I've sat through a lot of Last Planner sessions that went nowhere. Fifteen people crammed around a plywood table in the trailer, the GC's super reading commitments off a printout, and everybody nodding along while quietly planning to do whatever they were going to do anyway. The board looked full. The schedule looked collaborative. And come Thursday, half the tasks slipped and nobody was surprised.
The tool wasn't the problem. Sticky notes, a whiteboard, a proper look-ahead app, it doesn't matter much. What kills these meetings is bad facilitation. The Last Planner System lives or dies on whether the people in that room actually make real commitments to each other, out loud, and mean them. That's a human skill, and it's one most of us never got trained on. We got handed the role because we've been on jobs the longest, and we figured it out or we didn't.
This is the stuff I wish somebody had told me before I started running weekly work plan meetings. It's not soft-skills fluff. Run these sessions well and your Percent Plan Complete climbs, your subs start trusting the schedule, and Friday stops being a fire drill.
Your job is to run the room, not to run the schedule
The single biggest mistake new facilitators make is treating the meeting like a status report they deliver. You stand up, you go down the list, you tell everybody what they're doing next week. Wrong. If you're the one making the commitments, the trades never own them.
The trades make the plan. You make the plan possible. Your actual jobs in that room are narrow and they're hard:
- Keep the session moving so you're done in 45 minutes, not two hours.
- Get every foreman to actually speak, not just the loud ones.
- Pull the constraints out of people before those constraints become Thursday's excuse.
- Write down who committed to what, by when, in words everyone agrees on.
Notice what's not on that list: knowing every detail of the drywall scope. You don't need to. The drywall foreman knows it better than you ever will. Your value is the process and the coordination between trades, not the content inside any one trade's work.
Ask about constraints before you ask about work
Here's the question that separates a real look-ahead meeting from theater. Not "what are you doing next week," but "what do you need in place before you can start, and is it there yet?"
That's the whole game. The Last Planner distinction between should, can, and will comes down to constraints. A task should happen per the master schedule. It can happen only once its constraints are cleared, material on site, area released by the prior trade, inspection passed, RFI answered, manpower available. It will happen when a foreman looks you in the eye and commits to it as a sound assignment.
So make constraint-hunting the spine of the meeting. Walk the three-week look-ahead task by task and ask the annoying questions:
- "Is your material on site, or on a truck, or still a PO?"
- "Is the area actually going to be released by the trade ahead of you? Framing, you good with that date?"
- "Does this need an inspection to close before you start, and who's calling it?"
- "What's the one thing that would blow this up, and how do we handle it now?"
A task with an open constraint does not go on the weekly work plan as a commitment. It goes on a constraint log with an owner and a need-by date, and it stays off the "will" list until it's clear. This is the discipline that makes the whole thing work. A commitment you make on top of an unresolved constraint isn't a commitment, it's a hope, and hopes don't build buildings.
Make the quiet foreman talk and the loud one wait
Every planning meeting has the same cast. There's the sub who fills every silence, usually the one whose scope is loudest right now. There's the veteran foreman who's seen it all and says almost nothing, and who happens to know exactly why your ceiling grid date is fantasy. If you let the room run itself, you hear from the first guy and miss the second one entirely. And the second one is the one who saves you.
For the dominators, acknowledge and redirect. "Good, I've got your points, let me make sure I hear from the other side of the table." Don't shut them down, they're often right, just stop them from being the only voice. If a trade's issue is real but off-topic for the group, park it. Write it on a visible list and tell them you'll grab them after. That one move, the parking lot, saves more meeting time than anything else I know.
For the quiet ones, you have to go get them by name. "Mike, you've run this sequence before, what am I missing on the layout?" A direct, specific question aimed at their expertise gets an answer where a general "anybody else?" gets silence. If you've got a foreman who genuinely won't speak up in a group, catch him on the walk beforehand or let him hand you constraints in writing. The goal is the information, not the performance.
Handle conflict instead of smoothing it over
When two trades disagree about a date or an area handoff in the meeting, that's not a problem. That's the meeting doing its job. That collision would have happened on site next week, except now it's happening at a table where it costs nothing to fix.
