The team health check came back green. Trust, high. Communication, high. Psychological safety, high. Six weeks later the same team froze on a real decision, argued in the hallway instead of the room, and missed a deadline that mattered. The survey said the team was healthy. The work said otherwise.
This is the problem with most team health check templates. They measure how people feel about the team on a quiet Tuesday. They do not measure what the team actually does when the pressure is on. The signs of a dysfunctional team only surface when the stakes rise, and the gap between the calm-day survey and the real behavior is where projects go to die.
A team health check is worth running. You just have to check the right things, and you have to check them in conditions where the truth can show up. Here is a template that does both: the five dimensions that predict real team health, and the one change that turns a survey into a diagnostic you can trust.
Why Most Team Health Checks Lie to You
A typical team health check is a survey. People rate statements like "I trust my teammates" or "I feel safe speaking up" on a scale of one to five. You average the scores, color the dashboard, and call it a read on team health.
The problem is that you are measuring opinions in calm water. People answer based on how things feel when nothing is at stake. The Tuesday-afternoon version of a team is almost always healthier than the version that shows up when the timeline collapses and the budget gets cut. A survey cannot see the second version, because the second version only appears under pressure.
There is also a more subtle failure. People answer health surveys with the answer they wish were true, or the answer they think is expected. "Yes, we capture all our ideas." "Yes, we build on each other." Then you watch a real meeting and three good ideas die in the first ten minutes and nobody writes anything down. What a team says about itself and what a team does under pressure reveals everything — and they are rarely the same.
So the first rule of a useful team health check is this: measure behavior, not feelings. The five dimensions below are all behaviors you can observe, not opinions you have to take on faith. They are the characteristics of a high-performing team that actually predict performance, reframed as things you can watch instead of things a team claims.
The Team Health Check Template (5 Dimensions)
Check these five. Each one is a behavior, each one predicts performance under pressure, and each one is observable if you watch the right moment.
1. Decision-making. When this team faces a real decision with incomplete information, what happens? Do they decide and move, or do they stall, escalate, and revisit? Score it by watching an actual decision, not by asking how decisive they think they are. The tell: how many times the same decision comes back to the table. A decision that keeps returning is often a sign the team is not aligned in the first place — and most team alignment exercises miss it because they test agreement in a calm room, not a real call.
2. Idea flow. When someone offers a half-formed idea, does the team build on it or kill it? Most teams kill ideas reflexively with "yes, but." Healthy teams acknowledge the idea first, then build. The tell: count how many ideas in a meeting get a "yes, and" versus a "yes, but" or silence.
3. Context-sharing. Does information move to the people who need it, with enough context to act, or does it sit in someone's head until it is too late? People do not act on raw information. They act on context: why it matters, what is at stake, who it affects. The tell: how often someone says "I didn't know that mattered." This is the dimension that quietly sinks mixed teams — most cross-functional team challenges are broken hand-offs, not bad people.
4. Conflict. When two people disagree, does it happen in the room where it can be resolved, or in the hallway where it festers? Healthy conflict is visible and fast. Unhealthy conflict goes underground. The tell: the gap between what gets said in the meeting and what gets said after it.
5. Follow-through. Do the commitments made in the room actually happen, or does the team relitigate the same agreements next week? The tell: how many items on this week's list were also on last week's.
Rate each dimension red, yellow, or green based on observed behavior. Five greens is a team that holds together under pressure. Any red is a fracture line that a real crisis will find.
Copy This Team Health Check Template
Copy the scorecard below into a doc or a slide. Score each line right after you watch one real meeting, not after you ask the team how it feels. Use the same kind of meeting every time you run it, so the colors stay comparable.
1. Decision-making
- Question: Did the team make the decision on the table, or push it to next week?
- Green: decided in the meeting, owner named, nobody reopened it.
- Yellow: decided, and someone reopened it within a week.
- Red: no decision, or the same decision is back for the third time.
2. Idea flow
- Question: What happened to the first half-formed idea someone offered?
