It Breaks in the Joints

Somewhere in your organization there is a report that is completely accurate and completely useless.

It was researched properly. The numbers are right. It went to the correct distribution list. And it arrived after the decision it was meant to inform had already been made, or worse, after the moment when making it would have mattered had passed unnoticed.

Nobody did anything wrong, and that is the entire problem. The person who wrote it did good work. The person who received it read it. The information moved. It simply didn’t reach a decision while a decision was still available.

Information, decisions and execution are usually treated as three separate concerns, owned by three different parts of an organization, improved by three different initiatives. They are not three things. They are one movement with three stages, and it almost never breaks inside a stage. It breaks at the joints.

Information should move, not accumulate.

Most organizations are not short of information. They are drowning in it. Dashboards nobody opens, reports nobody acts on, data lakes filling steadily with things no decision will ever draw from. The scarcity is not information; it is information arriving where a decision is being made, at the moment it is being made.

An organization can double its reporting and become slower, because the cost of information is not the gathering. It is the reading, the interpreting, and the deciding whether it changes anything — and that cost is paid by exactly the people who have the least time.

The useful question is never do we have this information. It is whose decision does this reach, and when.

Decisions create momentum when they’re made.

Every organization has an unspoken clock, and almost nobody measures it: the time between a decision becoming necessary and that decision being made.

It’s a strange number, because it belongs to no one. The people waiting are not idle — they’ve moved to something else, which is precisely why the wait feels free. Nobody bills for waiting. There is no line in any budget for a decision that took eleven weeks, and no report that shows what the organization couldn’t do during them.

A useful test: ask how long the last significant cross-team decision took, then ask how long the deciding took. In a healthy organization those numbers are close. In most, one is measured in minutes and the other in months, and the gap is not deliberation — it’s scheduling, preparation, and waiting for the right people to be in the same place.

The delay is rarely caused by the difficulty of the decision. It’s caused by uncertainty about who gets to make it, which is an Ownership problem arriving in disguise.

Execution is the only stage anyone measures.

Which is why it’s the stage least often at fault, and the stage that gets the most improvement effort. Delivery metrics, velocity, throughput, utilization — all of it pointed at the one part of the loop that was usually working.

An organization can get significantly better at execution and not move any faster overall, because the constraint was never there. It was three weeks upstream, in a decision that hadn’t been made, waiting on a report that arrived correct and late.

The loop closes or it isn’t a loop.

Execution produces an outcome. Someone learns something from it — what worked, what didn’t, what the estimate missed. If that never travels back to inform the next decision, the organization has a line, not a loop. It will make the same decision again with the same information it had the first time.

That last stage is the shortest, the cheapest, and the one skipped almost universally. It’s the subject of the next chapter, and the next chapter is deliberately brief.

None of the three stages is where flow dies. It dies in the space between them: information that never reaches a decision, decisions that never reach execution, execution whose outcome never reaches anyone. Those spaces belong to no one, appear on no org chart, and are therefore nobody’s job to fix.

This is the sap. Everything else in Part II describes the tree; this chapter describes the thing actually moving through it — and a tree that stops circulating dies standing up, long before it looks dead from outside.

It almost never breaks inside a stage. It breaks at the joints, which is precisely where nobody owns it.

What moves this is not improving any one stage, which is what each function does and why the joints stay untouched. It is making the joints somebody’s — routing information to the decision it serves rather than to a distribution list, moving decisions to where the knowledge already sits, and sending outcomes back to whoever chose. Each of those is a small change at a boundary rather than a programme inside a department.

Cultivating this: Closing the Loop, in Part III.