Most construction dashboards are wallpaper. Somebody in the office builds a screen full of dials and pie charts, mounts a monitor in the trailer, and it becomes furniture. Nobody makes a decision off it. The trade partners never see it. Three weeks in, half the numbers are stale and everyone's back to the whiteboard and the group text.
A dashboard earns its keep only when it changes what you do next. That's the whole test. If you can look at a screen for fifteen seconds and know where to spend your morning — which constraint to chase, which sub to call, which area is about to stall — it's doing its job. If you have to think hard about what a widget is even telling you, it's decoration. Here's how to tell the difference, and what actually belongs on a superintendent's screen versus the stuff that just looks impressive in a demo.
What a Dashboard Is Actually For
The job of a dashboard is triage. On a busy jobsite you have more problems than hours, and you can't fix what you can't see. A good dashboard surfaces the two or three things that will hurt you this week and pushes everything else into the background. It answers one question fast: where do I need to intervene right now?
That means the best dashboards are ruthlessly short. When I see forty tiles on one screen, I know nobody uses it, because the human eye can't triage forty things. Pick the handful of indicators that predict trouble on your kind of work and let the rest live one click deeper. A dashboard is a smoke detector, not the whole fire report.
Schedule Health — Read It as a Trend, Not a Number
Every dashboard shows schedule status. The useless ones show a single percent-complete figure, which lies to you constantly — you can be "62% complete" and dead in the water because the remaining 38% is all on the critical path behind a submittal that hasn't come back. What you want instead is variance you can see moving: are we gaining or losing float week over week, and on which activities?
The layer that actually drives the week is your short-interval plan — the two-to-six-week look-ahead where the master schedule turns into named crews on named days. A schedule tile that tells you your look-ahead is 90% populated but your master schedule slipped four days last week is telling you the real story: near-term work is planned, but the finish date is bleeding. Watch both. The master schedule tells you if you'll make the milestone; the weekly work plan tells you if this week is real.
Milestones that mean something
Milestone tiles are only useful if the milestones are constraints you actually manage — permanent power, roof dry-in, elevator turnover, TCO. "Drywall complete" is not a milestone worth a dashboard slot unless drywall is gating something downstream. Keep the milestone list to the events that other trades and the owner are waiting on. Everything else is noise dressed up as progress.
Constraints — The One Screen I'd Keep If I Could Only Keep One
If your dashboard shows one thing well, make it the constraint log. Constraints are the reasons work can't start — missing material, an open RFI, incomplete predecessor work, no inspection sign-off, a sub who hasn't manned up. In pull planning and the Last Planner approach, clearing constraints ahead of the work is the entire game, and a dashboard that surfaces them by age is worth more than every safety chart combined.
Two numbers matter here. First, aging: a constraint that's been open eleven days on work scheduled to start Monday is a fire, and it should look like one on the screen. Second, ownership: every constraint needs a name and a date next to it. "Waiting on engineering" is not a constraint status; "Waiting on Priya, promised Thursday" is. Dashboards that let a constraint sit there anonymous and undated are how you get to a Monday coordination meeting and discover nobody owned the thing that just stopped forty guys.
A quick field rule I've used for years: any constraint on work inside your next two weeks that doesn't have a named owner and a commit date is escalated that same day. The dashboard's only job is to make those jump out. Tools built around look-ahead scheduling — LookAheadWall among them — tie the constraint straight to the activity it's blocking, so you're not cross-referencing a spreadsheet against a wall of sticky notes to figure out what's actually at risk.
Plan Reliability — PPC Is the Honest Number
Percent Plan Complete is the metric most crews avoid because it's the one that tells the truth. PPC is simple: of the tasks you committed to finishing this week, what fraction actually got done? Not "made progress on" — done, complete, off the list. A crew running 50% PPC is completing half of what it promises, which means every downstream trade is planning around commitments that fail one time in two.
Track it as a trend line, not a single week. One bad week is weather. Six weeks trending down is a planning problem — usually that you're committing to work that isn't actually ready because the constraints weren't cleared. The gold in PPC isn't the number; it's the reasons for the misses. A good dashboard captures the variance reason every time a committed task slips, and after a month those reasons cluster. When you see "prerequisite work not complete" show up as 40% of your misses, you've just found where your sequencing is broken, and no amount of yelling in the morning huddle would have told you that.
