Remote Delivery at Scale — What Worked, What Didn’t, What We Forgot
We are running a four-month experiment in accelerated knowledge loss, and we will not see the full cost until teams change and programmes enter new phases.
The Four-Month Experiment
Four months ago, most programme delivery teams walked out of their offices and did not come back. What followed was the largest unplanned experiment in remote programme delivery that the profession has ever seen. Now, with enough distance to think clearly but close enough to remember the details, it is worth asking honestly: what actually worked, what did not, and what have we already started to forget?
The answer matters because it will shape how programmes are structured and governed for years to come. If we draw the wrong conclusions — if we mistake survival for success, or confuse what was tolerable under duress with what is sustainable at scale — we will build the next generation of delivery models on foundations that cannot hold.
What Worked: The Unexpected Gains
The most striking discovery of the past four months is how much ceremony turned out to be unnecessary. Programme boards that once required two days of preparation and a half-day of face-to-face theatre moved to forty-five-minute video calls — and the decisions were no worse. In many cases, they were better. The compression forced clarity. When a senior stakeholder has committed to a video slot rather than a full afternoon, the papers must say what they mean. The meandering update, the political preamble, the careful positioning that characterised so many in-person governance sessions — all of it fell away, not because people chose differently but because the medium would not support it.
Sprint ceremonies adapted faster than anyone predicted. Daily stand-ups, retrospectives, showcases — the core rituals of iterative delivery — translated to video with surprisingly little loss. Teams that were already disciplined about timeboxing and preparation found the transition almost frictionless. The tools were ready: shared boards, collaborative documents, screen sharing. The bottleneck was never the technology.
Cross-geography collaboration improved. This is the finding that programme leaders are most reluctant to acknowledge, but it is real. When everyone is remote, everyone is equally remote. The power asymmetry that favours the team in the room — the side conversations, the whiteboard that the dial-in participants cannot see, the body language that the remote team misses — all of it vanished. Distributed teams that had spent years feeling like second-class participants suddenly had parity. The irony is sharp: it took a crisis that forced everyone out of the office to create the conditions that distributed delivery always needed.
What Did Not Work: The Gaps We Are Learning to See
But the honest account must also name the failures, and they are more structural than most post-mortems admit.
The informal information network collapsed on day one and has not been rebuilt. In a co-located programme, an extraordinary amount of coordination happens without anyone planning it — the overheard conversation that alerts a workstream lead to a dependency, the corridor exchange that resolves a blocker before it reaches the RAID log, the body language in a meeting that tells an experienced programme manager that a delivery lead is struggling. None of this translates to video. We have replaced it with nothing, and the consequences are accumulating.
The informal information network — the corridor conversations, the overheard exchanges, the unplanned moments of coordination — collapsed on day one and has not been rebuilt. We have replaced it with nothing, and the consequences are accumulating.
The result is a phenomenon that programme managers across sectors are now recognising: issues surface later. Not because people are hiding them, but because the ambient awareness that once caught problems early no longer exists. By the time a risk formally enters a status report or a dependency is raised in a structured meeting, it has already hardened. The early-warning system was never a system at all — it was a property of physical proximity, and we did not notice it until it was gone.
Remote delivery also exposed a brutal truth about the quality of programme documentation. When teams are co-located, they can survive with ambiguous requirements, incomplete designs, and assumptions that live in people’s heads rather than on paper. The informal network compensates. Remove the network, and every gap in documentation becomes a delivery risk. Programmes that had always had weak written discipline found themselves re-learning requirements and re-negotiating scope — not because anything had changed, but because the tacit knowledge that papered over the gaps was no longer accessible.
And then there is the question of pace. The initial sprint was exhilarating — teams moved faster in March and April than they had in years, driven by urgency, adrenaline, and the novelty of the new model. By June, the fatigue was unmistakable. Video meetings are cognitively more demanding than in-person ones. The absence of commuting time was filled not with rest but with more meetings. The boundary between work and home dissolved, and with it the natural rhythms that prevent burnout. Programme managers are now managing teams whose sustained output is declining, and the tools for addressing that remotely are limited.
