PLOT moves the space decision to the foremen. Here is why that makes decisions better — and what must be on the map for it to work.
There is a rule about organisations that is older than software and older than construction. The military learned it first, and gave it a name: mission command. The commander states the intent. The people on the ground decide the execution – because they can see the ground, and headquarters cannot.
The rule underneath it is simple: a decision belongs at the lowest level that can see everything the decision touches. Push it lower, and it becomes blind. Hold it higher, and it becomes slow and wrong.
The PLOT method is built on this rule. But I want to argue it carefully, because the easy version – “give control to the foremen!” – is wrong in three ways. This article is about the correct version.
1. First, credit where it belongs
Construction already had this insight once, and it deserves the credit.
Thirty years ago, Glenn Ballard built the Last Planner System around one observation: the plans that matter are made by the last planner – the foreman who assigns the work – and commitments are reliable only when the people who will do the work make them. Lean construction moved the time commitment down to the crews: what we promise for next week, we promise ourselves.
PLOT extends the same logic to the dimension Last Planner never touched: space. A crew can promise the week and still collide in the bay, because the weekly promise has no coordinates. The plot – a place, a date range, an owner, claimed by the crew itself is the last planner’s promise with ground under it.
So no, decisions-by-the-crews is not my invention. What is new is the dimension, and the map it happens on.
2. The inversion that is not really an inversion
Here is the trap in saying “PLOT hands control from management to the foremen”: it assumes management had that control. It did not.
Look honestly at how space is allocated on a site today. Whoever arrives first takes the spot. The trailer parks where it is convenient. The laydown grows because nobody said stop. The foreman decides – informally, invisibly, unaccountably – because he is the one standing there, and no office decision reaches the ground in time. Management holds a plan; the foremen hold the reality.
You cannot invert control that was never held. What PLOT actually does is different, and easier to say to a construction director with a straight face:
It takes the control foremen already exercise in the dark, and makes it legitimate, visible, and accountable.
Before PLOT: the foreman takes the space, and nobody knows. After PLOT: the foreman claims the space – publicly, in advance, with his name and his dates on it – and everybody knows. The decision stays where it always was. The visibility is what moves, and it moves up. Management does not lose control it never had. It gains sight it never had.
3. Why the foreman’s decision is better — under one condition
The case for deciding low is speed and truth: the two foremen whose plots collide can settle it in one phone call, days early, with full knowledge of their own works. No office can match that.
But local decisions have a known disease: they optimise locally. The two foremen’s neat swap can sit on the critical path, break a crane window, or push a third crew nobody phoned. Historically, this is exactly why decisions crawled up the hierarchy – not because managers are wiser, but because managers sat next to the information. The office had the schedule, the fabrication status, the delivery reports. The foreman had a radio.
This is the rule this article is named after: control follows information. A decision can only move down as far as the information moves down. Delegate without the information, and you have not empowered the field – you have abandoned it.
So the real question about PLOT is not “should foremen decide?” It is: what must the foreman see, for his decision to be better than the office’s?
4. What must be on the map
The answer is: both layers of the site, on one picture.
The temporary layer – the plots. Who has claimed what ground, for which dates. The foreman planning Tuesday sees the crew that plotted Monday, the crane radius over the bay, the hazard zone next door. This is PLOT’s core, and I have written about it from every side.
The permanent layer – with its true status. This is the half that usually gets forgotten, and it is the reason this article exists. The engineering model on the same map should show, for every tag, how the work actually stands: fabricated – delivered – erected. Colour on the model, straight from the project’s tracking systems.
Watch what the second layer does to the quality of the first decision. The foreman choosing where to claim space can now see that the neighbouring package’s steel is not even delivered – that bay will stay empty for weeks, claim it. Or that the piping crew’s materials landed yesterday – their plot is real, avoid it. Or that a work front everyone assumed open is, in fact, not erected. He is no longer negotiating space with a radio and a rumour. He is looking at the same truth the project controls office looks at – plus the one truth the office never had, which is where everybody will physically be.
That combination is the whole argument. Plots show who will be where. Status shows how the work truly stands. Together they put more information in front of the foreman than any office ever had. And that, and only that, is what justifies moving the decision to him.
Two honest warnings about the status layer. First, status data lives in fabrication and material systems of very mixed quality. Map that shows stale status teaches the foremen to distrust the whole picture, including the plots sharing the screen. The map must show the age of its data, and a project whose status systems are fiction should fix that before painting it on a wall. Second, to my AWP readers: PLOT does not want to manage your work packages. It displays their state as context, nothing more. The register stays about space, your packages finally land on real ground.
5. The boundaries – because “the crews decide” has limits
Subsidiarity is not abdication. Some space decisions must never be settled foreman-to-foreman, and the method has to say so in advance, not discover it in an incident report.
- Hazard interactions escalate by rule. A lift over a working crew, hot work near a live system, entry into a tested area – these are SIMOPS, and permit authorities exist precisely because peer negotiation is not an acceptable control for hazard decisions. The register’s job is to surface these overlaps automatically and route them to the authority that owns them – faster than any paper process, but never around it.
- Critical-path conflicts go to the phase planner. When a local swap touches a milestone or a shared resource like the main crane, the two foremen are no longer the lowest level that can see everything the decision touches – so, by the article’s own rule, the decision is not theirs. It goes to the tier that can still change the sequence this month, which is the phase planner, and the lookahead meeting exists for exactly this short list.
- Everything else stays low. Which is most of it. The register’s record from every industry that runs this pattern – and the bet PLOT makes – is that the overwhelming share of conflicts are exactly the kind two owners can settle, once they can see each other in time.
The same rule draws every line: the decision sits at the lowest level that can see everything it touches. The map’s job is to make that level as low as possible. The boundaries’ job is to catch the cases where it is not low enough.
6. What each side must give up
Honest inversion article require this section.
Management gives up the routine decisions — and gets exception-based control. They stop allocating ground they cannot see, and start seeing all of it: every claim, every conflict, every resolution, with history. They intervene where the rules route things to them, and audit everything else. This is not less control. It is control over the right things – the mission-command trade, and it has a century of evidence behind it.
Foremen give up the dark. This is said less often, so let me say it. An unregistered claim is an unaccountable one, and some of the field likes it that way. Plotting puts your name and your dates on the record; if you sat on a bay for three weeks you did not need, everyone saw it. The method’s answer is the same as always – plotting must pay the foreman back in permits, in space, in visible neighbours, or he will not plot. But the trade is real: the crews gain the decision openly, and lose the ambiguity. Most good foremen will take that trade.
7. An invitation
If you run lean or Last Planner and space is the dimension your weekly promises keep tripping over – this is your method’s missing half.
If you are a construction director and “the foremen decide” made you flinch in section 2 – you are exactly who I want to talk to, because the version where you gain sight while the field gains speed is the version worth building.
And if you are a foreman who read section 6 and thought “I would take that trade” – you are the person this whole method is for.
Control follows information. Put the information where the work is, and the decisions will follow it down – better, faster, and finally in the light.
