The first time I saw a Percent Plan Complete number posted on a jobsite trailer wall, it read 47%. The super who put it up wasn't embarrassed by it. He was proud, because two months earlier it had been 31%, and every point of improvement had come from a real conversation about why work didn't happen. That's the whole point of PPC. It isn't a report card you send to the owner. It's a flashlight you shine on your own planning to find out where it's leaking.
PPC is the signature metric of the Last Planner System, and it's about as simple as a metric gets: the percentage of the commitments your crews made for the week that actually got finished. Made 20 commitments, completed 16, that's 80%. What makes it powerful isn't the arithmetic. It's what the number forces you to do with the misses.
What PPC Actually Measures (and What It Doesn't)
Here's the distinction that trips people up. Traditional scheduling measures variance against a baseline: are you ahead or behind the dates in the CPM schedule? That tells you how late you are. It does not tell you why, and it does not tell you how to stop being late.
PPC measures something different and, frankly, more useful week to week: the reliability of your own planning. Did the work you promised to do actually get done? You can be behind a brutal baseline schedule and still have high PPC, which tells you your crews are executing well against a hard target. You can be ahead of an easy baseline with terrible PPC, which tells you you're coasting on slack and your planning is a mess that will bite you the moment the schedule tightens. The baseline tells you where you are. PPC tells you whether you can trust the plan that's supposed to get you the rest of the way.
That's why it belongs to the weekly work plan, not the master schedule. The commitments come out of your short-interval planning session, the review happens at the end of the week, and the number falls out of it automatically.
Count It Honestly or Don't Bother
PPC is only worth tracking if you count it straight, and the temptation to fudge it is constant. A few rules keep it honest:
- Completion is binary. A task is done or it isn't. No partial credit. "Drywall is 90% hung on the third floor" counts as zero for the week. This feels harsh the first few weeks and then it becomes the most valuable discipline in the whole system, because 90%-done tasks are exactly the ones that quietly wreck next week's plan.
- The commitment has to be specific. "Work on the east wing" is not a commitment you can score. "Rough-in plumbing, units 201 through 208" is. If you can't tell at Friday's cutoff whether it happened, it was never a real commitment.
- Measure against the original commitment, not the excuse. The weather being lousy doesn't turn a miss into a hit. It turns it into a miss with a reason code, which is a different and more useful thing.
- Same cutoff, every week. Pick a time, usually end of day Friday or first thing at the Monday planning meeting, and never move it. Consistency is what makes the trend line mean anything.
The number itself matters less than the questions it forces. Which brings us to the part almost everybody skips.
The Real Work Is the Reasons for Variance
PPC by itself doesn't improve anything. If all you do is post 68% on the wall and move on, you've wasted the exercise. The improvement lives in the analysis of every single miss. For each incomplete commitment, you ask one question: why didn't it happen? Then you tag it with a category. Over a construction career I've seen these show up again and again:
- Predecessor not complete. The work in front of yours wasn't finished, so you couldn't start. This is the most common miss on most jobs, and it's a make-ready failure, not an execution failure.
- Materials. The material wasn't on site, wasn't staged where the crew needed it, or showed up wrong. A procurement and logistics problem, not a crew problem.
- Prerequisite work / inspection. Ready to close the wall but the rough-in inspection hadn't happened yet. Ready to pour but the reinforcing hadn't been signed off.
- Manpower. The bodies weren't there. The sub pulled the crew to another job, or somebody didn't show.
- Equipment. Lift was down, the pump didn't show, the crane was booked by another trade.
- Information. An open RFI, a missing detail, a conflict in the drawings nobody resolved.
- Over-commitment. The foreman promised more than a realistic crew could physically do in five days. This one is on planning, and it's fixable with a couple of honest conversations.
- Weather. The genuinely unavoidable one, though it's smaller than people claim once you separate it from "we could have moved this work indoors and didn't plan to."
- Changed scope. A directive or design change landed after the commitment was made.
