The org chart nobody wrote down

Two layers exist in every project. Only one of them is being automated.

Someone says “sounds good” and you know immediately that it isn’t.

Maybe they were looking somewhere past your shoulder as you finished. Maybe there was a half second of hesitation before the answer. Nothing you could point to, nothing you could put in an email, but you leave the meeting knowing that what they meant was they weren’t listening and they wanted this to end.

That gap between what gets said and what gets meant is where most of the actual work happens.

Every project runs on two layers

There’s the documented layer. Who owns what, how approvals flow, what the process document says, what the RACI claims about escalation paths. All of it is written down somewhere, most of it is accurate in the way a map is accurate, and a model can absorb it faster than any human.

Then there’s how things work.

You don’t escalate that issue directly, you mention it to Sarah first because she has history with that stakeholder. When the project manager says “we’re tracking to plan” without looking at anyone, the timeline is cooked and everybody in the room knows it. “Let’s revisit this” from the VP means the thing is dead and you should stop spending energy on it. The same phrase from the PM means not this month.

None of that is documented. You learn it by being present, watching what happens, getting it wrong, and dealing with the consequences of getting it wrong.

Why one layer is disappearing and the other isn’t

The documented layer is being automated quickly. The analysis, the first draft of the specification, the workshop summary, the status report nobody reads. That work is getting cheaper by the month and I don’t expect it to reverse.

The undocumented layer resists automation for a structural reason that has nothing to do with capability. The training data doesn’t exist.

Nobody wrote down that Friday is the wrong day to push for changes with this team. Nobody logged that this particular stakeholder goes quiet when he disagrees rather than pushing back, which means silence from him is a warning rather than consent. Nobody recorded that the finance director will approve almost anything framed as risk reduction and almost nothing framed as opportunity, or that the reason two teams don’t cooperate goes back to a reorganisation four years ago that neither of them will mention.

This information exists only in the heads of people who were there. It’s transmitted by apprenticeship and by failure, and it decays when people leave.

What makes this hard to build deliberately

The obvious response is to say “go and learn the informal dynamics”, which is useless advice, because the informal layer isn’t a body of knowledge you can study. It only appears under conditions where the formal layer is inadequate.

Which means clean projects teach you very little. When ownership is clear and stakeholders agree and the process works, you never find out what the organisation does when those things break, and that’s the only information worth having.

A few things that have worked for me.

Volunteer for the messy ones. When nobody’s sure who owns what and the stakeholders openly disagree, the real structure of the place becomes visible for a few weeks. That’s the closest thing to an x-ray you’ll get.

Follow up on the silences. Somebody clearly disagreed in a meeting and didn’t say why. Catch them afterwards. “I noticed you had reservations, what am I missing?” That single question has taught me more about how teams operate than any process document I’ve read.

Notice who you go to when you’re stuck. It’s usually not your manager. It’s the person who knows how things really get done, and there are typically three or four of them in any organisation, none of whom appear at the top of a hierarchy. That’s the real chart.

Study your own mistakes closely. You escalate to the wrong person, push at the wrong moment, misread a room. The sting of that is the lesson arriving, and if you can stay with it long enough to be specific about what broke, you’ll know something next time that isn’t available any other way.

The uncomfortable part

I should be honest that this knowledge is less portable than I’d like it to be.

Some of it transfers. The general skill of reading a room, noticing who has informal authority, understanding that requests carry unstated constraints, all of that moves with you. But a decent proportion of what makes someone effective in a particular organisation is specific to that organisation, and it evaporates the moment you leave. Which is part of why capable people sometimes have a rough first six months somewhere new and slowly wonder whether they were ever any good.

You were. You just lost your map and you’re drawing a new one.

The documented work is becoming table stakes. The other half takes longer, doesn’t travel intact, and remains the thing that determines whether projects live or die.