Handing a foreman a login and a five-minute walkthrough is not training. I've watched it play out too many times: the new guy nods along, clicks a few buttons, says he's got it, and three weeks later the look-ahead is a graveyard of half-finished weeks and constraints nobody ever cleared. The tool wasn't the problem. The onboarding was.
A real training checklist isn't a list of buttons to press. It's a sequence of competencies you verify in order, because each one depends on the last. Somebody who can't read a look-ahead has no business building one. Somebody who can build one but can't identify constraints will publish schedules full of work that can't actually start. Below is the checklist I use to bring a superintendent, foreman, or PM from "has a login" to "I trust this person to run the weekly plan without me looking over their shoulder." Work it top to bottom. Don't skip ahead.
Before you train anyone: get the concept straight
The single biggest onboarding failure is teaching software to someone who doesn't understand the practice underneath it. A look-ahead schedule is not a smaller version of the master CPM schedule. It's a different animal with a different job. The master schedule tells you where the project should be. The look-ahead — typically a rolling three- to six-week window — is where you turn those milestones into work you can actually assign this week, and where you surface every reason a task might not start on time.
Before anyone touches the app, make sure they can answer, in their own words:
- Why the window rolls — you drop the completed week off the front and add a fresh week on the back, so you're always looking the same distance ahead.
- The difference between "should" and "can" — the master says a task should happen; the look-ahead exists to confirm it can, by clearing constraints before the crew shows up.
- Why three weeks is the common sweet spot for the detailed window (near enough to be real, far enough to react) and why some teams run four or six for long-lead trades.
If they can't explain why a rolling window beats a static one, stop. Ten minutes of concept now saves you a month of a foreman treating the look-ahead like a Gantt chart he updates when he remembers.
Navigation and orientation
Now the mechanics. This part is genuinely quick, and it should be — good construction scheduling software shouldn't need a manual to find your way around. Verify the trainee can:
- Log in on both a desktop and the device they'll actually use in the field — for most crew leaders that's a phone.
- Find the current week's plan without help, and switch between the project, the location or area breakdown, and the crew view.
- Tell you what they're looking at when they land on the wall: which column is which day, how activities are grouped by location, and where the trade sequence flows.
Small tell that separates real orientation from a nod: ask them to find last week's plan and next week's plan without you pointing. If they can navigate time, they understand the rolling structure. If they get lost, they were memorizing clicks, not learning the layout.
Reading a schedule before building one
This is the step everyone wants to skip, and skipping it is why so much bad data gets entered. A person who can't read a weekly work plan correctly will build one that reads wrong to everybody else.
Verify they can look at an existing week and correctly answer:
- Which crew is where, on which day, doing what — down to the location, not just the trade.
- How the trade flow connects: who hands off to whom, and what the sequence is telling them about dependencies. On a typical multi-family stack that's framing, then rough-in trades, then insulation, then drywall — and they should be able to read that flow off the wall, not recite it from memory.
- What's planned versus what's actually committed. A task sitting on the wall is a plan; a task with its constraints cleared is a commitment.
Hand them a real week from your project and quiz them. If they misread a hand-off or can't tell you why two activities are stacked in the same location, they're not ready to create.
Building a week from scratch
Only now do you let them build. Have them create a fresh week end to end, ideally shadowing a real upcoming week so you can compare their version to yours.
Watch for the things that separate a usable plan from a mess:
- Right-sized activities. The classic rookie error is dropping in a two-week "drywall" bar. In a look-ahead, work should be broken down to roughly the level you'd assign to a crew for a day or a few days — "hang unit 204–208," not "drywall, building C." If the task can't be finished and checked inside the window, it's too big.
- Honest durations. New planners pad everything or pad nothing. Coach them toward the crew's real production rate and remind them a look-ahead duration is a promise, not a hope.
- Sequence that respects reality. They should place work in the order trades actually flow, and leave the buffers that keep the job from tripping over itself. Frame-to-rough-in usually wants a day or two of slack for cleanup, layout, and inspection — not a same-day hand-off that collapses the first time a framer runs long.
The output you're grading isn't "did they fill the week." It's "would a sub read this and know exactly what to do Monday morning."
Constraints: the part that actually makes look-aheads work
If your trainee only masters one thing, make it this. The entire value of a rolling look-ahead is that it forces you to identify and clear constraints before the work is scheduled to start. Skip constraint management and you've just built a prettier wish list.
Verify they can:
- Look at a planned task and name what has to be true for it to start — and there's almost always something. Common constraint categories worth drilling: prerequisite work complete, materials on site, submittals or shop drawings approved, permits and inspections signed off, equipment available, adequate manpower, and clear access to the area.
- Flag a constrained task so it's visibly not ready, assign an owner to clearing it, and set a need-by date that lands before the task is supposed to begin.
- Tell the difference between a task that's genuinely make-ready — every constraint cleared, screwdriver-ready — and one that just looks scheduled.
The lesson to drive home: an unconstrained task on the wall is a lie until proven otherwise. The discipline of walking every task and asking "what would stop this from starting?" is the whole game. This is the heart of Last Planner thinking, and it's the difference between a schedule that predicts the future and one that just decorates it.
Updating and rolling the window
A look-ahead that isn't updated on a rhythm is dead within two weeks. The update habit is a trainable skill, and it has a cadence.
- Daily-ish: mark progress honestly. Did the task complete? If not, why — and was the reason a constraint that should have been caught?
- Weekly: roll the window. Drop the closed week, add a new week on the back, pull the next slice of the master schedule down into detail, and run the constraint walk on the new work.
Verify the trainee can complete a full weekly roll without hand-holding: update actuals, add the new week, re-detail the incoming work, and re-check constraints. And verify they update it truthfully — the most dangerous foreman is the one who marks everything green because he thinks that's what you want to see. Tell them plainly that a missed task honestly recorded is useful data; a missed task marked complete is a landmine for next week.
Coordination and sharing with the trades
The plan only pays off when the people doing the work can see it. Verify the trainee can share the current week with subs and crew leaders, and — just as important — knows what to do when two trades collide in the same location on the same day. That conflict showing up on the wall is the tool doing its job; resolving it before Monday is the planner doing theirs.
Good scheduling software earns its keep here by making the plan visible on a phone in the field, so a crew leader doesn't have to call the trailer to find out where his guys are working. If your team uses the LookAheadWall mobile companion, this is where you confirm the foreman can actually pull up his week on his own phone, standing in the parking lot, without help.
Reading the numbers: PPC and variance
This is the last competency and the one that turns a schedule-user into a schedule-improver. Teach them to track Percent Plan Complete — the share of tasks they committed to that actually got done — and, more importantly, to read the reasons tasks slipped.
The number itself matters less than the pattern behind it. If PPC keeps cratering because materials show up late, that's a procurement fix, not a scheduling one. If it's prerequisite work never finishing on time, your durations upstream are fiction. Verify the trainee can pull a week's completion, tell you the top reason things slipped, and connect that reason to something they'd change next week. A planner who tracks PPC but never acts on the variance reasons is just keeping score.
When is training actually done?
Not when they've clicked every button. Training is done when the trainee can run a full weekly cycle unsupervised — read the current state, build and right-size the next week, walk and clear constraints, roll the window, share it with the trades, and read back what the last week's numbers are telling them. The honest final test isn't a checklist at all: hand them next week's plan, walk off the job, and see whether the crews know where to be Monday morning. If they do, your training took. If they don't, go back and find the step you rushed — it's almost always the constraint walk.