Every job runs late somewhere. The rebar shows up bent, the inspector red-tags a run, a mechanical sub no-shows on the morning you sequenced them to lead. The question is never whether you'll take a delay — it's whether that delay stays a one-day problem or metastasizes into a three-week milestone slip nobody caught until it was too late to fix cheaply. The difference is almost entirely in how the superintendent reads the delay in the first 24 hours and what they do next.
This is where a good look-ahead process earns its keep. Not because software recovers time for you — it doesn't, only crews and decisions do that — but because it tells you fast which delays actually matter, and gives you a place to test recovery moves before you commit crews and money to them.
First: figure out if this delay even matters
Not all delays are equal, and the rookie mistake is treating them like they are. A two-day slip on an activity sitting on ten days of float is a shrug. The same two-day slip on your critical path can push a certificate of occupancy and cost you liquidated damages. Before you do anything else, answer one question: does this activity have float, or is it driving the finish?
Walk the dependency chain downstream from the delayed work. If drywall slipped two days, who was waiting on drywall? Tapers, then prime paint, then trim, then final paint, then flooring, then final clean. If any of those trades are already scheduled tight against a hard milestone, your two days just became somebody else's two days, all the way down the line. That's the ripple, and you want to see it on a wall or a screen, not carry it in your head.
A weekly work plan that shows trade flows connected — where the finish of one activity feeds the start of the next — makes this ripple visible instead of theoretical. In LookAheadWall you can trace the downstream chain from the slipped activity and see exactly which crews inherit the problem. That's the whole point of location-based, connected scheduling: the impact isn't a guess, it's drawn out in front of you.
The float trap
Here's the gotcha that burns experienced supers: float gets consumed silently. You've got a delay that lands on an activity with five days of float, so you don't sweat it. But three other minor slips already ate four of those days, and now you're one day from critical without ever having made a decision. Track float on the activities that matter, and treat a shrinking float buffer as an early warning, not a comfort. When float on a milestone-driving chain drops under two or three days, that chain is effectively critical — manage it like it.
Understand what kind of delay you're actually fighting
Recovery options depend entirely on the cause, so name it honestly:
- Manpower delay — the sub short-crewed or no-showed. Fixable with bodies, overtime, or a second shift, assuming the sub can find the labor.
- Material/procurement delay — the gear isn't on site. No amount of overtime fixes a switchgear that's six weeks out. This is a resequencing problem, not an acceleration problem.
- Predecessor delay — someone upstream handed off late, so the trade sitting idle isn't at fault. Chasing the wrong sub here wrecks relationships and fixes nothing.
- Rework / quality delay — failed inspection, wrong installation. Watch this one, because rework tends to breed more rework if you rush the fix.
- Weather / owner / design delay — outside your control, and the one you must document cleanly for a time-extension request.
Mislabel the cause and you'll pick the wrong recovery. Throwing overtime at a predecessor delay just pays a crew to stand around faster.
Recovery move #1: resequence before you spend money
The cheapest recovery is almost always resequencing, because it costs no premium labor — just planning. Ask whether downstream work truly needs the delayed activity finished, or only needs part of it. Can flooring start in the east wing while the west wing waits on a repair? Can the electrician rough-in the corridors while the delayed panel schedule gets sorted? Location-based planning shines here: if you're scheduling by area rather than by lump activity, you can peel work apart and keep crews productive in the zones that are ready.
This is where a rolling look-ahead beats a static bar chart. You're not trying to preserve the original sequence out of pride — you're finding the path through the work that keeps the most crews moving. Sketch the alternative sequence in your weekly plan, check it doesn't create a new pileup two weeks out, then commit.
Recovery move #2: acceleration, and its real cost
When resequencing won't cover it, you add horsepower — overtime, added crew, or a second shift. All three work, and all three have failure modes supers learn the hard way:
- Overtime is fast to deploy but productivity decays. The first Saturday is fine. By the third or fourth week of sustained six-day, ten-hour weeks, you're paying a premium for tired crews making more mistakes — and mistakes mean rework, which is the delay you were trying to escape. Use overtime for a short sprint, not as a lifestyle.
- Added crew helps only if there's room and work to add them to. Two crews stacked in the same corridor trip over each other and you lose the gain to congestion. Confirm you can actually deploy the extra bodies to separate work areas before you ask the sub to bring them.
- Second shift can double throughput on the right work, but it needs its own supervision, lighting, and a clean handoff from day shift. It also invites trade-stacking conflicts if two shifts are working the same zones. It's powerful and it's a management load — go in with eyes open.
