The first Last Planner meeting I ever ran went about forty-five minutes over, produced a list of "commitments" nobody actually meant, and ended with the drywall foreman staring at his phone while I read constraints off a laptop. It was a disaster, and it was entirely my fault. The software wasn't the problem. I was. Nobody had ever told me that facilitating a planning session is a skill of its own — separate from knowing the schedule, separate from knowing the trades — and that if you get it wrong, the best scheduling tool on earth just helps you fail faster.
This guide is what I wish someone had handed me that day. It's about running the weekly session well: how to prepare, how to keep it moving, how to pull real commitments out of people, and how to use your software to serve the conversation instead of hijacking it.
Your job is not to plan. It's to make planning happen.
The single biggest mistake new facilitators make is confusing themselves with the planner. You are not the person who decides the drywall crew will hang Zone B on Tuesday. The drywall foreman is. Your job is to create the conditions where he makes that call honestly, out loud, in front of the trades he's handing off to and receiving from.
Practically, that means you're doing five things at once: keeping the session on time, making sure the quiet people talk and the loud ones don't run the room, capturing what's actually agreed to (not what you hoped you heard), surfacing the handoffs between trades, and running the variance discussion so it teaches something instead of assigning blame. If you find yourself telling people what they're going to do, stop. You've stepped out of the chair.
The session is won or lost before anyone sits down
A session that runs clean almost always had thirty minutes of prep behind it. A session that turns into a swamp usually had a facilitator who walked in cold and tried to figure out the schedule live, in front of fifteen people. Don't do that.
Before the meeting, get these in order:
- Calculate last week's PPC before you walk in, not during the meeting. Percent Plan Complete — commitments completed divided by commitments made — is your headline number. Know it cold, and know which trades drove it up or down. Fumbling for it live kills your credibility in the first five minutes.
- Pull the two or three variances worth talking about. You will not discuss every missed commitment. Pick the ones with a pattern or a real lesson. Come in already knowing "materials showed up late three weeks running" so you can steer the room straight to the cause.
- Get the look-ahead current. Your rolling look-ahead — whether you run a three-, four-, or six-week horizon — has to reflect reality before the session, not get built during it. Updating a look-ahead live, while people watch, is the fastest way to lose a room. This is where a real short-interval scheduling tool earns its keep: in something like LookAheadWall you can have the weeks laid out visually so the walk-through is a review, not a construction project.
- Know your constraints and who owns each one. Every constraint — permit, submittal, RFI, material delivery, predecessor work, manpower, equipment — needs a name attached and a status. If you don't know the status before the session, you'll spend the session finding out, and that's dead time for everyone else.
- Test the tech and set the room. Screen up, everyone able to see it, schedule open to the right week. Sounds obvious. Watch how many sessions burn ten minutes on a dead HDMI cable.
A shape for the hour
Ninety minutes is plenty for a weekly session if you hold the boundaries. Here's the split that's worked for me on jobs from tenant improvements to five-story wood-frame multifamily:
- Opening — 5 minutes. Purpose, agenda, any announcements. Move.
- Review last week — 15 to 20 minutes. PPC, the variances that matter, what we learned.
- Look-ahead update — 20 to 30 minutes. Walk the horizon, update constraints, find work that's coming ready.
- This week's plan — 30 to 40 minutes. Pull commitments, work the handoffs between trades.
- Wrap — 5 to 10 minutes. Read the commitments back, assign the action items, confirm next steps.
Start on time even if half the room isn't there. If you wait for stragglers this week, you'll wait longer next week, because everyone learns that "starts at 7:00" actually means 7:12. Punctuality is a policy, and you set it in the first two sessions.
Running the review without turning it into a trial
Put PPC up where everyone can see it, and say the number out loud. Then, before you touch a single miss, name the wins — the trades that hit their commitments, the work that closed out clean. People will sit through hard feedback all day if you've shown you also notice when they deliver.
When you get to the misses, the whole game is asking "what happened?" without loading the question. There's a world of difference between "Why didn't you finish the risers?" and "The risers didn't close — what got in the way?" The first one makes a foreman defensive and he stops telling you the truth. The second one gets you the real answer, which is usually that a predecessor slipped or a submittal was still in review — information you need to fix the system, not the man.
And push past the first answer. "Material was late" is a symptom. "Why was it late?" gets you to "the PO went in three days after we identified the need," which is something you can actually change. Five whys is a cliché because it works. One warning: keep this tight. The review is where sessions bloat. If a variance needs a real investigation, park it and take it offline with the two people involved.
Walk the look-ahead from the back
When you update the look-ahead, start with the furthest-out week and work toward this one. It feels backwards, and that's the point — it forces the conversation onto making work ready instead of just reacting to what's in front of you. The whole purpose of a look-ahead is to find constraints early enough to clear them, so you spend your attention where you still have time to act.
