Ask any superintendent why last week's plan fell apart and you'll get some version of the same answer: the crew showed up ready to work and something wasn't there. The material was still on a truck in Sacramento. The inspector hadn't signed off on the deck below. The RFI answer never came back, so nobody could set the embeds. The work didn't fail because the guys were slow or the sequence was wrong. It failed because we scheduled an activity that was never actually ready to run.
That gap — between what the schedule says can happen and what the jobsite will actually allow to happen — is what constraints are all about. A look-ahead is only as honest as its constraint list. Fill the grid with pretty bars and no make-ready work behind them and you've built an optimistic wish list, not a plan. Get disciplined about identifying and clearing constraints two to six weeks out, and the weekly work plan stops being a hope and starts being a commitment you can actually keep.
What a Constraint Really Is
Strip away the jargon and a constraint is just anything that has to be true before a crew can start an activity and finish it without stopping. If it isn't true yet, the work isn't ready, and putting it on this week's plan is lying to yourself. In the field there are really only about seven kinds worth tracking:
- Materials — the actual product on site, not "ordered," not "on the way." A submittal approved but with a 10-week lead time is a constraint you needed to see two months ago.
- Information — approved drawings, a closed RFI, an ASI, a resolved detail. If the crew has to guess, the work isn't ready.
- Prior work (predecessors) — the deck poured and cured, the wall framed and inspected, the substrate accepted. "Mostly done" is not done.
- Labor — the right crew, the right headcount, the right skills, actually available that week and not buried on another job.
- Equipment — the lift, the pump, the crane pick, the welding machine — reserved and operational, not double-booked.
- Space and access — a clear, safe, physically reachable work area. Two trades stacked in the same 400 square feet is a space constraint even if everything else is ready.
- Approvals and conditions — permits, inspections, owner sign-offs, and the external stuff you don't control: weather windows for a big pour, a utility shutdown, a street-closure permit.
Most blown commitments trace back to one of these. The value of naming them isn't bureaucratic — it's that a named constraint has an owner and a due date, and an unnamed one just becomes Friday's excuse.
Make-Ready: The Work Behind the Work
The whole point of a look-ahead — as opposed to just tracking today — is that it gives you a runway. Activities roll into the window a few weeks out precisely so you have time to clear their constraints before they hit the commitment week. That clearing process is what the Last Planner crowd calls make-ready, and it's the single most important habit that separates a superintendent who hits 85% plan reliability from one stuck at 50%.
Make-ready runs on a simple loop, and it's worth doing it the same way every week:
- Screen. As each activity enters the look-ahead window, walk it and ask the seven questions above. What has to be true for this to run clean? Write down every "not yet."
- Assign. Every constraint gets one name and one date — the person who owns clearing it and the day it needs to be clear by. Not "the team." A person.
- Clear. The owner does the work — chases the submittal, schedules the inspection, confirms the crew, moves the material.
- Verify. This is the step everybody skips, and it's the one that matters. You don't mark a constraint clear because someone said it's handled. You confirm the material is physically on site, the RFI is actually answered (not just "in review"), the inspection actually passed. Assumed-clear is how work still fails on a plan full of green checkmarks.
Only activities that survive all four steps are allowed onto the weekly work plan. That's the discipline. An activity with an open constraint doesn't get committed — it stays in the look-ahead as a warning light until it's genuinely ready. This is exactly where a purpose-built tool earns its keep: LookAheadWall lets you hang constraints on an activity and keeps it visually flagged until they're cleared, so nobody quietly slides a not-ready task onto Monday's plan because the bar looked lonely.
How Far Out to Look — and When to Commit
Lead time is the whole game. A six-week look-ahead exists so that a constraint with a six-week resolution — a long-lead material, a design decision that has to climb three levels of approval — can still be caught in time. Shorten the window to two weeks and you'll keep discovering ten-week problems with two weeks to fix them, which is to say you won't fix them.
A rule of thumb that holds up on most commercial and multi-family jobs:
- Weeks 4–6 out: constraint identification. You're screening activities and surfacing the long-lead stuff — materials, major RFIs, permits, big equipment.
- Weeks 2–3 out: active make-ready. Owners are clearing constraints; the short-lead items get chased down.
- Week 1 (commitment): only fully constraint-free activities make the weekly work plan. If it's not clear by now, it doesn't get promised.
The other quiet killer is the buffer between trades. Frame-to-rough-in usually wants a day or two of cushion for cleanup, inspection, and the punch that always shows up. Rough-in to cover — drywall, close-up — wants its own buffer so the electrician can megger the runs and the plumber can hold pressure on the lines before anybody hides them in a wall. Schedule those transitions back-to-back with no air and the first inspection hiccup blows the next three activities. Treating "waiting on inspection" as its own constraint, with its own slot on the calendar, keeps that domino from starting.
