Coordinating subs is most of the job. You can have a beautiful CPM schedule hanging in the trailer, a clean set of drawings, and a super who knows the work cold — and still lose two weeks because the drywallers showed up to a wall that framing never signed off on, or the tile guy's crew got sent home because the mud set wasn't dry. The schedule is the plan. Coordination is what makes the plan survive contact with a dozen companies who don't work for you, don't answer to you, and each have three other jobs going.
This is the part nobody teaches you. You learn it by getting burned. What follows is the stuff I wish somebody had handed me twenty years ago — how to actually run the trades on a jobsite so the work flows instead of stacking up.
Coordination Is a Sequence Problem First
Before you worry about relationships or communication or any of the soft stuff, understand that most trade conflicts are really sequencing conflicts wearing a costume. Two crews want the same room on the same day because whoever built the plan never mapped the handoff between them.
Every trade has predecessors and successors. Rough-in follows layout. Insulation follows rough-in inspection. Drywall follows insulation. It sounds obvious written down, but on a live job with forty rooms at different stages, the sequence gets muddy fast. The single most useful thing you can do is think in terms of trade flow — the chain of who hands off to whom, in what order, in each physical location. When you build the sequence by location instead of by trade, the conflicts show up on paper before they show up on the floor.
This is exactly where a location-based look-ahead beats a bar-chart schedule. A Gantt chart tells you drywall runs weeks 6 through 9. It does not tell you that Unit 204 is ready for drywall while Unit 206 is still waiting on a plumbing correction. Short-interval scheduling — the weekly work plan, the three-to-six-week look-ahead — is where you get granular enough to catch the collision. Tools like LookAheadWall exist precisely to make those location-by-location handoffs visible, but the discipline matters more than the tool: if you're not planning by area, you're planning blind.
The Weekly Sub Meeting Earns Its Keep — If You Run It Right
The best coordination mechanism I've ever used is a tight weekly foreman huddle, and the worst waste of everyone's morning is that same meeting run badly. The difference is whether you make commitments or just read status.
Here's the format that works. Pull up the next two weeks by area. Go trade by trade and ask each foreman one question: "What are you doing next week, and what has to be true for you to do it?" That second half is the whole game. When the electrician says "I'm rough-ing in the north corridor Tuesday," your follow-up is "Is framing done and inspected up there?" If the answer is no, you just found next week's blowup a week early, in a warm trailer, with time to fix it.
Make people commit out loud, in front of the other trades. A foreman who tells the room he'll be done with the second floor by Thursday is far more likely to be done than one who got an email. Peer accountability does work no clipboard can. And write the commitments down where everyone can see them next week, because next week you're going to hold them to it.
Constraints Kill Schedules — Hunt Them Early
A task is constrained when something has to happen before a crew can productively work it: a missing submittal, material not on site, an RFI still open, prior work not inspected, equipment not available, an area not released. The whole point of look-ahead scheduling is to pull work forward far enough that you find those constraints while there's still time to clear them.
Run a constraint check on everything two to three weeks out. For each upcoming task ask:
- Predecessor complete and inspected? Not "framing's basically done" — done, signed, and released.
- Material on site or confirmed to arrive before the crew does? Long-lead items especially — switchgear, custom glass, elevator equipment will wreck a schedule if you assume instead of confirm.
- Design questions answered? An open RFI on a detail means the crew builds it wrong, or builds nothing.
- Access and area released? Is the space actually available, or is another trade still in it?
- Manpower confirmed? Did the sub commit the crew size the work actually needs?
A task that clears all five is ready. Only ready work goes on the weekly work plan. This is the core idea behind the Last Planner approach and it's the single biggest lever you have: stop scheduling work that can't actually be done, and your reliability jumps. Crews stop showing up to walls they can't build, and they start trusting the plan — which is worth more than any software feature.
Confirm Manpower Before You Believe It
The most common lie on a jobsite isn't malicious, it's optimistic. "Yeah, I'll have six guys Monday." Then Monday you get two, because the sub's other job called an emergency and you were the softer deadline.