One incomplete task and one reason code doesn't tell you much. Forty of them over six weeks tells you almost everything. When you aggregate the reasons and the pattern is 60% "predecessor incomplete," you don't have an execution problem. You have a make-ready problem, and no amount of yelling at foremen will fix it. You fix it upstream, in the look-ahead, by driving out constraints before work ever hits the weekly plan. This is where a tool that connects the look-ahead to the weekly plan and rolls up the variance codes for you earns its keep. LookAheadWall's trade-flow sequencing is built around exactly that handoff, so a predecessor slipping shows up as a downstream risk before it becomes a Friday miss. But the discipline matters more than the software. A whiteboard and honest tally beats a fancy dashboard nobody fills in truthfully.
What the Numbers Roughly Mean
Teams always want to know if their PPC is "good." It depends heavily on how mature your process is, but here's the honest read from the field:
- 50–60%: Normal for a team that just started Last Planner. Don't panic. This is the baseline reality of how unreliable most weekly planning actually is once you finally start measuring it. The upside is enormous.
- 60–70%: You're making progress, but your make-ready process still has real gaps. Look hard at predecessor and materials misses.
- 70–80%: Solid. Most of the constraint-management basics are working. You're squeezing out the remaining noise.
- 80–90%: Strong, reliable planning. This is a good target for a functioning short-interval scheduling process.
- Above 90%: Excellent, but stay suspicious. Sustained mid-90s sometimes means the crews are sandbagging, committing to less than they can do so they always hit. Reliable is the goal, not perfect. If you're at 95% every week, push the crews to commit to more.
And watch the trend more than the absolute number. A team climbing steadily from 55% to 75% over two months is in far better shape than a team that's parked at a flat 80%.
How PPC Gets Gamed
The moment PPC becomes a target that leadership beats people over the head with, people learn to game it instead of improving it. Three tricks to watch for:
- Under-committing. Promising fewer, safer, easier tasks so the percentage stays high. Your PPC looks great and your job is falling behind. This is the most damaging form of gaming because it looks like success.
- Redefining "done" after the fact. The task suddenly meant something smaller than what everyone understood on Monday.
- Quietly dropping commitments. Erasing the missed task from the board instead of marking it incomplete, so it never counts against the number.
The fix is cultural, and it comes from the top. The superintendent has to make it dead clear that PPC is not a personal scorecard, that a low number honestly reported and analyzed is worth far more than a high number that's been massaged. The day a foreman feels safe saying "I missed three commitments, here's why" is the day the system starts working.
Putting It to Work in the Weekly Meeting
PPC lives or dies in the Monday planning session. A workable rhythm:
- Post last week's PPC where everyone can see it. Number on the wall, no hiding.
- Walk the misses, not the hits. Spend your time on what didn't happen and why. Two minutes each, no blame, just the reason code.
- Name the pattern out loud. "That's the third week in a row we're waiting on the electrician's rough-in. Let's fix the sequence." Patterns are invisible until somebody says them.
- Adjust next week's commitments accordingly, and make sure every task on the new plan is genuinely make-ready before anyone commits to it.
- Get real commitments. A commitment is a promise between people in the room, not a date the office typed into a schedule. That distinction is the heart of Last Planner.
Break PPC out by trade or by sub, not just as one project number. Aggregate PPC tells you the job's overall health. PPC by trade tells you which sub keeps their word and which one needs a harder conversation, and it gives you the evidence to have it. When you can show a sub their own reliability number next to everyone else's, the conversation stops being an argument about opinions.
Why It Correlates With Everything That Matters
Reliable weekly execution isn't just tidy. It ripples into every outcome you actually care about. Work that starts on time and finishes on time keeps the trade behind it from stacking up, which keeps crews out of each other's way, which is a big chunk of jobsite safety right there. Preparation done properly means less rework and less overtime, which shows up in the cost report. And a job where the plan can be trusted is a job where people stop firefighting and start building.
None of that comes from the metric itself. It comes from the weekly loop the metric drives: commit, measure, find the reasons, fix the system, commit again. Track PPC honestly, chase the reasons behind every miss, and change the process that produced them. Do that for a few months and the number climbs on its own, because you've quietly rebuilt the way the job plans its work. The percentage was never the goal. Reliable work was. PPC is just how you know you're getting there.