The screen said green. Every indicator on the operations dashboard was in the healthy band. Complaints down, resolution time inside target, attendance fine, training completed. It was the sort of dashboard that makes a leadership meeting feel calm. Then we went to the site, and the calm did not survive the car park.

The floor was not fine. A process had quietly broken weeks earlier. People had worked around it so smoothly that the workaround looked like normal. The numbers had not caught up. They were still reporting the world as it used to be.

This is the pattern we meet most often in operations. A dashboard that is green because it is measuring yesterday, while the floor has already moved on.

What a dashboard actually measures

A dashboard measures what someone chose to put on it. That is all. It is a set of numbers, drawn from a set of sources, refreshed on a schedule. It is a model of the operation, not the operation itself.

Between the floor and the screen there are several steps. Someone decides what to count. Someone decides where the number comes from. Someone decides how often it updates and what counts as green. Each step is a place where reality can get smoothed into something tidier.

By the time the number reaches the screen, it has been through several hands. Usually those hands are honest. Sometimes they are not. And often, no one is being dishonest at all, and the number is still wrong, because the thing it measures stopped matching the thing it is supposed to describe.

The lag between number and reality

Numbers lag. Some lag by minutes, some by months. A complaint that is resolved slowly shows up as an open ticket today and a good resolution time next quarter. A process that broke last week shows up in the numbers next month, if at all.

This lag is the dashboard's quiet weakness. It reports the past with confidence. It says little about the present. When things change quickly on the floor, the dashboard is often the last to know.

So a green dashboard is not a statement that all is well. It is a statement that all was well, as far as the last data cycle could see. Those are different claims, and the gap between them is where problems hide.

Metrics that flatter

Some numbers flatter more than others. We have learned to be suspicious of any metric that can improve without the underlying thing improving.

• Complaints down. Could mean fewer problems, or fewer people willing to speak.
• Resolution time inside target. Could mean faster fixes, or cases closed without really being resolved.
• Attendance high. Could mean people are present, not that they are working well.
• Training completed. Could mean people learned, or that they clicked through a module.
• Backlog clear. Could mean the work is done, or that it was quietly written off.

Every one of these can be true and green while the floor tells a different story. The number is not lying. It is answering a narrower question than the one you think you asked.

Why green hides red

There is a second force at work, and it is human. People know what the dashboard wants. When a number is watched, it gets managed. Not always dishonestly, but always with an eye on the screen.

A team that knows its resolution time is tracked will close tickets faster. A team that knows complaints are counted will find reasons a complaint is not really a complaint. A team that knows attendance matters will make sure attendance looks right. None of this requires bad intent. It requires only that the number became the goal.

This is how a dashboard goes green while the floor goes quiet. The workaround that keeps the number healthy becomes more important than the problem the number was meant to reveal.

Going to the floor

The only reliable way to check a dashboard is to leave it. Go to where the work happens. Watch it. Talk to the people doing it. Ask them what the numbers miss.

We look for three things. First, workarounds: where are people doing something different from the process, and why. Second, silence: what is not being reported, and who decided it did not count. Third, drift: what changed recently that the numbers have not noticed yet.

Almost every time, the floor knows. The people closest to the work can tell you what is broken before the dashboard can. They just were not asked, or their answer did not fit the format.

Making the signal honest

You do not fix this by deleting the dashboard. You fix it by keeping the dashboard honest. Pair every headline number with a floor check. Track not just outcomes but the workarounds that hint at trouble. Ask the people doing the work what the number is missing, and act on the answer.

Add a few metrics that are harder to game, ones that describe what actually happened rather than how quickly it was closed. And put a name against each one, so someone owns the difference between the number and the truth.

A dashboard is useful. It is also a summary, and every summary leaves something out. The danger is not that it leaves things out. The danger is that we stop looking for what it left out.

Green is a claim, not a fact

When the screen says green, it is making a claim about the world. Usually a careful one, based on the best data available at the time. But it is still a claim, and claims can be wrong.

So treat green as a question, not an answer. Ask what the number measures, what it might be missing, and when it was last true on the floor. If you cannot answer, the dashboard is not telling you the operation is healthy. It is telling you the operation looked healthy to the last cycle, and the floor may already disagree.

We do not trust the screen. We trust the screen and the floor together, and when they disagree, we go and find out which one is right.