Every company has two org charts. There's the official one — boxes, lines, reporting structures, lovingly maintained in a slide nobody looks at. And there's the real one: who actually helps whom, who unblocks whom, whose work quietly makes everyone else's possible. The official chart is a theory of how the company works. The real one is how it works. And here's the fun part: your peer recognition data is one of the few honest windows into the real one — which is why cross-team recognition turns out to be such a sharp read on your collaboration health.
The logic is simple. Nobody thanks a coworker on another team out of politeness. A cross-team thank-you means something concrete happened at a boundary: a handoff worked, a favor was done, an incident got solved by two teams instead of one. Every one of those messages is a timestamped receipt of collaboration that actually occurred. Map them for a quarter and you get something no survey can give you: a live picture of which boundaries in your company are alive — and which have quietly gone dark.
Recognition Flows Are a Collaboration Map
Think of each team as a node and each piece of recognition as an arrow: who sent it, who received it, whether it crossed a team line. Inside-team arrows are the baseline — teammates see each other's work daily, so they should dominate. The interesting arrows are the ones that cross. Cross-team recognition is rarer, more deliberate, and far more informative, because it only happens when work genuinely flowed across a boundary and went well enough that someone said so.
That last clause matters. Cross-team recognition isn't measuring whether teams talk to each other — Slack traffic can tell you that, and it can't distinguish collaboration from arguing. It's measuring whether the collaboration produced gratitude. A boundary where lots of work crosses but zero appreciation does is arguably in worse shape than one with no traffic at all: those are your grinding, resentful handoffs, the ones people describe with a sigh.
So the map answers three questions at once: where is collaboration happening (arrows exist), is it healthy (arrows flow both directions), and where is it missing (boundaries with dependent teams and no arrows at all).
What a Healthy Map Looks Like
Before hunting for problems, know what good looks like. A healthy cross-team recognition map has three properties:
- Cross-team share in the 15–30% neighborhood. Most recognition should stay in-team — that's normal and fine. But if fewer than roughly one in ten thank-yous ever crosses a boundary, your teams are functioning as separate companies that share a logo.
- Reciprocity at the edges. Recognition between two teams should flow both ways over a quarter. It won't be perfectly balanced — service teams naturally receive more — but a healthy edge is a conversation, not a monologue.
- Arrows that match your dependencies. Engineering and product ship together, sales and support share customers, design and marketing share launches. Wherever real work crosses a boundary, some recognition should follow it. Recognition tracing your dependency graph is the map saying: the handoffs work.
Four Patterns That Should Worry You
Now for the diagnostics. Four shapes show up over and over in unhealthy maps, and each one points at a different problem.
1. The Island
A team that gives and receives plenty of recognition internally, and almost none across any boundary. Islands aren't disengaged — internally they might be your most vibrant team — they're isolated. Sometimes that's structural (a genuinely self-contained function), but more often it's the early stage of a silo: the team has started solving everything in-house, routing around other departments rather than through them. Islands are where duplicated work and we-didn't-know-you-existed surprises come from. We dug into the silo mechanics — and how recognition helps dismantle them — in breaking down silos with recognition data.
2. The One-Way Street
Team A thanks Team B constantly; Team B has never once thanked Team A. One-way edges usually mean one of two things. Either Team B is a pure service function whose own upstream contributions are invisible (a recognition-culture problem worth fixing — invisible work is where blind spots breed), or the relationship is genuinely lopsided: A depends on B, B barely registers A's existence, and the handoff quality reflects it. Ask the people on the quiet end of the arrow which one it is. They'll know instantly.
3. The Single Bridge
Two teams whose entire cross-boundary recognition runs through one person. Every arrow between platform and product has the same name on one end. That person is doing heroic connective work — and they're also a single point of failure. When they take parental leave, change teams, or burn out, the boundary goes dark overnight. Single bridges deserve two responses: celebrate the bridge-builder loudly (they're doing work the org chart doesn't capture), and deliberately widen the bridge before you need to.
4. The Silent Border
Two teams with a heavy, obvious dependency — and zero recognition in either direction, ever. This is the most alarming pattern precisely because it's the easiest to miss: silence doesn't show up in dashboards unless you go looking. A silent border between dependent teams almost always means the handoff has degraded into tickets, escalations, and mutual grumbling. Nobody says thank you because, honestly, nobody feels thankful. Find these first. They're where your next big cross-functional failure is incubating.
How to Read Your Own Map
You don't need a network-science degree. A quarterly 30-minute review covers it:
- Pull a quarter of recognition data — sender, recipients, and each person's team. A month is too noisy; cross-team events are rarer, so give them room to accumulate.
- Compute the cross-team share, overall and per team. Rank teams by it.
- Sketch the edges. For each pair of teams that should collaborate, note: recognition in both directions, one direction, or none.
- Flag the four patterns. Islands, one-way streets, single bridges, silent borders. You'll usually find one or two of each.
- Validate with humans. The map tells you where to look, not what's true. A silent border might be a broken handoff — or two teams whose dependency ended two quarters ago. Ten minutes with each team lead sorts signal from noise.
That last step is the whole game. Recognition data is an early-warning instrument, not a verdict — the same principle behind making recognition data actionable: the data buys you a better question, and the conversation delivers the answer.
Moving the Map: Interventions That Work
Once you've validated a weak boundary, resist the urge to mandate gratitude ("please recognize the data team more" is how recognition dies). Instead, fix the conditions that make cross-team recognition natural:
- Create shared wins. Cross-team recognition follows cross-team work. Joint incident reviews, paired launches, and embedded rotations generate the collaboration that generates the thank-yous.
- Make the invisible team visible. If a service team never gets thanked, have them demo what they do at all-hands once a quarter. People can't appreciate work they can't see.
- Celebrate bridge-builders publicly. When leaders visibly reward the people connecting teams, they're broadcasting that boundary-crossing is valued work, not extracurricular niceness.
- Re-check next quarter. A boundary you intervened on should show new arrows within a quarter or two. If it doesn't, the problem is deeper than recognition — likely structure or incentives.
The payoff for getting this right isn't abstract. Broken collaboration is one of the frustrations that quietly pushes good people out the door — and the Work Institute finds about 3 in 4 voluntary departures are preventable, at a replacement cost of roughly 33% of salary on the conservative end. Healthy boundaries are cheaper than backfills.
Where Propsly Fits (Yes, Propsly Is Ours)
Everything above works with any recognition system that records who thanked whom — a spreadsheet will technically do. But since you're on our blog: Propsly was built with exactly this analysis in mind. It's a Slack-native recognition app — teammates give props with the /props command, every give lands in a public feed, and the free tier covers unlimited users with 200 props per person per month, so you get real cross-team data instead of a sample. The Pro tier ($50/month flat for the whole workspace) adds the analytics that make this article a dashboard instead of a weekend project — including team-level breakdowns of who recognizes whom — plus automated monthly rewards, one of which is literally called Bridge Builder: a gift card for the person who gave the most cross-team props that month. We built an award for the exact behavior this post argues is precious. That's how strongly we believe the boundaries are where culture is won or lost.
The Map Is Already There
Here's the thing about cross-team recognition as a collaboration-health signal: you don't have to build anything to start. If your team recognizes each other at all, the map already exists — it's just unread. Pull one quarter of data, sketch the arrows, and look for islands, one-way streets, single bridges, and silent borders. You'll learn more about how your company actually collaborates in thirty minutes than the org chart has told you in years. And if you'd like the map to draw itself, well — see the section above this one.