Weather is the one constraint you can't buy your way out of, negotiate with, or expedite. You can pull a submittal forward, air-freight a long-lead valve, throw a second crew at a slab — but when it's raining sideways on your membrane installation, the job stops. What separates crews that lose a week to a storm from crews that lose a day isn't luck. It's that one of them already had the next play ready before the rain started.
The Last Planner System doesn't make weather stop. What it does is force you to see weather-sensitive work coming, screen it before it lands on a weekly plan, and keep a bench of make-ready and interior work you can swing to the moment the forecast turns. Software just makes that discipline faster and harder to skip. Here's how the two actually work together on a live job.
Why weather wrecks schedules that looked fine on paper
The mistake most crews make is treating a rain day as a one-for-one hit — lost one day, add one day. It's rarely that clean. A single hard rain on an open excavation can cost you three days: one for the rain, one to pump and let the subgrade dry, and one to reproof and get the inspector back out. Now shift that downstream. Your rebar crew shows up to a hole that isn't ready, your concrete supplier gave your Tuesday pour slot to someone else, and your pump truck is booked out to Friday. One weather event, four trades scrambled.
That cascade is the real cost, and it's why weather belongs in your look-ahead conversation weeks before it belongs in your weekly work plan. In a healthy Last Planner rhythm, the six-week and three-week look-ahead is where you spot that the exterior painting, the roofing dry-in, and the site concrete are all stacking up in October — right when your region's rainy season starts. That's the window to resequence, not the morning it's pouring.
Screen weather-sensitive work like any other constraint
In the make-ready process, you screen activities for constraints — design, materials, labor, prerequisite work, permits, inspections. Weather is a constraint like any of those. The discipline is simply to tag it and treat it as a real gate, not a footnote.
Know your hard limits cold, because they're not opinions — they're spec:
- Concrete: ACI puts cold-weather protection into play below about 40°F; hot-weather procedures — retarders, shading aggregate, cooling the mix — come into play as you push past roughly 90°F. Below freezing you're into blankets, enclosures, and heat, and you do not strip forms early.
- Membrane and single-ply roofing: the deck has to be dry and most adhesives want the surface above 40°F and rising. Trap moisture under a roof and you've built a warranty problem you can't see yet.
- Masonry: mortar doesn't like to be laid below 40°F without cold-weather measures, and green masonry has to be protected from freezing for its first day or two.
- Painting and coatings: read the data sheet, but most want a surface a few degrees above dew point and generally above 50°F — miss it and you get adhesion failures and blush.
- Crane picks: most machines shut down critical lifts around 20–30 mph sustained wind, and a large wind-sail load lowers that further. That's the manufacturer's chart, not the operator's mood.
When you build these into the plan as constraints, a weather-blocked activity simply can't get pulled into a weekly work plan while the constraint is live — the same way you wouldn't commit an activity whose material hasn't landed. In LookAheadWall, that shows up visually: the location-based board makes it obvious that your Tuesday exterior scope sits on the same corner of the building where the forecast says it'll be raining, and the trade-flow links show you exactly what stalls behind it.
Use the forecast you can actually trust
Match the forecast horizon to the planning horizon, and don't pretend to more precision than exists. A 5–7 day forecast is genuinely useful for your weekly work planning — it's accurate enough to decide whether Thursday's pour is real. Beyond about ten days you're looking at probabilities and seasonal patterns, which is fine for a six-week look-ahead but useless for committing a crew.
The practical move: pull the forecast into your weekly planning meeting and read it against the plan, out loud, with the trades in the room. "NOAA's got 70 percent rain Wednesday afternoon — roofers, does your Wednesday dry-in survive that, or do we pull it to Tuesday and slide the interior framing back a day?" That two-minute conversation, made in front of everyone who's affected, prevents the Wednesday-morning phone tree of confusion. This is exactly what short-interval scheduling is for: small adjustments, made often, before they become emergencies.
Always keep a bench of backup work
This is the single highest-leverage habit, and it's low-tech. Every weekly plan should carry a short list of weather-independent work you can swing a crew to on a bad-weather morning — identified in advance, made ready in advance. Not "find something for them to do." Actually ready.