For each activity out there, you're asking two things: what's the status, and what has to be true before this crew can start? That second question is where you make work ready. "What else needs to happen before framing can go vertical in Building C?" surfaces the inspection, the material, the predecessor, the crane window. Every answer becomes a constraint with an owner and a date. Constraints that aren't moving — the submittal that's been "in review" for three weeks — get flagged loud, because those are the ones that will eat your schedule if nobody chases them.
Pulling real commitments
This is the heart of the session and the part software cannot do for you. A commitment in the Last Planner sense is not an assignment you hand down — it's a promise a foreman makes, in his own words, after he's looked at his crew size and his constraints and decided he can actually deliver.
So you only ask for commitments on work that's ready — all constraints cleared. Asking someone to commit to work that's still blocked is how you manufacture broken promises. Then you ask, plainly: "What can your crew reliably get done this week?" Note the word reliably. You're not asking what's theoretically possible on a perfect day. You're asking what he'll stake his name on.
Three things make or break the commitments you capture:
- Kill the vague ones. "Make progress on the electrical" is not a commitment — it's a hedge. "Rough-in the east corridor, panels EA through EF, done Thursday" is a commitment. If you can't tell at week's end whether it was met, send it back for specifics.
- Work the handoffs out loud. This is the coordination that keeps trades from stepping on each other. "Maria's crew needs Zone A closed up by Wednesday to start finishes — does that land, or do we have a collision?" Half the value of the room is catching these before they become a field argument.
- Read it back. Every commitment, say it back the way you're recording it: "So that's east corridor rough-in, EA through EF, complete Thursday — you good with that?" It feels slow. It saves you the Friday argument about what was actually promised.
Capture each commitment as it's made, right in the plan on screen, so the record everyone leaves with is the record everyone agreed to. The value of doing it live in a shared weekly work plan — rather than scribbling notes to type up later — is that there's no gap between what was said and what's written, and the plan is ready to share the second the meeting ends.
Managing the room
Every session has the same cast of characters, and part of the craft is handling them without making it weird.
The dominant talker — the super or foreman who'd happily run the whole hour — gets a warm redirect: "Good, thanks Tom. Maria, you're downstream of that, how does it look from your side?" You're not shutting him down, you're moving the ball. The quiet foreman who nods at everything and commits to nothing needs a direct, gentle invitation: "Juan, this handoff runs right through your crew — what's your read?" Some of the best information in the room lives with the person least likely to volunteer it.
Side conversations get folded back in. Conflict — and there will be conflict, because two trades want the same space on the same day — you handle by naming both needs honestly and turning the room toward solving it rather than winning it. And when the tech glitches, which it will, you keep moving with a printout or a whiteboard. A facilitator who freezes because the projector died has told the room the meeting was really about the screen. It wasn't. It was about the commitments.
Let the tool serve the conversation
Good scheduling software is leverage, not a substitute for facilitation. Used well, it means everyone's looking at the same picture, you can move through the weeks fast because you know the tool, commitments land in the record the moment they're spoken, and the finished plan goes out to every sub before people have made it back to their trucks. When you've got crews or subs joining remotely, screen-sharing the live plan and deliberately calling on the remote folks by name keeps them in the room instead of on mute.
But hold the line on what the software is for. It captures and displays; it does not create the commitment or clear the constraint. I've watched facilitators disappear into the screen — clicking, formatting, tidying — while the actual planning conversation drifted sideways behind them. Learn your tool well enough that you're barely looking at it, so your eyes stay on the people. The plan is a byproduct of a good conversation. Run the conversation.
A working checklist
Keep this taped to the inside of your session folder.
Before:
- PPC calculated, the two or three variances worth discussing identified
- Look-ahead updated to current reality
- Every open constraint has an owner and a status
- Screen tested, plan open to the right week
- Attendance confirmed, prep reminders sent
During:
- Started on time, no waiting on stragglers
- Wins named before misses
- Every quiet foreman spoken to directly
- Commitments specific, measurable, read back before they're recorded
- Every trade-to-trade handoff surfaced out loud
- Variance discussion found a cause, not a culprit
After:
- Plan distributed to all trades immediately
- Action items assigned with names and dates
- Two minutes of honest reflection: what dragged, what to tighten next week
You get better one session at a time
Nobody walks into their first Last Planner session and runs it well — I certainly didn't. Facilitation is a muscle. Watch a superintendent who's genuinely good at it and steal their moves. Ask a foreman you trust, after the meeting, what dragged. And take two minutes when you're back in the trailer to think about what you'd do differently. Do that for a few months and you'll notice the room change: commitments get sharper, PPC climbs, and the trades start bringing you constraints before you have to dig for them.
That's the real payoff. The software gives you the structure and keeps the record straight. The facilitation is what turns a status meeting into a room full of people who actually mean what they just promised — and that, far more than any feature, is what makes the plan hold.