A Kill Decision Needs an Owner Who Is Not the Sponsor
Evidence and criteria do not stop a project if nobody holds the right to act on them. The kill decision belongs to whoever owns the trade-off, and the meeting should be built around that person.
A project survives because nobody holds the right to end it.
The earlier notes in this series gave leadership teams two instruments for stopping work: a condition written before the project starts and a framework for judging one already running. Both assume that once the evidence is on the table, somebody is able to act on it. In many companies nobody is. The evidence is assembled, the meeting is held, and the project continues, because the people in the room can recommend a stop and none of them can order one.
Where the right goes missing
Starting a project has a clear path. A sponsor writes a case, a budget holder approves it, and a code is opened. Ending one has no equivalent path, because the right to end it was never allocated. It sits by default with the sponsor, who is the person least able to use it. A sponsor who stops a project is announcing that the case they wrote was wrong, in front of the colleagues who approved it, and the incentive to find one more reason to continue is structural. It has little to do with character.
The other candidate is the steering group. A steering group can discuss, challenge and recommend. Because its members share the outcome and none of them owns it, a steering group decides by consensus, and a consensus about stopping is very hard to reach while any one member can name a reason to wait. A kill decision with no named owner defaults to continuation, because continuation is the only outcome that needs nobody's signature.
The same pattern appears wherever a programme has outlived its premise.
A pattern seen in a digital banking team shows the mechanism. A programme to rebuild part of a checkout flow had been approved on the belief that a simplified path would lift completion. Four reviews later the completion gain was small and the estimate to finish had moved twice. The sponsor was credible and the team was strong. At each review the steering group agreed that the numbers were disappointing, agreed that the direction was right, and agreed to look again next quarter.
Why the reviews did not decide
No one in the room was avoiding the question. The reviews were built to inform a decision that no attendee was authorised to take. The product lead could cut the scope but not the programme. The risk lead could delay a release but not end the work. The sponsor could end it, and had every reason not to. Each review produced a better-informed group and the same outcome.
The fifth review changed only one thing. The managing director attended, said at the start that the programme would be continued or stopped by the end of the hour, and asked for the three forward-looking answers from a finance partner instead of the sponsor. The programme was reshaped to the one part that carried the value, and the rest was released within the week. Nothing in the evidence was new. The right to decide had been placed in the room.
The right to stop belongs to whoever owns the trade-off.
If the sponsor is the wrong owner and the steering group is the wrong forum, the question is who holds the right. The answer follows from what a stop actually does. Stopping a project is rarely a verdict on that project alone. It frees people, budget and executive attention for something else, so the person who can use what is released is the person who should decide.
What the owner must be able to do
A usable owner can do three things that a sponsor or a steering group cannot:
- Reallocate the released capacity without asking permission from the project's own sponsor.
- Read the evidence without a stake in the case that created the project.
- Absorb the cost of the decision, including the sponsor's disappointment, without it ending a working relationship.
In most companies that is a portfolio owner: the managing director, a chief operating officer, or a head of strategy with authority over several initiatives at once. In a smaller company it is whoever sets the quarter's priorities. Priority lists grow because adding is cheap, and the same person who is able to say no to a new item is the one who can say it to a running one.
Delegation is allowed, ambiguity is not
The right can sit lower, for projects below a stated size, and it can be shared with a finance partner who holds the numbers. What it cannot do is stay unassigned. A rule that says "the portfolio owner decides, on the evidence the finance partner presents, at the quarterly review" is enough. A rule that says "the leadership team will decide together" returns the company to the fourth review.
The decision is made in one meeting, with three permitted outcomes.
Once the owner is named, the meeting becomes short. It has a fixed input and a fixed set of results, and it ends with a statement that can be repeated to anyone who was not there.
A fixed shape for the meeting
The input is the forward cost, the forward value and the work the project displaces, prepared by someone outside the project. The sponsor speaks after the numbers are read and not before them. The owner then names one of three results: continue with a new, narrower condition and a date, reshape to the part that carries the value, or stop. An extension granted because the team was close is a fourth result, and it is the one the format is designed to refuse.
Nobody should leave the meeting unsure whether the project is alive.
What the owner says aloud
The decision is closed when the owner states it in a sentence and the reasons in two more, which is the same discipline described in how a decision is closed in a leadership meeting. Then the practical questions are answered before anyone leaves: what is kept, where each person works from Monday, and who tells the stakeholders. A stop with those answers attached reads as a managed change. A stop without them reads as a failure, and the next sponsor learns to avoid proposing anything ambitious.
Naming the owner is where the thinking becomes work.
The practical start is a single page. List the initiatives currently running, and beside each one write the name of the person who could end it today without anyone else's agreement. Where the name is the sponsor, the right is probably unused. Where the cell is empty, the project is running on inertia.
That exercise, and the redesign of review cadence that usually follows it, is the kind of work our consulting practice does within the growth discipline pillar. It is often run as an outside read on a portfolio in which every project has a reasonable defence and no one is positioned to weigh them against each other.