Some programmes fail in a way that leaves no one obviously accountable, and those are the ones worth understanding. The strategy was sound; you could defend every line of it now. The delivery was competent, the team capable, hard-working and largely on target. And still the thing failed, or it landed so far from what was intended that calling it a success would be generous. When that happens the post-mortem struggles, because it goes looking for a mistake and cannot find one: each part did its job, and no one fell short of what they were asked to do. What no one was answerable for was whether the parts came together, and that is the gap the failure lived in.
That gap is no accident; it is built into how organisations do consequential work. Strategy and delivery are separated by design, set in one place and executed in another, by different people and often at different times. A strategy is formed, by a top team, a strategy function, or a firm brought in to write it, and then handed over to be delivered. The handover looks like a moment of completion, the plan is done, now we build it, yet it is in fact the most dangerous point in the whole arc, because it is where the people who hold the intent stop and the people who must realise it start, and the two groups are rarely the same.
What gets handed over is never the whole of what was meant. A strategy is set at a level of abstraction that no one can act on directly, so to deliver it the team has to make a hundred choices the strategy left open, most of them small enough that no one escalates them. The strategy says win the mid-market; delivery, facing a quarter-end, ships the feature the existing enterprise accounts were asking for, because that was the path of least resistance that month. Nobody decided to abandon the mid-market. It simply never got built. Each choice is made locally, reasonably, by someone doing their best with the part in front of them, and each shifts the thing a degree or two from what was intended. None is wrong on its own, but together, over months, they add up to a programme recognisably descended from the strategy and no longer the same animal. There is no moment of betrayal to point at, only drift, which is hard to govern because it never announces itself.
The reason the drift goes uncorrected is that no one’s job is to correct it. The people who set the strategy own the strategy, and their accountability ends when the plan is signed off and they move to the next thing, or leave the building entirely if they were brought in to write it. The people who deliver own the delivery, and their accountability is to build what the plan specifies, on time and on budget, which is exactly what they are measured on. Between the two sits the one thing that decides the outcome, whether the original intent gets realised, and it belongs to neither. The org chart has a name against the strategy and a name against the delivery and no name against the seam where they meet, so when the programme drifts each side can look at the other in good conscience, because each did the job it was given. The job nobody was given was holding the whole arc together.
You can see the gap most clearly in how the two ends describe a failure afterwards. Ask the board and you hear that the strategy lacked clarity, that it was never sharp enough to execute against; ask the executives running delivery and you hear that the strategy was fine and the problem was execution, the resistance, the competing priorities, the speed of decisions. The current surveys capture exactly this split, with directors naming strategic clarity as the barrier while chief executives name execution. Both are describing the same failure from opposite banks of the same river: the board sees a strategy that did not survive contact, the executives a delivery that wandered from its intent, and neither is looking at the handover in between, because the handover is no one’s vantage point. The disagreement is the gap itself, seen from its two edges.
So give the seam an owner. The instinct then is to reach for one of two temperaments, and both of them fail. The first is the consensus-builder, who will not move without the room: every decision goes to a meeting, every meeting defers to more consultation, and the thing slows until the calendar makes the choice that no one would. The danger here is subtler than a bad decision, because no call ever feels as though it has been taken, and so the programme settles onto the do-nothing path without anyone choosing it or being answerable for it. The second is the autocrat, who decides fast and alone, on the strength of a position rather than the facts, without asking the people who can see what is happening on the ground. That clears the paralysis and installs a different failure in its place: confident decisions made on poor information, briskly and beyond recall. Most capable people collapse into one of these under pressure, because each is the easier thing to be.
The two look like the only options, decide slowly with everyone or quickly alone, and the way out of the dilemma is an old one. The British military calls it mission command. A commander gives those below the intent, what is to be achieved and why it matters, together with the boundaries they must stay within, and then leaves them to work out how. The person on the ground has the authority to act, the speed to do so at once, and the freedom to draw in whatever perspective they need first, without passing the decision up a chain that would only slow it or hand it to someone too distant to judge it well. They decide, and they move. That dissolves the dilemma, because the two things absent in paralysis and in autocracy are exactly what mission command supplies: an intent clear enough that the person can act without seeking permission, and enough trust that they are allowed to.
