Part of THREAD™, the operating lens inside the Minotaur Method™: six kinds of attention that complex work asks for. You enter where the work is actually stuck, not in a fixed order. This piece reads Evaluate.
Picture a leadership review at a Series-C SaaS company.
The VP of Engineering walks through the slides. Velocity is up. Burn-down looks healthy. Defect rate is steady. The trend lines all bend the right way. He’s been reporting these numbers every quarter for two years and the lines have always done the same thing.
The CEO is nodding politely. She’s stopped asking follow-up questions on this part of the deck. The VP notices that, and it bothers him, but he can’t quite say why.
She’s stopped asking because the numbers stopped meaning anything to her about a year ago. She just hasn’t said that out loud yet. The metric has gone decorative and nobody on either side of the table has named it.
This is the most common evaluation failure inside otherwise healthy companies.
A Metric Goes Decorative Quietly
Watch a metric over its life cycle.
It starts useful. Someone designs it to answer a question the team actually had. Are we shipping fast enough. Is the platform stable. Are customers staying. The metric gets reported. People look at it. It moves and they ask why. It moves the other way and they ask why again. The metric is doing its job. It is producing conversation.
A year passes. The team has gotten better at the work the metric was measuring. The number is consistently in the green zone. People stop asking why it moved. They stop asking why it didn’t move. The metric is still on the dashboard. Nobody mentions it.
Another year passes. The conditions underneath the metric have shifted. The team is bigger. The product is different. The customers are different. The metric is still measuring the same thing. The thing it’s measuring used to be the question. It is no longer the question.
The metric is now decorative. It produces a chart. It does not produce information.
A decorative metric is more dangerous than no metric at all. A team without the metric would notice the gap. A team with a decorative metric believes the gap is being measured, and stops looking.
The Test Is What You Would Do If the Number Changed
Take any standing metric on your dashboard. Ask two questions about it.
What would we do if this number doubled in a week.
What would we do if this number halved in a week.
If both answers are concrete and specific, the metric is alive. We would dig into the recent feature releases. We would call the top five customers. We would adjust the on-call rotation. Concrete answers mean the metric still has a connection to real moves the team would make.
If either answer is vague, the metric is drifting. I guess we would investigate. Probably look at it more closely. Vague answers mean the team has lost the bridge between the number and any action.
If both answers are vague, the metric is decorative. The team has been reporting it for so long that the act of reporting has become the work, and the connection to action has quietly broken.
Decorative metrics survive a long time. They’re easy to report. They never cause friction. They never spark hard conversations. The fact that they never spark hard conversations is exactly the problem. A metric that never produces friction is a metric the team no longer takes seriously, even if they still put it on the slide.
The dangerous part is not the metric itself. It is what the team is not looking at because they believe the metric covers it.
Every Standing Metric Should Have a Calibration Date
A leader who runs Evaluate well treats metrics the way a careful captain treats instruments.
A captain doesn’t assume the compass is right because it has always been right. He checks it against a known landmark periodically. If the compass has drifted, he resets it. If the compass is broken, he replaces it. The fact that he’s been using this compass for ten years does not give it permanent authority. It earns its place every voyage.
A metric earns its place the same way. Periodically, the leader asks whether the conditions that made the metric meaningful are still present. Whether the number is still connected to a real decision. Whether the act of measuring it is still producing thinking, or has become a ritual.
Most metrics deserve a calibration check every six months. Some deserve one every quarter. Some can run for a year between checks because the underlying conditions move slowly. The point is not the frequency. The point is that the metric is on the schedule at all.
A dashboard full of metrics that have never been recalibrated is not a dashboard. It is a museum exhibit of what the team used to believe was important.
Confident Wrongness Is the Failure Mode
Take the engineering team from the opening.
Their velocity metric was useful in year one. It captured the throughput of a small team shipping a v1 product. The number meant something. The team pushed against it and the product shipped.
In year three, the team is bigger. The product is more complex. The work is less about shipping new features and more about reducing the support burden of existing features. Velocity, as defined, no longer measures what the company cares about. The team is still high-velocity. The customers are still escalating. The CEO has stopped trusting the deck.
Nobody is wrong, exactly. The team is doing the work the metric measures. The metric is measuring the work the team is doing. The mismatch is between the metric and the actual question the business needs answered, and that mismatch has been growing quietly for eighteen months while the chart kept bending the right way.
This is confident wrongness. It is not lazy reporting. It is honest reporting against a question the company has outgrown. It is the most expensive kind of evaluation failure because it is invisible from inside the team that produces the numbers.
The fix is not better dashboards. The fix is a recurring practice of asking whether the dashboard is still pointed at the right questions.
What This Means in Your Week
Three moves before your next operating review.
- Print your current dashboard. For each metric, write the date it was first defined and the question it was originally designed to answer.
- For each metric, write the two-question test underneath it. What would we do if this number doubled in a week. What would we do if it halved. Be honest. If either answer is vague, mark the metric.
- For each marked metric, write one of three decisions next to it. Recalibrate the metric to match the current question. Replace it with a metric that does. Or retire it entirely and stop pretending the company is watching it. Pick one before the next review.
Where This Leads
A metric that has not been questioned is not a measurement. It is a ritual. The companies that get this right are not the ones with the cleanest dashboards. They are the ones who treat their metrics as instruments that earn their place each cycle.
The TORCH™ Brief is a thirty-item assessment that surfaces the pattern your decisions are actually being shaped by. About ten minutes. The result is yours.
Read across the lens: Target · Horizon · Resource · Evaluate · Act · Develop.
