Is Your Dashboard Reporting the News? How to Design Manufacturing KPIs That Actually Drive Action

A couple of weeks ago I sat in on a monthly leadership meeting where the COO pointed at a number on the dashboard and asked his engineering lead about the queue. Seventeen jobs sitting there which was nearly double the target.

She walked him through it. Ten of those seventeen lines came from a single customer order with multiple analytes. A few more were complex multi-part jobs that always take longer. The actual order queue was three.

By the end of the meeting, they had not fixed anything in production. They changed a column on the dashboard to start tracking orders instead of lines.

The KPI they originall picked was lying. Not on purpose, but it was telling the team a story that was not true, and the team was about to use their collective brainpower to act on it.

That moment captured the entire problem with most manufacturing dashboards, and it is the reason most weekly leadership meetings feel like theater. The metrics on the board are reporting the news. They are telling you what already happened. By the time you see them, you have already lost the chance to change the outcome.

The “reporting the news” test

When I am sitting in a leadership meeting and somebody pulls up a metric, I ask myself one question. Is this reporting the news?

If a number crosses a line and the only thing the team can do is talk about it, you are looking at a lagging indicator with no operational counterpart. You can review it, you can be frustrated by it. You cannot influence it. The cake is already baked.

If a number crosses a line and somebody knows exactly what to do tomorrow, you are looking at an actionable KPI. Same dashboard, different metric, completely different meeting.

Most manufacturing dashboards are stacked with metrics from the first bucket. Revenue, on-time delivery, win rate, customer satisfaction, cost of goods, etc. They all matter. They tell you something. But when you build your weekly leadership meeting around them, the conversation becomes an interrogation about results nobody can change in the next seven days.

The deeper version of leading and lagging indicators lives in this article if you want the foundational framework. What I want to talk about here is what happens when you build a leadership meeting around the wrong half of that distinction, and what to do about it.

Most of your dashboard is decoration

For years I have been sitting in leadership meetings as a fractional COO, and I have started to notice a pattern. The vast majority of meetings are theater. They are not designed to drive specific outcomes. They are scheduled because they have always been scheduled, populated with metrics because somebody has to put metrics in the deck, and they end without anybody being able to point at one decision that changed the trajectory of the next week.

If the meeting does not prompt action and the numbers do not move for the next cadence, the meeting is pointless. It could be an email or slack message or teams thread. Meetings cost money. They consume the most expensive hours your company produces, and the first question you should ask before scheduling any of them is whether the intended outcome can be achieved with a faster, cheaper channel.

But this article is not about meetings in general. It is about a specific kind of leadership meeting that exists in almost every low to mid-market manufacturer I work with. The weekly or biweekly leadership cadence built around a dashboard. The KPI review. The scorecard. The thing your CEO insists is essential and your operators dread.

The reason that meeting feels like theater is that most of the dashboard is decoration. The numbers are there for show. Nobody knows what to do with them when they move. So everybody discusses them, nods, and moves on. Next week, same numbers, slightly different positions, same discussion.

The fix is not to add more metrics. The fix is to apply one test to every metric you already have and start cutting.

The one question every KPI has to pass

Pick any metric on your weekly dashboard. Ask one question.

If this number crosses red, what does a specific person do tomorrow?

If the answer is “we discuss it next week,” the metric is decoration. Take it off the weekly board. Maybe it belongs on a monthly review where the cadence matches the metric’s response time. Maybe it belongs on a quarterly retrospective. Maybe it does not belong on a dashboard at all.

If the answer is “I increase outbound activity by twenty percent and shift Tay and Aileen to two additional blocks of prospecting,” the metric is actionable. Keep it.

That test cuts brutally. Most leadership dashboards I see lose half their metrics the first time a team runs this exercise honestly. People do not like cutting metrics because every metric has somebody who fought to put it there. But the cost of carrying decoration on your weekly board is that the meeting becomes about reviewing decoration. The signal gets buried in noise.

The COO who caught the engineering queue lying did not catch it because he was clever. He caught it because his team had already done the work of cutting decoration metrics. The dashboard was lean enough that when one number looked wrong, they actually dug in.

The leadership team that loaded four CRMs in five years

A client I have been talking to recently runs an industrial services business with active sales motions across multiple regions. The CEO has been trying to install a working sales process for years. He told me on a recent call that he has loaded four different CRMs in five or six years. Each one launched, ran for a few months, and died.

When I asked one of his senior leaders to walk me through their current process, he pulled out a notebook. Ten pages of handwritten notes from every meeting he has taken in the last quarter. Excel sheets with prospect names and site counts. Nowhere to put any of it. He kept saying “I want to be able to look at this daily and know exactly what’s happening.” He had no system that let him do that.

I asked the CEO what he wanted to see on a working dashboard. He paused, then said something that has stuck with me.

“I don’t even know what I want to see. That’s part of the problem.”

That sentence is the entire dysfunction of meeting theater in one line. He does not know what he wants to see because he has never had a working dashboard to push back against. So he keeps showing up to the weekly sales meeting, asking activity questions, getting activity answers, and watching the company miss revenue targets.