That is also, precisely, what the seam needs. Intent gets lost between strategy and delivery because it has to survive a handover, passed from the people who hold it to the people who must act on it. Mission command removes the handover. The person who owns the seam carries the intent down into the hundred small choices where it would otherwise drift, making each of them in service of what was meant. The same judgement now sits on both sides of the seam, and the intent travels with the decision rather than being handed across in hope.
This is the point at which a borrowed idea has to be handled with care, because mission command is easy to admire and hard to run. The military can delegate this way because it has spent heavily on the two things that make delegation safe: officers are drilled in the craft of expressing intent, which is far harder than it sounds, and trust is built on purpose, through shared training and doctrine, so that a commander can let go and the person below can act. Most organisations have done neither, and want the speed of delegated decision-making without the clear intent that makes it safe or the trust that makes it real, so they get the two failures instead. Tell people they are empowered without giving them an intent sharp enough to act against and they will not dare to, which is paralysis dressed in the language of autonomy. And the first time a delegated decision looks risky, the senior figure reaches back in and overrules it, something the military understands well enough to have named, the long screwdriver, and the watching organisation learns that the authority was never real. Modern reporting deepens this rather than easing it, because a dashboard hands a sponsor just enough visibility to believe they grasp the ground better than the person standing on it.
None of this is the work of the programme manager, which is the role most organisations reach for and the reason the gap persists. A programme manager owns the plan and its machinery, the schedule, the budget, the status report, and watches whether it is being followed; but following it faithfully is often how the drift happens, because the plan was written before the hundred questions arose and cannot answer them. Governing the seam is a separate job from running the plan: it means owning whether the intent is realised, a question quite apart from whether the schedule holds, and sometimes the opposite one. The sponsor does not do it either, since they set the programme up, fund it, and then, rightly, step back.
The obvious rejoinder is that this is what a transformation office is for, and the fate of most transformation offices is the answer. The instinct behind them is right: someone notices the gap and stands up a function to own it. But the function is handed the seam’s accountability with only the PMO’s authority, so it answers for whether the transformation lands while the power to make the business change sits with line leaders who do not report to it. It does the part it can, collecting status, aggregating reports, convening the steering meeting, and passes a consolidated picture up to the sponsor, which is monitoring rather than accountability. Watching the drift and reporting it is a long way from having the standing to arrest it, and a function measured on a result it cannot command will fall back on the things it can do, which are exactly the things that do not move it. So when the transformation underdelivers, the office is the visible overhead with nothing demonstrable to show, and it becomes the easy thing to disband. The idea was right; what undid it was a reporting line where it needed a mandate.
So when a programme fails and no one is obviously accountable, this is usually what happened. The intent was sound and the delivery diligent, and between them lay a gap that belonged to no one, crossed by a handover that quietly turned the first into something the second could not realise.
The places to look fall out of everything above. Start with who is answerable for whether the intent is realised, rather than whether the plan is on track, and whether that is one person or a committee that meets once a month. Look at whether they hold the authority to settle the decisions that cut across the silos, or whether each one has to climb the chain and wait. Notice whether they were there when the intent was set and remain there as it is built, or whether they took delivery of a plan at the handover and never saw what it was for. And watch the small choices being made this week, the hundred that never reach a steering meeting, for any sign that someone is weighing them against what was meant. Where you find a committee, an escalation, a handover, or no one at all, you have found the gap.
I am not going to pretend there is a single answer for what to do about it, because how you close the gap depends entirely on how your organisation works, where its real authority sits, and who is trusted to make the call. What does not change is that it comes down to a person: someone who can stand in the gap and lead, carry the intent the whole way, and answer for whether it arrives. A title and a box on the chart are not the same as that, and the difference is the whole of it. Get that person right and the gap never opens; get it wrong, or leave it to no one, and the soundest strategy and the most diligent delivery will drift apart all the same.