Every superintendent has lived some version of this: it's fourteen months after substantial completion, you've moved on to two other jobs, and an email lands from the owner's attorney asking why the second-floor slab pour on the east wing slipped eleven days. You genuinely don't remember. The only thing standing between you and a guess is your daily log. If it says "0.9 inches of rain 6 AM to noon, subgrade saturated, pump truck released, no pour," you win that conversation in thirty seconds. If it says "rainy," you've got nothing.
That's the whole reason daily logs exist, and it's why the daily log is one of the few features that makes subcontractor management software worth logging into every day instead of once a week. A log that lives inside the same tool as your schedule and your subs isn't just a diary — it becomes the audit trail that explains every variance on your look-ahead. Below is how the pieces actually fit together on a real job, and how to write logs that hold up when it matters.
Write the log for the version of you that has forgotten everything
The single most useful mindset shift is this: you are not writing today's log for today. Today you remember everything. You're writing it for the you who has slept 400 nights since, or for a project engineer who wasn't even on the crew. Contemporaneous documentation — written the day it happened — carries enormous weight in a claim precisely because it wasn't reconstructed after the dispute started. A memo dated three weeks late reads like a story built to win an argument. A log entry timestamped the afternoon it happened reads like the truth.
So the standard is specificity, not volume. "Framing behind" is useless. "Framing crew of 4 (planned 8), 2nd-floor demising walls unstarted, drywall stack scheduled Thursday now at risk" is a record that tells you something and protects you later. Every field below lives or dies on that same rule.
What each subcontractor actually did today
The core of the log is a per-sub account of work performed. Not "electrical on site" — which electricians, in which area, doing what. "Acme Electric, 3 hands, roughed-in branch circuits gridlines 4–7, third floor" is a line you can defend and, more importantly, use.
Here's where the connection to your schedule earns its keep. When your daily log lives in the same system as your weekly work plan, that per-sub entry cross-references against what you planned. Your look-ahead said Acme would finish the third-floor rough-in Wednesday; your log says they got two-thirds of it done with half the crew. That's not two facts, it's one story — and it's the story you'll tell at Friday's coordination meeting. Good short-interval scheduling depends on an honest read of actual versus planned, and the daily log is where "actual" gets recorded in the field, in the moment, by the person who watched it happen.
Manpower counts — the number subs argue about most
Log headcount by trade, every day, without exception. This is the field that catches the slow-motion failure you feel but can't yet prove: the sub who committed eight and has quietly been running four for two weeks. On any single day, four instead of eight is an off day. Logged across ten days, it's a manpower trend you can put in front of the PM with a straight face, and it's the difference between a vague "you're behind" and "you've averaged 4.2 hands against a committed 8 since the 6th."
Two practical notes from experience. First, count heads yourself or have your foreman do it — don't take the sign-in sheet's word for it, because the sheet counts who badged in, not who's actually swinging a hammer at 1 PM. Second, log planned crew size alongside actual. A count with no baseline is just a number; a count against the commitment is leverage.
Weather — the field superintendents shortcut and later regret
Weather is the field most people phone in, and it's the one that most often decides money. If you're ever going to file a weather delay or defend against a liquidated-damages assessment, the log is your case. So don't write "rain." Write the amount, the timing, and — this is the part people skip — the specific impact on specific work.
"0.9 in. rain, 6 AM–noon; parking-deck subgrade saturated; ABC Concrete pump truck released 7:15 AM; slab pour gridlines A–D pushed to next dry window" is a delay claim in one sentence. Temperature matters too, and not just for concrete and masonry cold-weather protection — sustained heat kills productivity and belongs in the record when it does. A rule of thumb worth keeping: if weather changed what a crew could do, or how much they got done, it goes in the log with the number attached.
Deliveries, materials, and the constraints they clear
Log deliveries with quantity, condition, and any damage — the day they land, not the day you get around to it. This is straightforward record-keeping, but it does something valuable inside a scheduling system: material on site is often the last constraint standing between a planned activity and a ready-to-start one.
