Every time someone thanks a teammate in public, they leave behind a tiny data point: who appreciated whom, for what, and when. One of those data points is a nice moment. A few thousand of them are a live map of how your organization actually works — who's carrying teams, which groups talk to each other, and who's quietly fading from view. That map is the heart of this recognition data playbook: seven signals every people leader should track, why each one matters, and what to do when one of them moves.
Here's what makes recognition data special compared to almost every other people metric: nobody is performing for it. Surveys capture what people are willing to say. Recognition captures what people actually noticed and appreciated, in the flow of work, with no HR prompt in sight. It's behavioral, it's continuous, and it moves before the lagging indicators — engagement scores, regrettable attrition — catch up. Given that Gallup and Workhuman found inadequately recognized employees are about twice as likely to say they'll quit within a year, ignoring this data is like owning a smoke detector and never putting in the battery.
A monthly review of these seven signals takes about thirty minutes. Here's the playbook.
Signal 1: Participation Rate
What it is: the percentage of your people who gave at least one piece of recognition this month.
This is your program's pulse. Everything else on this list is noise if only 15% of the company is participating — you're not measuring the organization, you're measuring the enthusiasts. Healthy programs typically see half or more of employees giving monthly; if you want reference points, we've collected them in our recognition program benchmarks.
Watch the trend more than the absolute number. Participation that slides three months in a row is a program problem (friction, forgetting, fading sponsorship) long before it's a culture problem. The fix is usually mechanical: make giving easier, make it visible, and make leaders go first. Recognition that lives where people already work — a /props command in Slack beats yet another portal nobody logs into — removes most of the friction on its own.
Signal 2: Reach
What it is: the percentage of people who received at least one recognition this month.
Participation tells you who's giving; reach tells you who's being seen. And the gap between them is where the pain lives. A team can have great participation and terrible reach — twenty people enthusiastically recognizing the same five stars while everyone else watches from the bleachers. The watchers are the ones your engagement survey will flag in six months and your exit interviews will explain in twelve.
Pull the list of people who received zero recognition this month and read it like a manager would. Some names are noise (new hire, someone on leave). Some are your quiet infrastructure people — internal tools, ops, compliance — whose work is only visible when it breaks. Those people deserve a deliberate spotlight, not an algorithmic shrug.
Signal 3: Equity of Distribution
What it is: how concentrated recognition is — what share of all recognition goes to the top 10% of recipients versus everyone else.
Some concentration is natural; your best people really do earn more shout-outs. But when a small group absorbs the overwhelming majority of recognition month after month, you've built a highlight reel, not a culture — and everyone outside the reel learns that recognition isn't for people like them. Cut the data by team, tenure, seniority, and location too: hybrid teams routinely discover that remote folks receive systematically less recognition for identical work, which is exactly the kind of invisible unfairness that curdles into disengagement.
We've written a full guide to diagnosing and fixing recognition inequality, but the short version: don't fix it by rationing praise for stars. Fix it by expanding what counts as recognizable — glue work, mentorship, unglamorous reliability — and prompting the people who never give.
Signal 4: Cross-Team Flow
What it is: the share of recognition that crosses team or department boundaries.
This might be the most underrated line in the entire playbook, because it's really a collaboration metric wearing a recognition costume. People thank colleagues on other teams for one reason only: they actually worked together and it actually went well. When cross-team recognition between, say, engineering and support flows steadily, that boundary is healthy. When it flatlines — or was never there — you've found a silo, mapped for free, without a single consultant workshop.
Track the ratio monthly and watch specific team pairs. A boundary that goes quiet right after a reorg or a rough shared project is telling you where relationships got damaged. We dig into reading these patterns in breaking down silos with recognition data.
Signal 5: Individual Trend Drops
What it is: individuals whose recognition activity — given or received — has fallen sharply versus their own baseline.
This is the early-warning signal, and the key word is baseline. The absolute number doesn't matter; the delta does. Someone who gave props eight times a month for a year and has now been silent for six weeks has changed, and disengagement is one of the most common reasons. People who are checking out stop noticing others first — appreciation requires attention, and attention is the first thing disengagement takes. The receiving side matters too: a steady performer whose inbound recognition dries up may have been moved to invisible work, or lost the collaborators who saw them.
A trend drop is a conversation prompt, never a verdict — sometimes the story is a heads-down project or a house move. But given that the Work Institute finds about 3 in 4 voluntary departures are preventable, a free, automatic prompt to check in early is absurdly good value. This signal is the entire premise of detecting quiet quitting with recognition data.
Signal 6: Tag and Theme Patterns
What it is: what people are recognized for — the hashtags and themes attached to recognition.
If the first five signals are about volume and flow, this one is about content, and it answers a question executives pay survey vendors handsomely to approximate: which values does the culture actually practice? If recognition requires a tag (in Propsly, every props ends with a hashtag), the aggregate is a real-time values audit. Lots of #teamwork and #customer-love, but your official value of #innovation appears twice a quarter? The poster in the lobby and the culture on the floor have diverged — and now you know in which direction.
Theme data is also operational intelligence. A spike in #firefighting or incident-related thanks is genuinely heartwarming and also a flashing sign that some system — technical or organizational — is burning people's weeks. Celebrate the heroes, then fix the fire.
Signal 7: Streaks and Consistency
What it is: how many people give recognition consistently — week after week — rather than in occasional bursts.
Two programs can post identical monthly totals and be in completely different health. In one, recognition spikes after the all-hands reminder and dies within days; in the other, a growing core gives every single week without being asked. The second program has something the first doesn't: a habit. Habits survive busy quarters, leadership changes, and the long stretch between kickoff enthusiasm and actual culture. Bursts don't.
So track your consistent-giver count as a leading indicator of durability, and celebrate the streaks themselves — recognition for recognizing is not too meta, it's reinforcement. (Propsly tracks weekly giving streaks automatically and celebrates milestones in the feed, because a 26-week streak deserves confetti as much as any shipped feature.)
Running the Monthly Review
Seven signals, one meeting. A cadence that works:
- Health check (5 min): participation, reach, and consistency versus last month. Trending up, flat, or down?
- Fairness check (10 min): equity and cross-team flow. Who's invisible? Which boundary went quiet?
- People check (10 min): individual trend drops. Route each flagged name to the relevant manager as a nudge to check in — never as a scorecard.
- Culture check (5 min): tag themes. Do this month's tags look like your stated values?
Two rules keep the whole thing honest. First, every review ends with at least one action — a manager conversation, a spotlight on an invisible team, a friction fix — because a dashboard that never changes anyone's Tuesday is a screensaver (more on that in making recognition data actionable). Second, this data informs conversations; it never becomes a quota. The moment recognition counts show up in performance reviews, people will game them, and every signal above turns to mush.
Getting the Data in the First Place
None of this works without a recognition system people actually use — frictionless enough for healthy participation, structured enough (tags, timestamps, sender and recipient) to be analyzable. That's precisely what we built Propsly to be — yes, Propsly is ours, bias fully disclosed. The free tier gives every team unlimited users, 200 props per person per month, leaderboards, and a public recognition feed — enough to start generating several of these signals on day one. The Pro tier ($50/month flat for the whole workspace) adds the advanced analytics that automate the rest: distribution and concentration views, reciprocity, team breakdowns, engagement gaps, and tag stats.
Companies with strong recognition cultures see up to 31% lower voluntary turnover, per Deloitte — but "strong" isn't a vibe, it's something you verify. These seven signals are how. For more on turning people data into decisions, the Propsly blog covers everything from retention math to program design.