Why Change Without Buy-In Always Fails

Back to Blog

I came across a Harvard Business Review podcast recently that stopped me in my tracks. Not because it was about the NHS, or local government, or staff engagement. But because it was about the UN Refugee Agency - and every word of it felt like it was describing the organisations I work with every day.

Kelly Clements is the deputy head of UNHCR - the United Nations High Commissioner for Refugees. Over the past decade she has overseen the most ambitious transformation programme in the agency's 75-year history - decentralising decision-making, overhauling culture, and rebuilding how a 20,000-person organisation operating across 550 locations actually functions. All while the number of people depending on them doubled and funding repeatedly contracted.

The parallels to public sector transformation in the UK are striking.

The consensus trap

When Kelly Clements walked into UNHCR, she found an organisation built on consensus. Decisions travelled upward, got discussed at length, and travelled back down - often changed beyond recognition, or arriving too late to matter. A hierarchical culture where your level dictated who had a voice. A Geneva-centric model where decisions were made far from the people they affected. A way of working that made sense for a smaller, simpler organisation but was buckling under the weight of a much larger, faster-moving reality.

Sound familiar?

The NHS, local councils, and government agencies - these organisations weren't designed to be agile. They were designed to be accountable, consistent, and safe. These are admirable values. But they can create structures that make the kind of change the modern world demands incredibly difficult to achieve.

Why top-down transformation struggles

When change is imposed from above, I've seen three things happen fairly reliably.

First, people comply without committing. They adopt the new language, attend the new meetings, and fill in the new forms - while continuing to work the way they always have. The transformation looks real on paper and feels hollow in practice.

Second, the people closest to the problem - the ones who actually know what is broken and what would fix it - get excluded from the solution. Their knowledge goes untapped. Their frustration quietly builds.

Third, and perhaps most damaging of all, trust erodes. When staff feel that change is something that happens to them rather than with them, disengagement tends to follow. And disengaged staff in a resource-constrained public sector organisation is not an abstract problem - it has direct consequences for the people those organisations exist to serve.

Clements identified this early. Her response was to fundamentally redesign where decisions were made - not from headquarters, but "as close as possible to the people they were going to affect, and with the people they were going to affect."

The buy-in problem is really a voice problem

Leaders often talk about buy-in as if it is something you can engineer after the fact. You make the decision, then you bring people along. You communicate the change, then you manage the resistance.

But in my experience, genuine buy-in is very hard to retrofit. It tends to be built upstream, in the moments before a decision is made, when the people it will affect have a real opportunity to shape it.

This isn't about endless consultation or decision-making by committee. Clements is clear that UNHCR still needed to move quickly and decisively. The goal was not consensus on everything - it was ensuring that the right voices were in the room, that people closest to the work had a genuine channel to surface what was broken and what was working, and that leadership actually listened and responded.

That distinction matters. Listening and acting on what you hear is fundamentally different from listening and filing it away. The first builds trust. The second, in my observation, tends to erode it.

What the best transformation leaders do differently

The leaders I see making real progress on transformation in the public sector tend to share a common approach. They don't try to change the whole organisation. They find the space within their division, their department, their team - where they have genuine autonomy - and they prove something.

They create structured, time-bound opportunities for their people to raise what is not working. They assign ownership so that issues don't disappear into a void. They close the loop - visibly, transparently - so that staff can see their voices had an effect.

And then they talk about it. Peer to peer. To the divisional leader two floors up who has been watching with cautious interest. To the director who is looking for evidence that agility can actually work in a large organisation before they commit to anything.

That bottom-up proof point strategy is slow. It requires patience and political capital. But it's the only version of public sector transformation I've seen that actually holds.

The moment we are in

Clements made one observation that I keep coming back to. In times of pressure - funding cuts, restructuring, uncertainty - the temptation is to centralise. To tighten control. To retreat to command and control because it feels safer.

But she argues, and the evidence from UNHCR bears out, that the opposite tends to be true. The organisations that come through periods of pressure strongest are the ones that trusted their people with more, not less. That opened up channels for honest conversation rather than closing them down. That made it possible for the people closest to the problem to be part of solving it.

The public sector in the UK is under more pressure right now than at any point in recent memory. That pressure will either accelerate the shift toward genuinely inclusive, agile ways of working - or it will push organisations back into the patterns that created the problems in the first place.

The leaders who choose the first path will, I believe, build something that lasts. The ones who choose the second may find themselves having the same conversations, about the same problems, in another five years.