What We Are Forgetting: The Dangerous Revisionism
The most concerning pattern is not what went wrong but what is already being rewritten in organisational memory. Four months is long enough for revisionism to begin, and it is happening in two directions.
The first revision is the narrative of triumph: we proved we didn’t need offices. This is seductive but misleading. What we proved is that experienced, established teams — people who already knew each other, already understood the work, and already had the relationships that enable collaboration — could sustain delivery remotely for a period of months under crisis conditions. That is a genuinely important finding. But it is not the same as proving that remote delivery works at scale, over time, for all types of work and all stages of team maturity.
New team formation has been severely impaired. Onboarding new staff into a programme remotely is measurably harder — the learning curve is longer, the social integration is slower, and the cultural absorption that once happened naturally now requires deliberate and often clumsy intervention. Programmes that have been running on their pre-lockdown teams have performed well. Programmes that have needed to scale, restructure, or bring in new suppliers have struggled. This distinction is being lost in the aggregate narrative.
“What we proved is that experienced, established teams could sustain delivery remotely under crisis conditions. That is a genuinely important finding. But it is not the same as proving that remote delivery works at scale, over time, for all stages of team maturity.”
The second revision is subtler and more corrosive: the quiet downgrading of what was lost. Organisations are already normalising the absence of things they once considered essential — face-to-face workshops for complex design work, physical war rooms for critical phases, the in-person relationship building that underpins successful supplier management. These were not luxuries. They existed because they solved real problems, and the problems have not gone away simply because the solutions are no longer available.
The Structural Questions We Are Not Asking
Beneath the surface narrative — can we deliver remotely? — sit harder questions that the profession has barely begun to address.
The first is about governance. Remote governance is faster, but is it producing better-informed decisions? The compression that makes video boards efficient also strips out the peripheral information that experienced board members use to calibrate their judgements. A programme director reading the room — sensing the confidence behind a green status, testing the conviction of a delivery lead who claims a risk is mitigated — is doing something that a forty-five-minute video call may not support. We have made governance more efficient without knowing whether we have made it less effective.
The second is about knowledge. Every programme generates knowledge — about the domain, about the organisation, about what works and what does not. In a co-located setting, much of that knowledge transfers organically. In a remote setting, only the knowledge that is explicitly documented and deliberately shared survives. We are running a four-month experiment in accelerated knowledge loss, and we will not see the full cost until teams change and programmes enter new phases.
The third question is about equity. Remote delivery is not experienced equally. Team members with dedicated home offices, reliable broadband, and no caring responsibilities have had a materially different experience from those working at kitchen tables with intermittent connectivity and children in the next room. These differences map onto existing inequalities in ways that programme managers cannot address but also cannot ignore. The performance data from the past four months is contaminated by factors that have nothing to do with the delivery model and everything to do with the circumstances in which people are living.
What Comes Next
The temptation will be to draw binary conclusions: remote works or it does not, offices matter or they do not. The reality is more textured and more interesting.
What the past four months have demonstrated is that the pre-crisis model — the assumption that effective programme delivery requires co-location by default — was wrong. There is a large and valuable set of programme activities that can be delivered remotely without loss, and in some cases with gain. Structured meetings, status reporting, iterative delivery ceremonies, document review, testing coordination — all of these have survived the transition and some have been improved by it.
But we have also demonstrated that there is a set of activities — early-stage design, complex problem-solving, team formation, relationship building, informal coordination — where physical proximity creates value that we have not yet learned to replicate remotely. Pretending otherwise is not optimism; it is a failure of observation.
The programmes that will navigate the next phase most successfully will be the ones that resist both the romantic return to the old model and the uncritical embrace of the new. They will ask, activity by activity and phase by phase, what the delivery model should be — not as a matter of policy preference but as a matter of observed evidence.
We have the evidence now, if we are willing to read it honestly. The risk is that we are already forgetting what it actually says.