The move is not to jump in with your own answer. It's to make both sides state what they actually need. Nine times out of ten the fight is over something narrower than it sounds. The MEP guy doesn't need the whole floor, he needs the north bay for two days. The framer isn't refusing to leave, he's waiting on a strap detail from the RFI. Get the real interests on the table and the overlap usually appears. When it genuinely doesn't, you make the call, or you name who has to make it and by when, and you move on. What you never do is let it dissolve into vague agreement so the meeting can end. Vague agreement is exactly what falls apart Thursday.
Protect psychological safety like it's a schedule constraint
The Last Planner System only works if people tell you the truth about what's not ready. And people only tell you the truth if telling the truth is safe. The first time a foreman admits "we're not going to make Wednesday" and gets chewed out in front of the room, you've taught everybody to stop admitting it. From then on you get comfortable-sounding commitments that mean nothing, and you find out the real status when the work doesn't happen.
So when a task misses, the question is never "why did you fail." It's "what got in the way, and what do we change so it doesn't happen again." That's the weekly PPC and reasons-for-variance conversation, and it's the most valuable five minutes of the meeting. Treat a missed commitment as data about your process, not a personal failing, and you'll get honest planning. Treat it as a chance to assign blame, and you'll get theater. Model it yourself, too, admit when the GC dropped a ball. Nothing loosens up a room faster than the super owning his own miss first.
Keep the plan visible and keep it moving
The schedule has to be up where everyone can see it and point at it, not buried in a laptop only you can read. When a trade commits, the change goes on the board in real time, in front of the room, so the commitment is public and shared. This is where a proper look-ahead tool earns its keep over sticky notes, an app like LookAheadWall keeps a location-based weekly work plan on the screen, lets you drag a task and re-sequence the trade flow live, and pushes the updated plan to the crew leaders' phones the second the meeting ends so nobody's working off a printout that's already wrong. But the visibility is the point, not the software. A whiteboard everyone can see beats a slick app nobody's looking at.
On timing: guard the clock hard. Set a start time and actually start, even if two subs are late, they'll be on time next week. Timebox discussions. If a topic's eating the room, park it. A weekly work plan meeting should run 30 to 60 minutes. If yours runs two hours, you're either solving problems that belong in the parking lot or you're doing detailed coordination that should happen in a smaller pull-planning session with just the affected trades. Protect people's time and they'll keep showing up engaged. Waste it and you'll get bodies in chairs and brains checked out.
Close by nailing down who does what by when
Never let the meeting drift to an end. Land it. In the last few minutes, read back the commitments, plain and specific: "So next week, framing finishes the north wall by Tuesday, that releases the east rooms, MEP rough starts there Wednesday, and drywall's staged but not committed until we clear the inspection Thursday. Everybody good with that?" Then wait for actual verbal agreement, not nods.
That read-back does two things. It catches the misunderstanding that was hiding all meeting, the two trades who thought "Wednesday" meant different things. And it turns a discussion into commitments people heard themselves make. A commitment made out loud, in front of peers, gets kept at a rate a task silently assigned on a printout never will. That's the entire mechanism of the Last Planner System, and it happens in the last five minutes if you run them right.
You'll get better at this the same way you got good at building
Nobody's a natural at this. The first dozen meetings you run will feel clumsy, you'll talk too much, you'll let the loud guy steamroll the room, you'll forget to log a constraint that bites you on Thursday. That's fine. Watch a super who's good at it and steal what works. Ask a foreman you trust how the meeting landed. Track your PPC week over week, it's the honest scorecard, and when it climbs you'll know the facilitation is working, not just the schedule.
The short-interval planning tools matter, and good software makes the mechanics easier, keeps the plan visible, connects the trade-flow sequences, gets the weekly plan onto every crew leader's phone. But the meeting itself is a human thing. It's a room full of tradespeople deciding, together and out loud, what they can really promise each other for the week ahead. Your job is to make that conversation honest, fast, and specific. Get that right and the plan holds. Get it wrong and no app in the world will save your Friday.