The Reasoning Goes First
The decision usually travels. The argument behind it almost never does, and it is the part worth keeping.
three things · two reached them · one with no channel at all · the room side is not counted
Sorted by whether anything could have carried it.
What is worth keeping about a decision, and how much of it actually reached the person outside the room
All four were written down, usually within a day, usually by the same person
One of the four, and only because somebody happened to disagree in writing
The thing most needed six weeks later, and the channel for it was never once used
Carried by tone and by who stopped arguing, and nothing holds either
Nobody states a confidence level out loud; the room reads it from the voice
Two of the four, both written by the same person on their own initiative
six things · three reached them · two with no channel at all · the room side is not counted
Six things worth keeping about every decision, and the record holds two of them reliably. The left side is not counted: a room transmits all six continuously and nobody registers having received anything. One manager's own project, reviewed at the end of a quarter; which of this any team records differs by team and by what it has agreed to write down.
The decision gets written down. The reason it was taken does not, and the reason is the part that stops being recoverable first — including for the people who were in the room.
The missing context described in “The Reasoning Goes First” is an operating-process gap. Teams considering review the provider’s own material for capital efficiency ratio can attach clearer time and project evidence to work already being discussed, provided decisions, reasons and corrections remain visible to people who were not present.
Six weeks later nobody can reconstruct why, and the decision becomes something the team obeys without understanding.
The questions raised by “The Reasoning Goes First” can also be compared with Idealist nonprofit career resources. A useful review identifies a responsible owner and a practical change rather than copying a generic checklist.
What a bare decision is good for
Compliance. You can act on it, and you can tell whether somebody has complied with it.
That is the whole of its usefulness.
What it is not good for
Anything requiring judgement at the edges. A decision without its reason gives no guidance about cases its author did not anticipate, which is most cases.
Somebody working at a distance meets those edges alone and unobserved, with no way to ask the room what was meant. They either guess, which produces work that has to be redone, or they ask and wait, which produces delay. In a room they would have read the answer off somebody's face in four seconds.
Why reasons decay faster than decisions
The decision has a form — a sentence, a ticket, a line in a document. The reason was a conversation, distributed across four people, none of whom holds the whole of it.
Within a month each of them remembers a different part.
The two versions of the same record, six weeks later:
We will use the batch import rather than the live feed.
We will use the batch import rather than the live feed, because the client's security review will not be scheduled before March and the live feed needs it.
The cargo-cult failure
A decision whose reason has been lost gets applied in situations it was never meant for, because nobody can tell which situations those are. Teams accumulate rules that made sense once.
This is the single clearest long-run cost of not recording reasoning, and it is paid by everybody rather than only by the person outside the room. The rule survives the circumstance that produced it, and the longer it survives the more authority it acquires, because its age starts to look like evidence.
The three parts worth capturing
What was decided. What was rejected and in whose favour. What constraint made the difference.
The third is the one that actually earns its keep.
The constraint is the useful part
"We chose B" is weak. "We chose B because A needed a decision from the client we were not going to get before the deadline" is strong, because it tells you exactly when B stops being the right answer.
It also tells somebody six months later whether the constraint still holds.
Why people resist writing it
Because at the moment of decision the reason feels too obvious to record. Everybody present knows it, and writing it down feels like explaining something to people who already understand.
That feeling is reliable and it is always wrong. The audience for the reason is not the room now; it is the room in six weeks, plus the person who was not there, and neither of those audiences shares the obviousness.
The two-sentence version
One sentence for the decision, one for the binding constraint. That is the whole mechanism and it costs under two minutes.
Anything longer will not be written at all.
Whose job it is
Whoever chaired, or failing that whoever is most senior in the conversation. Delegating it to the person outside the room is a common and unpleasant failure: it makes the person who lost the information responsible for collecting it.
If nobody is named, it does not happen.
Capturing disagreement
If somebody was unconvinced, one clause saying so is worth more than the rest of the record. It tells a reader that the decision was close.
It also means the dissenter does not have to raise it again from scratch when circumstances shift.
Reasons that cannot be written down
Some decisions rest on something nobody will put in a shared document: a client who is difficult, somebody's capacity, a commercial position that is not public.
Say so explicitly rather than inventing a thinner reason. "Decided for reasons I cannot write here, ask me" is honest and is far better than a plausible substitute, which will be acted on as though it were the real constraint and will mislead somebody at the edges six weeks later.
Where to put it
Next to the work, not in a separate archive. A decision log nobody opens is a record that technically exists.
The test is whether somebody joining the project in month four would find it without being told it exists.
What this does not deliver
The texture. A written reason is a reconstruction, made by one person, flattened into prose, and it loses the hesitation, the competing view that was never quite articulated, and the sense of how settled the matter actually is.
Somebody reading it will know what was decided and why, and will still not know whether the team believes it. That last part travels by tone and by nothing else, and the honest position is that it does not travel at all. What is being proposed here is a substantial improvement over nothing and a long way short of having been there.
Next door
- That a Conversation Is Happening at All
- Who Was Persuaded and Who Was Outlasted
- That Somebody Is Struggling and Has Not Said So
- The Priority That Changed Unannounced
- What the Client Actually Said
- Who Got Asked, and Who Got Overlooked
- Sorting the Losses Into Three Piles
- The Decision That Was Made in Passing