Ownership: One Name, Not a Committee
A constraint assigned to everyone is owned by no one. Every open item needs a single owner, and — this is the part people get wrong — the owner has to be someone who can actually resolve it. Assigning a subcontractor-lead-time constraint to your project engineer when only the sub can expedite the order just launders the delay. Match the constraint to the person with the authority and reach to clear it: material lead times to whoever controls purchasing, design questions to the PM who owns the RFI log, inspection scheduling to the super who calls the inspector.
Split the tracking honestly between GC-owned and sub-owned constraints, too. Plenty of what stops a trade is the GC's obligation — the deck the sub is waiting on, the RFI answer they need, the layout that hasn't been shot. If your make-ready only tracks the subs' homework and never your own, you'll be the constraint nobody wrote down. In the coordination meeting, the subs should be able to point at a GC-owned item and say "we're stuck on that," and it should already be on the board with your name on it.
Run the Weekly Meeting on Constraints, Not Status
The biggest change most teams make when they get serious about this is what they talk about in the weekly planning meeting. A status meeting asks "how's it going?" and gets a fog of optimism. A constraint meeting asks three sharper questions:
- What did we commit to last week, what got done, and for anything that didn't — what constraint did we miss? (This is your plan-percent-complete, and the misses are pure gold. Log the reason.)
- What's rolling into the window, and what does each new activity need to be ready?
- Which constraints are stuck, and who needs to escalate what today?
That last one is where good teams separate themselves. A constraint that's been "in progress" for three weeks isn't in progress — it's stuck, and it needs to jump a level. Build escalation into the meeting so a stalled submittal or a silent RFI gets pushed to the PM or the owner before it eats the sequence, not after.
Tracking the misses over time turns the whole exercise into a learning loop. When you log why each commitment failed, patterns fall out fast. If a third of your misses are material constraints, your buyout and lead-time process is the problem, not your crews. If they're mostly information, your RFI turnaround is strangling the field. You can't fix a pattern you never wrote down, which is why a look-ahead tool that keeps constraint history — reasons for variance, resolution timing, category breakdowns — is worth far more than a prettier bar chart.
Hidden Constraints and How to Flush Them Out
The constraints that hurt worst are the ones nobody put on the list — the assumed ones. The crew you were sure would be free got pulled to another job. The "obvious" dependency nobody bothered to write down. The weather window you treated as a given until the pour got rained out. Two habits flush these out before they bite:
First, verify availability instead of assuming it — especially labor and equipment. "The framers are always here" is not a plan; a confirmed headcount for that specific week is. Second, get the field to talk. Your foremen and crew leaders can see constraints from ten feet away that never appear on a drawing — an access conflict, a stored-material pile blocking the work, a trade ahead of them running behind. Give them an easy way to flag it. This is a big part of why the LookAheadWall mobile companion exists: a crew leader standing in the actual space can raise a constraint the moment they spot it, instead of it surfacing as a surprise at Friday's meeting when it's already too late to fix.
When a Constraint Won't Clear
Sometimes make-ready fails and the constraint just isn't going to be resolved in time. The wrong move is to leave the activity on the plan and hope. The right move is to have a bench. Keep a short list of workable backlog — real, constraint-free tasks the crew can pivot to when the primary work gets blocked. When the inspector no-shows and the wall can't close, the crew doesn't stand around; they hit the ready work you set aside for exactly this. Defer the blocked activity, pull a clean one forward, and you've protected the week's production even though one thing went sideways. That flexibility only exists if you screened enough activities to have a backlog in the first place — another argument for a real look-ahead window instead of living week to week.
Getting Started Without Boiling the Ocean
You don't need a six-month rollout to make this work. Start narrow and let it prove itself:
- Adopt the seven constraint categories above so the whole team names things the same way.
- Add one question to your existing weekly meeting: "What has to be true before this can run?" — and write down every answer with an owner and a date.
- Only commit constraint-free activities to the weekly work plan. Hold the line on this even when it's uncomfortable.
- Track your plan-percent-complete and, more importantly, log why each miss happened.
Do that for a month and the pattern is almost always the same. Plan reliability climbs, the Friday excuses thin out, and the crews start trusting the plan because it stops sending them to work that isn't ready. That trust is the real prize. A short-interval schedule earns its authority not from how good it looks on the wall, but from how often the work it promises actually turns out to be there. Master the constraints behind the bars, and your look-ahead becomes what it was always supposed to be — a reliable guide to the next few weeks, not an optimistic guess.