Walk into almost any commercial MEP shop doing $10 million or more, and you'll find some version of the same scene.
There's a monitor somewhere. A dashboard on the screen. Numbers populating in real time. And a leadership team that stopped looking at it sometime around Q3.
Nobody turned it off. It still runs. It just stopped meaning anything.
The technology is fine. Great even. But the software implementation missed one critical piece: what to do with all the data. The decisions that actually move a business, whether to hire, how to price, which service lines to grow, are still getting made on gut. Even when the data is right there, in plain view, on a monitor no one looks at.
What goes wrong (It’s not just you)
Most dashboards are built by someone who finally got the data connected and wants to show it all. That instinct makes sense. You've got job costing tied to field ops. Technician productivity pulling automatically. Service agreement renewal rate sitting right there. Why not surface everything?
Because “everything” is an optimistic way of saying “nothing”.
When a dashboard tries to answer every question, it answers none of them fast enough to matter. Leadership opens it, scans 24 tiles of numbers, and closes it. Nothing changes.
The contractors who've actually cracked this don't build dashboards that report on the business. They build dashboards that tell them what to do next. Those are two very different things, and the gap between them is where most operators are quietly losing ground.
A reporting dashboard shows you what happened. A decision dashboard tells you something's off before the month is over. One captures the past. The other changes what you do next. And saves your margin in the meantime.
No participation trophies: Some metrics just don’t matter
No two businesses will run identical dashboards. But after enough time in commercial MEP shops across HVAC, plumbing, and electrical, the same five categories keep separating contractors who use their data from those who just collect it.
Revenue per technician. Total revenue tells you how big the business is. Revenue per technician tells you how efficiently it's running. Best-in-class commercial HVAC operators are typically running $350,000 to $500,000 per tech annually. If you're well below that, you don’t need to hire more technicians; you need to figure out how to optimize the team you have.
Service agreement attach rate and renewal rate. For commercial contractors, service agreements are working capital. Predictable, recurring revenue that funds payroll between project cycles and gives you a floor when the installation market softens.
Most operators who struggle with these numbers share the same problem: neither metric is visible to the people who can actually influence it.
Service managers need to know every time a tech closes a PM visit without presenting a renewal. When your team has that information as soon as it happens, they’re able to capitalize on the very next opportunity, instead of just surveying weeks of missed revenue potential at the end of the month.
Job costing variance. Winning a job at the right price means nothing if your actual costs routinely exceed the estimate. Track the gap between what you bid and what you spend — consistently, by project type — and you'll find the same patterns every time. Certain crews. Certain geographies. Certain service categories where you're systematically off.
Is your dashboard telling you what’s eating away at your margins? Or is it hiding costs that won’t show up until the end of the year?
Technician utilization and capacity. You know your technicians are busy; you don’t need data to prove it. But are they busy with the right work? In the right sequence? With the right materials in hand before they get there? Billable hours as a percentage of available hours, broken down by service line and crew, give dispatchers the information to make real-time adjustments. You're not chasing 100%. You're understanding where slack exists and why.
Invoice aging. Commercial contractors extend credit to property managers, facilities teams, and GCs. That's the business. But how long it takes for completed work to convert into collected cash is where many enterprise operators are bleeding working capital.
A decision dashboard makes aging visible by account, service line, and bucket — 30, 60, 90+ days. It tells the AR team where to apply pressure this week to stop the bleed, rather than at the end of the quarter.
Build a dashboard that actually matters
The technical side matters less than the design process, and most dashboards fail before a single metric gets configured.
The shift from the “What data do we have?” to the “What decisions do we need to make?" mindset generates two very different dashboards.
Start with your operations leadership. Ask them: What are the five to eight decisions you make every week that most directly affect the bottom line? Hiring decisions. Pricing decisions. Dispatch sequencing. Service line investments. Those decisions — and only those — should drive what goes on the dashboard.
Then work backward. For each decision, ask what you'd need to know to make it with confidence rather than instinct. That's how you get to the right metrics. Not by pulling everything you can measure, but by starting with what you're actually trying to decide.
Then build for roles, not for everyone. The owner doesn't need the same view as the service manager. The service manager doesn't need the same view as the dispatcher. One enterprise-wide dashboard that everyone shares is almost always a compromise that serves no one.
Three ways everyone messes this up
They optimize for completeness. Every metric has someone who cares about it and will push to include it. The result is a dashboard that grows until it's comprehensive and useless. A decision dashboard with seven sharp metrics beats a reporting dashboard with thirty. When in doubt, cut. Even if you’re not in doubt, still cut.
They build the dashboard without building the behavior. A dashboard nobody looks at is just expensive infrastructure. Or a place to drink coffee so you look productive. The operators who get real value from this build the review cadence alongside the dashboard. A 15-minute Monday morning operations meeting. A weekly service manager check-in. The dashboard changes behavior only when looking at it is a scheduled, expected part of how the business runs.
They treat it as finished. Business priorities shift. New service lines create new metric needs. An acquisition changes what the GM needs to see. Build with the expectation it'll look different in 12 months, and put a quarterly review on the calendar to ask: are these still the right metrics?
The unified view no one wants to build by hand
Decision dashboards built on disconnected systems are a maintenance burden. When job costing lives in one place, service agreements in another, and AR aging in a third, producing a unified view takes manual exports, spreadsheet work, and time — which means it doesn't happen consistently, and when things move fast, you're already behind.
Market-leading commercial MEP contractors do this by consolidating on a single platform. When dispatch, job costing, service agreements, invoicing, and technician performance all share the same source of truth, the dashboard becomes a live view of the business, not a quarterly snapshot.
ServiceTitan is built for exactly this: real-time visibility into the metrics that drive decisions, not just the ones that describe outcomes. Technician performance scorecards. Service agreement dashboards. Job costing variance by project type. The leading indicators that matter to growing, profitable MEP businesses — surfaced when you need them, not after the month closes.
The #1 newsletter for the trades.
The data is already there, so the decisions are all yours
The commercial MEP contractors who pull away from the pack over the next decade won't do so by hiring larger analytics teams or running more complex operations.
They're going to do it by making a simpler, more disciplined choice: figure out the decisions that drive the business, then build the visibility to make those decisions well.
The data is already there. Are you using it? Or just watching it populate on a screen?