Before you commit to any of these, model it. Drop the overtime hours or the second crew into your look-ahead and see whether the recovered finish actually clears the milestone. Half the time the math shows you're spending real money to recover a day that float would've absorbed anyway. The other half it shows you need more than you thought, and better to know now.
Recovery lives or dies in trade coordination
Almost no recovery is a single-trade fix. Speed up the drywaller and you've just moved the pressure onto the taper and the painter, who planned their crews around the original dates. If you accelerate one trade without renegotiating the handoffs downstream, you've built a new jam a week out. Recovery is a conversation with everyone in the chain, not an order to one sub.
Get the affected foremen in a room — or on the phone — the same day. Walk the revised sequence together. The taper needs to know the wall's coming two days early; the painter needs to know he's losing his weekend buffer. When subs can see the same weekly plan you're looking at, the conversation is about the work instead of about whose fault it is. Sharing the schedule directly with subs — which is exactly what a tool like LookAheadWall is built to do — turns a finger-pointing meeting into a coordination meeting.
Protect your buffers, don't burn them all at once
Veteran supers build small buffers into the plan on purpose — a day or two between frame-out and rough-in for cleanup and inspection, a cushion ahead of any inspection gate. When a delay hits, resist the urge to spend every buffer immediately to look good on paper. Buffers are your shock absorbers. Spend them deliberately on the constraints that actually threaten a milestone, and keep some in reserve, because another delay is coming before this job closes out. It always does.
Document as you go — future-you will need it
If the delay wasn't your fault — owner-directed change, late design response, weather, differing site conditions — the documentation you create in the first few days is what supports a time extension or a claim months later. Reconstructing it after the fact never holds up. Capture, contemporaneously:
- The cause, in plain language, with the date and who or what drove it.
- What it delayed, and the downstream activities affected.
- The recovery steps you took and what they cost — overtime hours, added crews, resequencing.
- Photos, daily reports, RFI and inspection references tied to the event.
Your daily reports and weekly work plans, kept honestly, become the paper trail. The super who can show a clean before-and-after of the schedule and a dated record of the disruption is in a far stronger position than the one relying on memory and a stack of loose photos.
Escalate early, and communicate the plan — not the panic
There's an instinct to sit on a delay and try to quietly recover it before anyone upstream notices. Resist it. If a delay threatens a milestone, your PM and the owner need to know while there are still options, not after the finish date is already blown. Escalate with a recovery plan attached, not just bad news — "we lost two days on the panel, here's how we're getting one back through resequencing and where we still need help" lands completely differently than "we're behind."
Keep the recovery plan visible to management through the look-ahead. A four-week or six-week look-ahead showing the recovery sequence tells the whole story at a glance: what slipped, what you're doing about it, and whether the milestone still holds.
Then track the recovery like it's its own job
A recovery plan you don't monitor is a wish. Once you commit to a recovery sequence, measure against it daily. Did the accelerated crew actually hit yesterday's target, or did they fall a half-day short and quietly re-open the gap? Recovery plans decay fast when nobody's watching, and a half-day miss each day rebuilds the delay you just paid to erase. Update the weekly plan every day of the sprint, and if the recovery isn't holding, catch it in day two, not week two.
Close the loop: why did this happen?
Every delay is also a data point. The best superintendents run a quick post-mortem on the ones that hurt — not to assign blame, but to catch the preventable pattern. A huge share of field delays trace back to a constraint that was known and simply not cleared in time: material not confirmed on site, prerequisite work not inspected, a permit or submittal still open. That's the whole logic behind constraint-based look-ahead planning — you screen the next few weeks of work for what's not ready, and you clear those constraints before the crew shows up to a task they can't actually start.
Track which delays were preventable and which were genuinely outside your control. Over a few months the pattern tells you where your process leaks — maybe your material confirmations are always the weak link, maybe one sub is chronically short-crewed. Fix the pattern and you stop fighting the same delay on the next job.
The bottom line
Delays don't distinguish the good superintendent from the mediocre one — everybody takes them. What separates them is the response: reading fast whether the delay actually threatens a milestone, naming the real cause, reaching for resequencing before reaching for the checkbook, coordinating the whole trade chain instead of leaning on one sub, and then tracking the recovery until it holds. A disciplined look-ahead process — with a tool that makes the trade flows, downstream ripple, and recovery sequence visible to you and to your subs — is what turns that response from a scramble into a routine. The delay was always coming. Handling it well is the job.