What it costs and what it buys

A team that starts naming commitments notices the change within about a fortnight, and not usually as an improvement. The first thing it buys is discomfort, because the things it makes visible were being avoided on purpose. What comes after that is worth the fortnight.

What a fortnight of it buys

"I'll look into it" is the most expensive sentence in professional life. It is not a promise and not a refusal; it is a way of ending a conversation without answering it. There is no act for it here, so the question gets asked instead: promising, with a date? Declining, so I can go elsewhere? Countering — not that but this, not then but then? All three are fine answers, two of them are refusals, and a team that cannot say them out loud is a team whose plans are made of guesswork.

Ask a room who owns a piece of work and you get a team name. Teams do not promise; people do — and a named performer can decline, where a team has nobody in it to do the declining. That is the uncomfortable corollary and it is not optional: if declining is career-limiting in your organisation you do not have commitments, you have assignments wearing the language of commitments, and they will keep being late for reasons nobody will say aloud.

And most rework traces back to a conversation nobody had at the start, about what would count as done. Not a specification — a sentence. Agreeing it beforehand costs ten minutes and feels like pedantry; discovering afterwards that the two of you meant different things costs a week and feels like somebody's fault.

Each of those is one quarter of the loop failing, and where a promise comes apart takes them apart properly, one per quarter, with the acceptance that never comes. The rest of this page is what they cost a team to put right, which is the less comfortable question.

Delay becomes traceable

A promise is usually made of other promises. Once those are drawn as connected loops, a date has a derivation instead of an origin story: you can follow a slipped week to the person waiting on the person waiting on an answer nobody has given. That is a different conversation from asking a team why it is behind, and it has a different outcome, because it ends with something somebody can actually do.

Now the bill

It makes accountability visible, and some of that visibility will be unwelcome. Drawn honestly, a network shows work nobody asked for, approvals no person stands behind, and promises made by people who were not free to decline. Those are findings about the organisation, not about the notation, and somebody will prefer they had stayed unavailable.

It can also be used badly, in a specific way worth naming — and it is not the way usually offered. The usual worry is surveillance: write the promises down and somebody has a record to hold people to. That worry has lost its force, for reasons set out in what it does to software, and the short version is that the record exists whether or not anybody draws anything.

What actually goes wrong is imposition. Handed down — draw your work this way, use these words, here is the template, it is due Friday — the vocabulary becomes something people comply with, and compliance is the one thing this notation is least able to describe. It works when somebody picks it up because they have a coordination problem they cannot see, and it is close to worthless as a standard that has been rolled out. The tradition has been criticised for exactly this since the 1990s, and the criticism lands every time.

The last cost is the plainest: it takes talking. This notation makes it obvious when a conversation is missing, and cannot have it for you. Nothing on this site improves any coordination at all until two people say something to each other they were previously leaving implied.

A reasonable way to start

Not with a meeting about notation. Take one thing that is currently late, or one handover everybody complains about, and draw it on your own, as it actually happened rather than as the process says. If two of the four moments turn out to have no answer, you have found the problem, and it was never the one on the status report — where a promise comes apart has the questions to walk it with.

Then show it to the one other person who is in it. That conversation is the whole method. Everything above is about whether an organisation can survive having it.