A schedule is a promise about the future. Progress tracking is the honest accounting of whether you kept it. Most of us are good at the first part and terrible at the second. We build a beautiful weekly work plan on Friday, and by the following Wednesday nobody actually knows how much of it got done — we just have a general sense that "framing's a little behind" and a growing pit in our stomach. That gap between what you planned and what you can actually see is where jobs quietly slip.
Good look-ahead software closes that gap by making progress something you record, not something you feel. Here's how progress tracking actually works when it's done right, what to capture, and the ways it goes wrong on real jobsites.
What "progress" really means on a jobsite
The first mistake crews make is confusing time spent with work done. An activity that's on day 3 of a 4-day duration is not automatically 75% complete. It might be 40% complete because the deck was wet Monday and half the crew got pulled to chase a punch item. Time-elapsed is the lie your gut tells you; installed quantity is the truth.
So the useful unit of progress isn't "how many days has this been going" — it's some version of how much of the scope is physically in place and would survive an inspection. On a drywall activity that's sheets hung on a wall. On underground it's linear feet of pipe in the trench, backfilled and shot. When you track the physical thing instead of the calendar, your numbers stop flattering you.
The status categories worth using
You don't need fifteen statuses. You need four that mean something and that everyone reads the same way:
- Not started. Nothing's happened. The value here is watching the ones whose planned start has already passed — that's your early-warning list before it becomes a delay.
- In progress. Work is underway. This is where a percent-complete or a quantity matters, because "in progress" covers everything from "just broke ground" to "one connection from done."
- Complete. Finished and it meets the completion criteria (more on that below). Not "the crew left the area." Left-the-area and actually-done are different things, and the difference is usually a missed inspection.
- Blocked. The most valuable status on the board and the one most software buries. A blocked activity isn't behind because the crew is slow — it's stopped for a reason outside their control: missing material, a predecessor that didn't finish, an RFI hanging, an inspection not scheduled. Flag the block and name the reason, because that's the thing you can actually go fix.
Blocked is where short-interval scheduling earns its keep. In a Last Planner-style weekly work plan, an activity should only be on this week's plan if its constraints are cleared. When something still ends up blocked, that's a signal your make-ready look-ahead missed a constraint — and that's a lesson, not just a delay.
Define "complete" before you ever mark anything complete
This is the single biggest source of phantom progress. A foreman marks rough-in done. The GC's schedule shows it green. Two weeks later the inspector red-tags it because the plumber's straps are missing and the electrician's boxes are set proud of the finished wall line. Was it complete? On paper yes, in reality no, and now you're reopening a wall you already insulated.
For anything with an inspection or a downstream trade depending on it, "complete" should mean at least three things: the physical work is finished, it meets the quality standard, and any required inspection has passed. Electrical rough-in isn't done until it's inspected — so megger the runs and confirm the sign-off before you let the insulator and drywaller in behind it. Underground isn't done until it's been shot and the as-built quantities are recorded. Bake the inspection gate into your definition of complete and you stop closing walls on unfinished work.
Percent complete, without lying to yourself
If you're going to track percentages, make them mean something. A few rules that keep the numbers honest:
- Tie the number to a quantity, not a calendar. 200 of 500 linear feet installed is 40%. "Feels like halfway" is not a number.
- Use coarse increments. Nobody on a jobsite can tell 63% from 68%. Quarters — 0, 25, 50, 75, 100 — are honest and fast to report. Fine-grained percentages just invite made-up precision.
- Watch the 90% trap. Activities love to sit at "90% done" for a week while the last 10% — the fiddly connections, the fire-caulk, the cleanup, the punch — takes as long as the first 60%. When something's been near-complete for several updates, treat it as blocked and go find out why.
Where the data has to come from: the field
Progress data is worthless if it lives in the trailer and gets updated once a week by someone guessing. It has to come from the people doing and watching the work — the superintendent walking the floor and the foremen running the crews. That's the whole reason a mobile companion app matters: a crew leader should be able to stand in the room, look at the wall, and tap it to 50% before he's even back to the gang box. The moment you make someone walk to a laptop to report progress, the reporting stops happening.
A few things that make field updates actually stick:
- Make it a 10-second action. Tap the activity, set the status, done. If updating a plan takes longer than the work of remembering it, it won't get done consistently.
- Photos over prose. A photo of the installed work attached to the activity settles more disputes than any note. It's your proof for progress billing and your record when a sub swears they finished something they didn't.
- Assume no signal. Basements, elevator shafts, and rural sites eat cell service. Field tools need to capture updates offline and sync when they're back in range, or you'll get a hole in your data exactly where the tricky work happened.
Comparing actual to plan — the whole point
Capturing progress is just bookkeeping until you lay it against the plan. That comparison is where the software pays for itself. When your look-ahead schedule shows planned versus actual side by side, three things jump out:
- Variance you can see today. Which activities are behind, which are ahead, and by how much — before the weekly meeting, not during the argument at it.
- Trend, not snapshot. One activity behind is noise. Four weeks of the same trade consistently finishing 70% of what it committed to is a signal. In Last Planner terms that's your Percent Plan Complete telling you a sub is over-promising, and the fix is a hard conversation about realistic commitments, not more yelling.
- Forecast impact. Real progress lets the schedule re-forecast honestly. If drywall is running a day slow, the software can push the downstream paint and trim so you're not staring at a critical-path date that everyone knows is already dead.
This is also where the "reason codes" on your blocked and incomplete activities become gold. When you review a month of variance and see that half your misses trace back to "material not on site," you don't have a scheduling problem — you have a procurement problem, and now you can prove it.
How often to update
Match the cadence to the stakes. Critical-path and constrained work — the stuff that holds up three trades behind it — deserves a daily touch, even if it's just the super confirming status on the morning walk. Everything else can ride on a weekly rhythm tied to your weekly work plan cycle: update actuals at the end of the week, replan the next week against reality, roll the look-ahead forward. On top of that, capture the big events when they happen — the failed inspection, the material that didn't show, the change that added scope. Those don't wait for update day.
Keeping the data honest
None of this works if the numbers are optimistic fiction. Two habits protect you. First, verify against your own eyes — a status a sub self-reports gets confirmed on the walk, not taken on faith. Subcontractor self-reporting is a huge time-saver, but unverified it drifts toward wishful thinking fast. Second, apply the same standard every time. If 75% means "sheets hung but not finished" on one wall, it means that on every wall. Consistency is what makes trend analysis trustworthy; inconsistent reporting just produces confident-looking garbage.
And set the tone that honest bad news is welcome. The foreman who tells you on Tuesday that he's blocked and won't hit Friday is doing you a favor. The one who reports 90% every week to avoid the conversation is setting you up for a much worse one later. Progress tracking only works in a culture where reporting reality doesn't get you chewed out.
Putting it to work
If you're standing up progress tracking on a job, keep the rollout simple. Agree on what your four statuses mean and what "complete" requires — write it down so every foreman reads it the same way. Get the update tool into field hands on their phones so reporting happens where the work is. Train the crew that a blocked activity with a named reason is a good report, not a confession. Then spend five minutes a week actually looking at variance and trend, because data nobody reviews is just data entry.
Plan with a look-ahead, but manage with the actuals. A tool like LookAheadWall gives you the visual weekly plan and the field updates in one place, but the discipline is yours: capture what really happened, lay it against what you promised, and let the gap tell you where to point your attention Monday morning. That loop — plan, do, measure, replan — is the entire difference between a schedule that's a wish and one that runs the job.