Good bench work is real work, not busywork:
- Interior scopes on floors that are already dried-in — framing, in-wall rough-in, drywall, blocking.
- Make-ready and prep for the next weather window: layout, staging material, pre-fab, pulling wire you'll terminate later.
- Punch and closeout on completed areas.
- Equipment maintenance and a genuine, planned safety or skills training block — a rain day is a legitimate time to run one, but only if it's queued up, not improvised.
The reason to name this in the plan and not on the fly is that a foreman deciding at 6:45 a.m. in the rain will pick whatever's closest, not whatever's most useful. When your look-ahead already flags "if it rains, the framers go to Building B interior," the redeployment takes five minutes and everybody already knows the play. Location-based planning software helps here in a specific way: because the plan is tied to physical areas of the job, you can see at a glance which spaces are enclosed and available as wet-weather fallback, and which aren't there yet.
The morning the weather actually hits
When it lands, the sequence is boring and that's the point. Boring means repeatable:
- Call it early. Make the go/no-go before crews mobilize, not after they've suited up and driven in. A 5:30 a.m. text beats a 7:15 a.m. shrug.
- Redeploy to the bench. Everyone already knows the fallback, so this is execution, not a meeting.
- Notify the affected trades immediately. The concrete supplier, the pump service, the inspector — cancel and rebook before the day gets away from you. The pour you push to Friday only happens if you call the plant this morning.
- Document what actually happened. Date, conditions, what work stopped, how long, and what you did to recover. More on why below.
A shared, live weekly work plan matters most in exactly this moment. When the superintendent moves an activity and re-links the trade flow, every foreman and sub sees the updated plan on their phone instead of working off a printout that went stale at dawn. That's the difference between the mobile companion app being a nice-to-have and being the thing that keeps forty people rowing the same direction on a chaotic morning.
Recovery: resequence before you throw money at it
Once the weather clears, resist the reflex to immediately buy your way back with overtime and Saturdays. First, assess the real damage — usually less or more than it felt like — and re-plan the sequence. Often you can recover most of a weather hit just by reordering, running two activities in parallel that were planned in series, or shifting a crew to the critical path work first and letting non-critical scope float.
Acceleration — extended hours, weekend work, a second crew, a pump truck you didn't budget — is the tool of last resort because it's expensive and it degrades quality and safety when overused. Tired crews make mistakes; that's when the rework and the recordables show up. Use acceleration surgically on the activities that actually sit on the critical path, and let your rolling look-ahead show you which those are. Chasing overtime on float work just burns money to finish early on scope nobody's waiting for.
Document weather as if you'll need it in a claim, because you might
Most contracts distinguish between normal weather — the rain days a reasonable schedule should have anticipated for that season and location — and abnormal, unusually severe weather that can support a time extension. You almost never get compensated for weather you should have expected. You can get relief for weather that genuinely exceeds the norm — but only if your records are clean and contemporaneous.
That means, every affected day: the actual measured conditions, which specific activities were impacted, how long the impact ran, and the concrete steps you took to recover. Notes written the morning of are worth ten times a reconstruction you cobble together during a dispute six months later. A scheduling system that timestamps plan changes and keeps the history gives you that trail almost for free — the record of what you planned, when the weather forced a change, and how you responded is just sitting there in the log.
Plan the season, not just the storm
The best weather management happens at the look-ahead level, long before any single forecast. Sequence your weather-sensitive scope — site concrete, dry-in, exterior skin, roofing — into the seasons that suit it. Push hard to get the building enclosed before your cold or wet season so you can run interior work through the ugly months. Schedule big pours and paving into the moderate-temperature windows on purpose. Watching a six-week look-ahead against your region's seasonal patterns lets you make those moves while they're still cheap to make.
None of this is exotic. It's the same Last Planner discipline you already run — screen constraints, make work ready, commit only what's truly ready, learn from what didn't go as planned — applied to the one constraint you can't control. Software doesn't replace that judgment. It just makes it faster to see the collision coming, quicker to swing the crews, and easier to prove what happened afterward. Weather is certain. Only the timing is up for grabs — and the timing is exactly what a good look-ahead is built to handle.