In look-ahead practice you screen upcoming work for constraints — materials, information, prerequisite work, labor, equipment, permits — and you don't commit an activity until they're clear. The delivery log is the field-side proof that a material constraint just cleared. Twelve pallets of the right tile showed up in good shape? Your tile-setting activity is now genuinely ready. Half the order arrived and the rest is on backorder? The log captures the miss the day it happened, with quantities, so when that finish activity slips you already have the cause documented instead of reverse-engineering it a month later.
Equipment on site and what it was doing
Track equipment presence and use — the 40-ton crane, the two scissor lifts, the pump. Two payoffs. One, rental billing: when the invoice says the lift was on site 22 days and your logs say 15, the logs win, and that's real money. Two, the log surfaces the equipment conflict you didn't see coming — the boom lift two trades both needed Thursday, or the crane that didn't show and stalled the steel set. When shared equipment is the bottleneck, the daily log is what makes the collision visible in hindsight so you plan around it next time.
Safety — the entries you hope you never need
Log the safety work: toolbox talks and their topic, inspections completed, PPE and fall-protection observations, and every incident or near-miss, however minor. Two reasons, and they pull in opposite emotional directions. The routine entries — "toolbox talk, silica exposure, 22 attendees" — are boring right up until an OSHA visit or an insurance question, at which point a consistent record of talks and inspections is your evidence of due diligence.
The incident entries are the ones that matter most and get soft-pedaled most. Document near-misses honestly and specifically. A near-miss logged today is a corrective action you can actually take; a near-miss logged vaguely, or not at all, is the incident report you're writing next month. Nobody enjoys writing down that a load swung too close to a crew, but the log that records it and the fix is the log that keeps someone out of the hospital.
Visitors and inspections — closing the loop on constraints
Record who came on site: inspectors, the owner's rep, the EOR, the AHJ, anybody official. For inspections specifically, log the result, because a passed inspection usually clears a constraint on the very next activity. Your look-ahead can't release framing to cover the wall until rough-in inspection passes; the log entry "electrical rough-in inspection PASSED, third floor, 10:40 AM, Inspector Ramos" is the green light, documented. A failed inspection is even more important to capture — reason, corrections required, and the reinspection date — because that's the delay everyone will ask about later.
Issues and delays, written the day they happen
Every problem that touched the work goes in the log: the delay, what caused it, and what it hit. This is where the contemporaneous-record principle pays its biggest dividend. When your schedule shows a slip, the daily log is the paired explanation — the "why" that lives next to the "what." Reconstructing that why weeks later is exactly the exercise that makes claims fall apart, because reconstruction always looks self-serving.
This is also the strongest argument for a log you can write from your phone in the field. The best time to record that the plumber hit unmarked conduit at 9:15 is 9:20, standing there looking at it — not 4:30 back in the trailer trying to remember which gridline it was on. A mobile-accessible log turns "I'll write it up later" (which becomes never, or wrong) into a thirty-second note captured while it's true.
Photos, attached to the entry
A photo tied to the day's log is worth more than either alone. Snap the saturated subgrade, the damaged delivery, the exposed rebar before the pour, the near-miss condition. Written descriptions plus dated images make a record almost nobody argues with. Build the habit into your daily walk: a handful of purposeful shots — progress, problems, anything you'd want a stranger to see later — attached to that day's entry. Software that prompts for photos as part of closing the log is doing you a favor by making the good habit the default one.
Where the daily log meets the look-ahead
Pull it together and the value isn't the log by itself — it's the log wired into the schedule. Tools built for short-interval scheduling, LookAheadWall among them, treat the weekly work plan and the daily log as two sides of the same record: the plan states what should happen, the log records what did, and the gap between them is your variance. Track that gap honestly and it becomes the raw material for real improvement — you start seeing which trades habitually overpromise crew, which constraints keep biting, which sequences never go as drawn.
None of this requires software to be true — a superintendent with a legal pad and discipline can keep a bulletproof log. What the integrated tool buys you is that the log stops being a separate chore and becomes a byproduct of running the job: the constraint that cleared, the inspection that passed, the crew that came up short all land in one place, cross-referenced to the schedule they affect. Do that consistently and, fourteen months later when that attorney's email lands, you won't be guessing. You'll just open the log.