- Green: the team built on it before anyone challenged it.
- Yellow: it got a fair hearing and then quietly dropped.
- Red: it got a "yes, but" or silence in the first minute.
3. Context-sharing
- Question: Did anyone say "I didn't know that mattered"?
- Green: people came in knowing why the work mattered and who it affects.
- Yellow: the context arrived, late, from one person.
- Red: a key fact surfaced after the decision was made.
4. Conflict
- Question: Was the real disagreement said out loud in the room?
- Green: the disagreement was named, argued, and settled in the meeting.
- Yellow: it came up, and nobody pushed it to a decision.
- Red: the meeting was calm, and the hallway was not.
5. Follow-through
- Question: How many items on this week's list were also on last week's?
- Green: none, or each one has a new date and a reason.
- Yellow: one or two carried over with no reason given.
- Red: the list barely changed.
What to do with the score
- Five greens: run the check again next quarter, or sooner if the team changes.
- One or two reds: fix the red that shows up earliest in the work first.
- Three or more reds: the team needs more than a fix list. Start with what high-performing teams do differently when the stakes rise.
The One Change That Makes the Check Real
Here is the piece that separates a real team health check from a survey. You cannot score these five dimensions honestly by asking people. You have to put the team in a situation with enough pressure that the real behavior comes out, then watch.
That does not mean waiting for an actual crisis. By the time a real crisis hits, the cost of finding out is the project itself. It means creating a contained, high-pressure situation on purpose, watching how the team decides, builds, shares, fights, and follows through, then debriefing what you saw.
This is exactly what a team simulation does. In the Save the Titanic experience, a team becomes the senior officers on the ship in the minutes after the iceberg strikes. Limited time, limited resources, 2,200 lives on the line. In three and a half hours, all five dimensions show up in the open. You see who decides and who stalls. You see which ideas get built on and which get killed. You see whether context moves or gets hoarded. You cannot fake any of it under that kind of pressure, which is the whole point. The simulation is the health check, run in conditions where the truth has nowhere to hide.
Then you debrief. The debrief is where the read becomes useful, and most teams do it badly or skip it. A structured team debrief template turns what the team just did into a clear picture of where it is healthy and where it will crack. The pattern is simple: diagnostic, then pressure, then debrief. Survey the team to get the calm-water baseline. Run the simulation to get the under-pressure truth. Debrief to turn both into a plan.
How to Run Your Team Health Check
Start with the five dimensions. If you only have a meeting to work with, watch a real one and score what you see. Decision-making, idea flow, context, conflict, follow-through. Be honest about the colors. A yellow you call green is a fracture you chose not to see.
For a real read, add pressure on purpose. That is the difference between experiential team development and personality assessments: a personality test tells you who is in the room, and a pressure test tells you what the room does together when it counts. Run the simulation, watch the five dimensions, then debrief against them. How the experience works walks through the setup, the pressure, and the debrief.
Re-check on a cycle. Team health is not a fixed trait. It moves with new members, new pressure, and new stakes. A team that scored five greens in March can develop a red by June. The teams that stay healthy are the ones that check behavior regularly and fix the reds while they are still small.
What a Real Health Check Is Worth
The cost of a bad team health check is invisible until the crisis, and then it is enormous. A team that looked green on the survey freezes on the decision that mattered, and you find out their real health at the worst possible moment.
The teams that check behavior instead of opinions get the read early, while there is still time to act. At ArcelorMittal, 710 emerging leaders went through the experience via Duke Corporate Education, with a focus on health and safety and speaking up when something does not look right. Participants left committed to specific courageous leadership actions. That commitment did not come from a survey. It came from seeing their own behavior under pressure, then deciding what they would do differently.
A team health check template gives you the dimensions. Checking them under pressure gives you the truth. Put them together and you stop being surprised by your own team — because you found the fracture lines in a simulation, not in the crisis.
Read next: What Your Team Does Under Pressure Reveals Everything