“We’ve got about ten percent left.” The PM says it in March and everyone nods. In June they say the same thing, in the same words. The team is tired, the client replies more and more slowly, and the sponsor has stopped asking about the date, which is a worse sign than if they asked every day.
A project that has been “almost done” for months has three real ways out. You can cut scope, you can extend, or you can close it. There is also a fourth way out, the one used most often.
The fourth option nobody chooses
The fourth option is: we carry on as before. Nobody announces it and nobody signs off on it. It simply happens. Week after week, because each of the other three requires a difficult conversation.
In my experience projects rarely die from a bad decision. Far more often they drag on for the lack of one. The first step of a rescue is to name the three options out loud and put them in front of someone who can choose.
Option one: cut scope
It makes sense when what is left to do is not what the project was started for. That happens more often than it seems. The last ten percent can be a wish list added along the way, a report nobody will open, or an integration with a system that is due to be replaced anyway.
There is one test: after the cut, does the project still deliver the thing someone once said “yes” to?
The trap is that a team is happy to cut what is hard rather than what is unnecessary. If the part that drops out is the very one the sponsor cared about, the project will finish on time and be of no use to anyone.
Option two: extend, but on a condition
It makes sense when the cause of the delay is known and has been removed. The supplier replaced the team. The person everyone was waiting for is back. The client has finally named someone who answers questions.
The test is: what has changed that makes the new date more credible than the previous one?
If the answer is “we’ll try harder”, you are looking at the fourth option in disguise. An extension without a change in conditions ends in another extension.
It also helps to give the new date a checkpoint halfway there. At that point you check one agreed thing, and if it isn’t there, you go back to the conversation about the three options.
Option three: close it
This is the hardest option to say out loud, because behind the project there is money spent, months of work and somebody’s name.
One question helps: knowing what we know today, would we start this project again? If not, every further month adds to a decision nobody would make today. A review that judges only the result will not show this, which is why good decisions and bad outcomes need to be told apart.
Closing does not have to mean throwing everything away. Often you can hand over what works, describe honestly what was not built, and release people to the work waiting in the queue. The team usually takes such a decision with a relief nobody dared to show earlier.
Who chooses
The sponsor chooses. The PM has no mandate to cut scope, move the date or end the project on their own.
What the PM does have is something the sponsor lacks: they know what each of the three roads looks like from the inside. Their job is to bring one page on which every option has three lines: what we get, what we lose, and what has to happen in the next two weeks. Underneath, a recommendation, one, with the reasoning in a single sentence. It is the same shape as a sponsor update built on status, consequence, options and recommendation.
A sponsor who gets a page like that usually decides quickly. A sponsor who gets another “ten percent left” has nothing to decide.
One question for this week
Do you have a project in your portfolio that has been “nearly ready” for three months? Which of the three options is real for it, and who should hear that?
All situations described in this article are based on real events, but contain no company names, no individual names, and no data from any specific project.
