Large programmes rarely fail dramatically. There is no single catastrophic moment, no obvious point of failure, no event that triggers the post-mortem. Instead, something quieter happens. They drift — slowly, gradually, almost invisibly — until the distance between where the programme is heading and where it was supposed to go becomes too large to ignore.
By that point, the correction is expensive. The drift has been embedded in decisions, dependencies, and team behaviours for months. Understanding why it happens — and how to recognise it early — is one of the most practically valuable things a programme leader can develop.
What Drift Looks Like in Practice
Drift is insidious precisely because it does not look like failure. Milestones are being hit. Deliverables are completed. Status reports are mostly green. The programme appears to be performing. And yet something is wrong — a feeling that the accumulated activity is not converging on the original objective in the way it should be.
In one programme, progress looked solid on paper: milestones hit, deliverables completed, reports green across the board. But when we dug beneath the surface metrics, a different picture emerged. Workstreams were optimising locally — making sensible decisions for their own area that were quietly diverging from the decisions being made in adjacent workstreams. Dependencies that had looked manageable were becoming strained. Outcomes, measured against the original objectives, were no longer fully aligned. We were not failing. But we were no longer moving in a single direction.
The absence of visible failure is not evidence of health. Drift conceals itself behind green traffic lights and hit milestones, which is precisely what makes it dangerous.
Why It Happens
The causes of drift are consistent across programmes and sectors. Each is individually reasonable; the problem is their combined effect over time.
Programmes don't drift because people make bad decisions. They drift because small, reasonable decisions accumulate in slightly different directions — and nobody pauses to measure the cumulative divergence until it is significant.
Why Drift Is Dangerous
The specific danger of drift — as distinct from the more visible programme risks that governance processes are designed to surface — is that it does not trigger alarms. It does not look like failure. It can continue for months while status reports remain green and leadership remains reassured. And then, when the divergence from the original objective becomes large enough to produce tangible consequences — reduced value, outcomes that do not match expectations, significant rework requirements — the cost of correction has been compounding quietly for the entire period.
The correction of an early-stage drift is a recalibration. The correction of a late-stage drift is a recovery programme.
Pulling It Back
Recovering from drift is possible, but it requires deliberate intervention — not additional process, but a genuine pause to reassess direction. The steps are straightforward; the discipline to take them, particularly when the programme appears to be performing, is less so.
The real skill in managing large programmes is not avoiding drift entirely — that is impossible in complex, long-running environments where context changes and people make independent decisions. The skill is recognising drift early, before it becomes structural, and having the discipline to intervene before the correction becomes a crisis.
The programmes that succeed are not those that never drift. They are those whose leaders notice the drift while it can still be corrected without dramatic intervention.