Break PPC out by trade and by area if you can. A jobwide 75% can hide a framing crew at 95% and a mechanical sub at 45% who's quietly wrecking your finish sequence. The average lies; the breakdown convicts.
Trade Flow and Sequencing — Where Location Beats a List
Bar charts are lousy at showing whether trades are stepping on each other. A Gantt tells you drywall runs three weeks; it doesn't tell you that on Wednesday the drywallers and the MEP rough-in crews are both scheduled in the same corridor. The failures that actually cost you money are collisions in space, not slips in time.
This is why a location-based, trade-flow view earns a dashboard slot on any project with repeatable areas — floors, wings, units, pours. When you can see the wave of trades moving through your building floor by floor, a stacked-up area is obvious before it becomes a stop-work. If two crews are colored into the same zone in the same week, you spotted the clash at planning instead of at 6:45 a.m. when both foremen are standing in the same room glaring at each other. Visual, location-aware planning is the whole reason to move off a spreadsheet, and it's a core part of how look-ahead tools like LookAheadWall lay out the week.
The Widgets That Look Great and Do Little
Some tiles demo beautifully and change nothing. Be honest about these:
- Live weather. You already know it's going to rain; you don't need a dashboard for the sky. Weather is only useful on a dashboard if it's tied to specific weather-sensitive activities — "pour scheduled Thursday, 60% rain" is actionable; a five-day forecast is a phone widget.
- Photo feeds. Great for the owner and for documentation. Nearly useless for deciding what to do next. Keep them, but don't confuse a wall of jobsite photos with a management tool.
- Vanity safety counters. "127 days without an incident" is a lagging number that tells you nothing about today's exposure. If safety's on your dashboard, make it leading — observations logged, near-misses, open corrective actions. Those predict; the day counter just congratulates.
- Document-count tiles that show "14 open RFIs" with no aging. Fourteen fresh RFIs is fine; one RFI open sixty days on critical-path work is a crisis. The count without the age is worse than nothing because it looks like information.
None of these are wrong to have. They're wrong to lead with. Put the decision-drivers up top and let the reference material live below the fold.
Drill-Down — The Difference Between a Report and a Tool
A summary you can't open is a dead end. When a tile shows red, you need to click it and land on the actual list — which constraints, which tasks, which sub, which area — without exporting anything or emailing the office. If your dashboard can't do that, it's a report someone printed, and it's already out of date. The whole value of software over a whiteboard is that the overview and the detail are the same live data, one click apart.
Mobile Isn't Optional
Superintendents don't sit at a desk, and foremen never do. If the only place to read the dashboard is a monitor in the trailer, the people making the day's decisions are flying blind out in the building. The metrics that matter in the field are a short list — today's committed work, open constraints on my area, anything red on my trades — and they need to render on a phone in the sun with gloves on. A crew leader should be able to pull up this week's plan for his area at the tailgate, not walk back to the office to find out what he's building today. That field-first access is exactly what a mobile companion is for.
Multi-Project Views — For the Office, Not the Trailer
If you're running one job, skip the portfolio dashboard. It's built for the operations manager comparing six projects, and on a single site it just adds a layer between you and your work. Where it earns its place is upstairs: someone overseeing multiple superintendents needs to see which of five jobs is trending down on PPC or piling up aged constraints, so they can send help before it's a claim. Right tool, wrong altitude for the field.
How to Judge a Dashboard Before You Trust It
Next time you're evaluating software, or auditing the dashboard you already have, run it through a few blunt questions:
- Can I tell in fifteen seconds what to do this morning? If not, there's too much on it.
- Does every alert have a name and a date attached, or does it just show a count?
- Can I click a red tile and land on the real list, live, right now?
- Does it show constraints and PPC, or just progress bars? Progress bars flatter; constraints and PPC tell the truth.
- Will it render on a phone in the field, and does it pull from the same plan the crews are actually working to?
A dashboard that passes those is worth mounting in the trailer and, more importantly, worth opening on your phone six times a day. One that doesn't is a screensaver with a budget line.
The honest version of all this: a dashboard doesn't manage the job — you do. What it can do is make the right problems loud and the noise quiet, so that when you walk the building you already know which room to walk to first. Keep it short, keep it live, tie it to the constraints and the weekly plan your crews are actually committed to, and it stops being wallpaper and starts saving you the one thing there's never enough of on a jobsite, which is time.