Why the Same Decisions Keep Coming Back
Teams do not reopen decisions because they lack discipline. They reopen them because nothing about the original decision was ever closed.
A decision that keeps returning was never actually closed.
The complaint is always phrased the same way. We agreed this in March, we are arguing about it again in June, and nothing about the business has changed in between. The agenda item has a new name and the same content, and everyone in the room recognises it within a minute.
It is usually diagnosed as a discipline problem. People are told to commit, to disagree and commit, to stop relitigating settled matters. Sometimes a rule is added: decisions are final unless new information arrives. The rule holds for roughly a quarter, and then the same question reappears wearing a different title.
Not a failure of will
That framing is comfortable because it locates the fault in individuals rather than in the way decisions are made. A decision that keeps returning is not evidence of weak commitment; it is evidence that it was never closed in the first place. Closing is a specific act with specific parts, and most organisations perform only the visible one — announcing an outcome — while skipping the rest.
The announcement is the part everyone remembers. The meeting ended, someone summarised, heads moved, the deck was updated. What was not established in that room was who owned the choice afterwards, what it was chosen against, and what would have to be true for it to be looked at again. Without those, an announcement records the mood of a room on a particular afternoon, and moods are reversible.
Reopening is rational when nobody owns the outcome.
When a decision has no single owner, reopening it costs nothing. There is nobody to persuade and nobody whose standing is attached to the result. Anyone who was unhappy in March can raise it in June at the price of one agenda slot, and the expected value of doing so is positive, because they might win and they cannot lose.
The economics of raising it again
This is worth stating without moral colour. The person reopening the decision is not being difficult. They are responding accurately to a system in which objection is cheap and closure is free to overturn. In an organisation where reopening costs nothing, every decision is provisional, whatever the deck says.
The pattern shows up in a few recognisable forms:
- A decision taken in a forum nobody treats as authoritative, which then gets tested in a more senior one.
- A decision taken without the function that has to live with it, which arrives there as a fresh problem rather than an old answer.
- A decision recorded as an outcome with no reasoning attached, so the argument has to be rebuilt from the beginning every time it surfaces.
Ownership is not approval
Approval is a signature at a moment. Ownership is a person who will be asked, six months later, how the choice is going, and who therefore has reason to defend it, revise it deliberately, or say plainly that it was wrong. A decision with approvers and no owner has many people who can reopen it and nobody whose job it is to hold it shut.
The strategy was agreed; the trade-off inside it was not.
There is a second source, and it is the more expensive one. Many decisions that keep returning were agreed at the level of the statement and never agreed at the level of the trade-off. Everyone signed up to move upmarket. Nobody signed up to lose the small-customer revenue that moving upmarket costs, and that bill arrives monthly.
Agreement at the wrong altitude
Statements are easy to agree because they cost nothing at the point of agreement. The cost lands later, in a particular quarter, on a particular team, and the person carrying it reasonably asks whether this was really what was decided. It was. But the decision was made in a form that did not name what would be given up, so the argument about the price was deferred rather than settled.
An organisation rarely relitigates its decisions. It relitigates the costs it never agreed to pay.
This is why reopening tends to cluster rather than arrive evenly. It comes when the sacrifice becomes visible: the quarter the deprioritised segment misses its number, the month a migration slows a roadmap everyone still expects to ship on time. The decision looks sound on the slide and expensive on the floor, and the distance between those two views is where the meeting keeps restarting.
A named cost travels further than a statement
A decision that names what it gives up survives contact with the quarter in which the giving-up happens. One that names only what it gains does not, because the first person to pay the price has no record showing that the price was chosen rather than incurred by accident.
A decision closes when the conditions for reopening it are written down.
The practical correction is small, and it is not another forum. It is the habit of recording four things at the moment a decision is taken, in the same place each time:
- The choice, in a sentence that names what was rejected as well as what was chosen.
- The owner, one name, not a function and not a committee.
- The basis — what was believed to be true that made this the better option.
- The trigger — the observation that would justify opening it again.
The trigger is the part that does the work
The first three are common enough. The fourth is rare, and it is the one that converts a decision from provisional to closed. A stated trigger gives an objector something better than an agenda slot: a test. It turns “I still think this is wrong” into “the condition we agreed to watch for has occurred”, which is a conversation worth having and takes twenty minutes rather than a morning.
It also makes revision honourable. Teams reopen decisions badly in part because reopening carries a stigma, so it happens sideways — through erosion, quiet under-resourcing and delayed implementation rather than a clean second decision. A decision that says in advance what would change it can be changed without anyone having to lose.
The asymmetry underneath this is the same one that sits under one-way and two-way decisions: the expense is rarely in the choosing, it is in the conditions built around the choice. Most of the quality is set before anyone votes, which is the argument in the preparation that precedes a decision. The wider family of these patterns sits under decision architecture.
Holding decisions shut is where the thinking becomes work.
Nobody disagrees with any of this in principle. Every leadership team asked whether decisions should have owners and stated triggers says yes, and means it. The reason it does not happen is that closure has no home in the operating rhythm. The meeting produces minutes, the minutes record outcomes, and nothing in the cadence asks what would reopen this.
Where it actually breaks
So the work is to put closure somewhere it cannot be skipped: the same template, the same page, reviewed at the interval the numbers are reviewed. In practice that means fewer decisions taken per meeting and more of them surviving the quarter, which is a trade most teams accept once the arithmetic is in front of them.
That is the shape of the work we do with commercial and product teams. Not making the decisions, but building the conditions in which a decision made once stays made, and the ones that should change do so deliberately and in the open.