Protect yourself with specifics. Don't accept "I'll have a crew." Ask how many, doing what, in which area. Then look at the quantity of work and do the math yourself — if hanging that floor is a four-man, three-day job and he's promising two guys for two days, the numbers don't close and you both know it before he no-shows. When a sub is chronically light, escalate to the PM early with documentation, not on the day it hurts. Confirming manpower against the actual scope, week over week, is dull work. It also prevents about half the fires you'd otherwise be fighting.
The Handoffs Are Where Money Leaks
The riskiest moment in any trade sequence is the handoff — one crew finishing, the next starting. That's where quality problems get buried inside a wall and where a day of buffer saves you a week of rework.
A few handoffs worth building explicit buffer around:
- Rough-in to cover-up. Don't let insulation or drywall chase MEP rough-in the same day. Build a 1–2 day buffer for cleanup, correction, and inspection. Megger the electrical runs and pressure-test the plumbing before anything closes the wall — you cannot un-hang drywall cheaply.
- Concrete and wet trades to follow-on work. Respect cure and dry times, and don't just take the manufacturer's minimum on faith in a cold or humid building. Flooring over a slab that's still giving off moisture is a callback with your name on it. Meter it.
- Framing to drywall. Walk it with both foremen. Backing in place, blocking for grab bars and casework, framing inspection signed. Once the rock goes up, a missing piece of backing is a demo job.
- Finishes into finished space. Once an area is painted and trimmed, every trade that re-enters is a scratch-and-dent risk. Sequence so you're not sending crews back through completed work, and when you must, make protection a condition of access.
The rule of thumb: never let a successor start the instant the predecessor "finishes." Leave room to inspect and correct. That gap feels like lost time to an office scheduler and it's actually the cheapest insurance on the job.
Access, Laydown, and the Fight Over Space
On a tight site, physical space becomes a scheduling constraint as real as any predecessor. Two trades can be perfectly sequenced on paper and still collide in a stairwell, a hoist, or a laydown yard.
Coordinate the shared resources explicitly: crane and hoist time, the loading dock, elevator access for material, laydown areas by trade. If the mason needs the hoist all morning to load block, the glazier planning to fly panels up at nine needs to know now, not at nine. Put major equipment moves and big deliveries on the look-ahead the same way you put labor — because a delivery landing in the wrong spot on the wrong day shuts down whatever was supposed to happen there.
Communicate the Plan So Nobody Can Claim They Didn't Know
Coordination falls apart at the information gap between what you know and what the foreman in the field knows. The schedule that lives only in your head or only in the trailer isn't coordinating anything.
Get the current weekly plan into the hands of every foreman, and get it to them in a form they'll actually look at — on their phone at 6 a.m., not a PDF buried in an email thread. This is one place field software genuinely pulls its weight: a shared, always-current look-ahead means the crew leader sees the same plan you do, and the companion mobile app puts it in their pocket on the walk from the truck. When the plan changes — and it will — the update has to reach the field the same day, or you're coordinating yesterday's job.
Document Performance — Quietly, Consistently
Keep an honest record of which subs hit their commitments and which don't. Not to build a grudge file — to have facts when you need them. Track planned-versus-actual completion week over week. Note the no-shows, the light crews, the rework. The sub who reliably does what he says is worth protecting and worth calling first on the next job. The one who's always a day late and a crew short needs a documented conversation with the PM before it becomes a claim.
The number that tells you the most is simple: of the tasks a sub committed to this week, how many actually got done as planned? Track that percentage over time and the reliable trades separate themselves from the hopeful ones fast. It also keeps you honest — if the whole job's completion rate is low, the problem might be your planning, not their crews.
What It Comes Down To
Good subcontractor coordination isn't a personality trait and it isn't a software license. It's a weekly discipline: sequence the work by location, pull constraints forward and clear them before crews arrive, confirm manpower against real scope, protect the handoffs with buffer, and get the current plan into every foreman's hands. Do that consistently and the trades start to trust the plan. Once they trust it, they show up ready, and the job runs itself a little more each week. That trust — built one kept commitment at a time — is the real product of everything above, and it's the closest thing to a competitive advantage a superintendent has.