Not every programme I have been called in to recover should have been recovered.
That is an uncomfortable thing to say after thirty years of doing the work. But the most valuable judgement I bring to a failing programme is not how to rescue it. It is whether it should be rescued at all – and the industry routinely collapses those two questions into one.
The question I am handed, almost always, is “can this be recovered?” It is only half a question. The right one is “should it be?”, and the gap between the two is where a great deal of money is lost.
Recovery capability answers the first. Given enough time, money and discipline, most programmes can be dragged across a finish line – genuine impossibility is rare. What recovery capability cannot tell you is whether the thing waiting at that finish line is still worth having. That is a different judgement, and it belongs to the owner, not the recovery team.
Continuation is not a decision. It is a default no one chose.
The bias toward carrying on is not stupidity. It is structural, and you cannot resist a structure you have not named.
Start with sunk cost. A board that has spent £60m of an £80m budget feels the £60m as a reason to continue. It is nothing of the kind. That money is gone whether you stop or carry on; the only money that matters is the money still to be spent, against the value still to be gained. Everyone knows this in the abstract. Almost no one feels it in the room.
Then add reputation. Someone sponsored this programme. Someone approved the business case, defended it at board, staked something on it. Stopping is not only a financial decision for those people – it is an admission. The human instinct is to defer the reckoning by continuing. A programme that limps on is a problem postponed. A programme that stops is a problem owned, today, by name.
And then the part the industry prefers not to say aloud: look at who is in the room. The contractor is paid to deliver. The delivery partner is paid to deliver. The advisors are paid to deliver. Continuation is the product that every party at the table sells. When you ask a room whose entire commercial interest is delivery whether delivery should continue, you should not be surprised by the answer. This is not corruption. It is incentive – and incentive does not have to be corrupt to be blinding.
Put the three together – sunk cost felt as investment, reputation defended by delay, and a table on which no one’s interest is stopping – and continuation stops being a decision at all. It becomes the path of least resistance, chosen by default because no one was positioned to choose otherwise.
The distinction that matters
Here is what I have learned to separate, because conflating them is the error.
“We cannot finish this” is rare. Genuine impossibility – the technology will not work, the ground will not take the structure – is unusual.
“Finishing this is no longer worth what it now costs” is common. The programme can be delivered. But the cost to complete has doubled, the schedule has slipped two years, the market it was meant to serve has moved, or the business case that justified it has quietly evaporated while everyone was busy managing the schedule. The programme is deliverable and no longer worth delivering.
Those are the ones that should stop and usually don’t – because the machinery of the programme is still working, the reports are still being written, the milestones are still being hit, and competent delivery against a dead rationale looks, from the inside, exactly like success.
If we were standing here today, knowing what we now know, with none of the remaining money yet committed – would we start this programme from here?
The test I use to cut through it is simple to state and hard to sit with. If the honest answer is no, then every pound from this point is being spent to avoid admitting that the answer is no. The sunk cost falls out of the question, and what remains is the only thing that ever mattered: value still to come, set against cost still to pay.
Stopping is not the failure state
The word that does the damage is “failure”. Stopping a programme is treated as failing; continuing is treated as the responsible course. This is backwards.
Continuing to pour capital into a programme whose business case has gone is the failure. The stop decision is what protects the capital that remains. An owner who halts a programme that should be halted has not failed – they have done the single hardest and most valuable thing an owner can do: read the position honestly and act on it against every structural pressure not to.
I have watched a board spend more defending a decision than the decision was ever worth. I have also watched – less often, because it is rarer and harder – a board look at a programme it had personally championed and say: this no longer makes sense, we stop here. That second decision takes more competence and more nerve than any recovery I have led. It is the mark of a mature owner, not a weak one.
Build the stop in before you need it
There is a practical consequence to all this, and it is the part most owners miss.
If stopping is structurally hard to do in the moment – and it is – then the time to design the stop is before the moment arrives. Most stage gates are theatre. They exist on the plan, the meeting is held, the pack is reviewed, and the programme proceeds, because by the time anyone is standing at the gate the sunk cost and the reputations are already too large for “stop” to be a real option on the table. A gate that has never once stopped anything is not a control. It is a ritual.
A gate with genuine kill authority is different. It is defined in advance, before anyone is emotionally or commercially committed: these are the conditions under which we will stop; this is who holds the authority to call it; this is the question we will actually ask. Naming the stop conditions early – when the programme is still abstract and no one’s name is yet attached to its survival – is how an owner gives their future self permission to make a decision that will, by then, be almost impossible to make from scratch.
You cannot remove the pressure to continue. But you can decide, in the cold, what you will do when the pressure is on.
This sits with the owner, and only the owner
None of this can be delegated to the delivery side – not because they lack the skill, but because it is not their decision to make and not their interest to make it. Judging whether a programme should continue means holding the business case, not the schedule; the value, not the delivery. It is, in the end, the intelligent client’s most consequential act. Not interrogating a report or challenging a forecast, but asking the one question no one else at the table is incentivised to ask.
The best recovery and the well-judged stop come from the same discipline: an honest reading of where the programme actually is, uncontaminated by what has already been spent or promised. Most of my work is recovery. But the advice I am most certain was worth the fee is, once or twice, the advice to stop.
If you sit on the owner’s side of a major programme, the question is worth asking long before you are forced to.
If we were deciding today, from here – would we still begin?