He is not lazy. He is not stupid. He is operating without a system that lets him separate signal from noise, and he is in good company. Most mid-market manufacturing CEOs I work with are in exactly the same spot. They feel the dysfunction. They cannot name what would replace it. So they keep meeting.

Why your leadership team cannot design the dashboard together

Here is the move that most teams get wrong, and it is the reason I think this stuff is so hard to fix from the inside.

When a leadership team realizes their dashboard is not working, the instinct is to put metric design on the agenda for the next meeting. Everyone comes to the table with ideas. The marketing lead proposes a customer satisfaction index. The ops lead proposes a throughput metric. The HR lead proposes a workforce productivity composite. The CEO listens, asks questions, and the room cycles through five or six alternatives without landing on anything.

I have been in that meeting more times than I can count. The conversation goes something like this. Somebody proposes a metric. Somebody else points out it would be hard to calculate weekly. A third person points out it could be gamed. A fourth person points out it might not even be the right thing to measure. By the end of the hour, the team agrees to “think about it” and revisit next week and nothing changes.

It is not that the team is incapable. It is that consensus design is the wrong process for this kind of work. Metric design is detail work. It requires somebody to sit alone with the data, draft a version, look at how it would behave week over week, and then iterate. You cannot do that in a meeting because as soon as you start the detail work, half the team gets bored and the other half starts proposing alternatives before the first version has been tested.

The leadership teams I have watched actually fix this do something different. They take the design work offline.

The version-five dashboard

One of my clients went through this exact transition over about five weeks. The first week, the leadership meeting was the cycling-through-ideas version. By the end of an hour, nobody had landed on metrics. The COO closed the meeting by asking everyone to “brainstorm your department’s key metric in preparation for next week.”

Between that meeting and the next one, the COO did something different. He went away and built a draft dashboard himself. He worked on it across multiple sessions, alone or with one other person as a thinking partner. By the time he came back to the next leadership meeting, he had what he called “version five.” He had iterated through five drafts in about a week.

Here is what he said when he walked into the next meeting. “This is version five. I really focused on leading indicators and weekly actionable things.” Then he walked through it.

The room did something different. Instead of proposing alternatives, the team started refining. The COO realized that the win-rate metric was useless weekly because a single quote could swing it from twenty percent to ninety-four percent. He swapped it to a ninety-day rolling window. The team agreed.

The next week, they refined more. Two weeks after that, the engineering queue caught itself lying, the team revised to an orders column, and the meeting shifted from cycling-through-ideas to actively diagnosing a misleading metric in real time. Three weeks after that, the COO told me it was the first time in the team’s history they had run consecutive weeks without changing the dashboard. They were not avoiding change, they had built something they trusted enough to leave alone.

The unblock was not a brilliant insight, it was a structural move. The design work happened between meetings, with one owner and one thinking partner. The team meeting became where the draft got stress-tested, not where it got invented.

What to do Monday morning

Pick one metric on your weekly leadership dashboard. Just one. Apply the question.

If this number crosses red, what does a specific person do tomorrow?

If the answer is “we discuss it next week,” you have your starting point. Replace it with the behavior underneath. If your scoreboard metric is win rate, the behavior underneath might be the number of new conversations your BD team started that week. If your scoreboard metric is revenue, the behavior underneath might be the number of meetings booked or quotes touched. If your scoreboard metric is on-time delivery, the behavior underneath might be the number of jobs released to the floor by Wednesday.

You do not have to design the new metric in a meeting. In fact, do not. Take it offline. Sit with the data. Build version one. Look at how it would have behaved over the last four weeks. If it would have caused you to act, it is a candidate. If it would have been just as quiet as the metric it replaces, try again.

Then bring it to your team as a draft. Use that exact word. Draft. Not the answer, not the new policy, just a version one to refine together. Give yourself permission to iterate. Tell the team you expect to revise the metric three or four times before it settles. That permission is what frees the team from feeling like they have to land on the perfect metric in the first conversation.

And one more thing. Watch for the gaming risk before you commit. Any time a metric goes on the board, somebody on your team will figure out how to influence it. That is not a bug, that is a feature. The question is whether the behavior they adopt to influence the metric is the behavior you actually wanted. If you put “number of quotes issued” on the board, you may get more quotes for things that should have been a quick price check. If you put “booked meetings” on the board, you may get meetings that did not need to happen. The fix is to track behaviors and outcomes together. If your behavior metric goes up but your outcome metric flatlines, you have a gaming problem. If both move together, you have alignment.

That is the work. It is iterative. It takes weeks, not hours, and the leadership teams that actually do it stop running weekly meetings that feel like theater and start running meetings where one number crossing a line tells one person what to do tomorrow. The rest of the time gets shorter, sharper, and a lot more honest.

That is what a working dashboard buys you. Not better numbers but a different conversation that prompts action.

That is it for today.

See you all again next week!

Dave