboard with the sales leader and two front-line managers
- Check whether the metric moved the actual quarterly outcome or just created motion
- Rotate the metric, resize the cohorts, or sunset the board based on what the retro surfaces
- Lock the next quarter design before the quarter opens so there is no ambiguous first week
Tip: Keep a short written log of every leaderboard quarter: metric, cohort shape, winners, and what the retro surfaced. Six quarters in, the log will tell you which designs actually produce results on your floor.