If Not 70%, Then What? Pointers on People-Centric Change

If not 70%, then what? Some pointers worth thinking about

Mark Blackwell

Industrial Transformation | Strategy, Commercial Excellence, Innovation, Integrated Business Planning | Founder Arkaro | Host of the Arkaro Insights podcast to help you adapt & thrive in complexity

This is the final article in a three-part series on rethinking organisational change. Article 1 set out how I came to question the famous “70% of change initiatives fail” statistic. Article 2 followed the trail back through the consultancy literature and showed that the headline numbers do not match the underlying data. This piece is about the more useful work waiting on the other side of the argument.

A note on what this piece is and is not. It is not a comprehensive map of how to lead change well. It is a small collection of pointers from people whose work I have found especially useful, offered as starting points for thinking rather than as answers. The fuller version of this argument will publish here shortly. The community responses to the earlier posts have shaped both, and I am grateful for them.

The whole series, in a sense, started with one sentence from Professor Julie Hodges, in a conversation we recorded for the Arkaro Insights podcast earlier this year.

I had asked her, gingerly, whether the 70% statistic was real. Her answer:

“That statistic, about 70% of change programs fail, is complete fable. There is no evidence whatsoever to justify that. It seems to be one of the sort of myths of change that seems to be perpetuated. We need to put that aside and focus on how we can do change better.”

The first two articles dealt with the first half of that sentence. This one is about the second.

So if we put the figure aside, where does the better thinking actually live? Four pointers stand out for me, drawn from four people whose work has shaped how I am now approaching client conversations on change.

Julie Hodges — change has to be done with people, not to people.

The most quotable sentence from our conversation, and the one I have caught myself repeating most often since, is hers: “Change has to be done with people rather than to people.”

This sits at the centre of what she calls people-centric change. It is not anti-process. Her language is deliberate: “It’s about getting a balance between the process and the people side of it. It’s not about throwing out the process and just focusing on people, but it’s getting a much better balance.”

The practical consequence is small but profound. Before deciding what the change is, leaders engage the people who will live with it. The objection, “we don’t have time for that, we have to deliver”, is the one Julie hears most often, and her answer is empirical: change that is co-created, where the people affected have shaped it, is more likely to be sustained. The upfront engagement saves time later. The change designed at people, and imposed on them, is the change that tends to leak away once the launch is over.

Sense-making is the other practical move from her work that has stuck with me. When people are facing a change, what they need is not more communication at them, it is space and support to work out what the change means for them and their work specifically. That is a manager’s job, not just the change team’s, and it is one of the highest-leverage things a manager can do during any significant change.

Dave Snowden’s Cynefin framework — linear models are no longer fit for purpose.

Julie’s critique of traditional change models — “these models have been around for a long, long time… they tend to come from the notion that change is done from the top down… and that’s no longer the case” — sits naturally alongside Dave Snowden’s Cynefin framework, which makes the same point at a deeper level.

In Cynefin terms, most large change sits in the complex domain — the territory where the relationship between cause and effect is only knowable in retrospect. In that domain, you do not get to plan the answer and then execute against it. You set a direction, probe, sense what is happening, and respond.

The implication is uncomfortable for anyone who has ever sold a change programme on the strength of a Gantt chart. Goals in complex work are hypotheses, not promises. The work teaches you what is actually achievable, where the leverage really sits, and what the organisation can and cannot absorb. By the end, the best version of the change may look quite different from the plan that justified the original budget. That is not failure. It is what learning looks like when the system you are working in is complex rather than complicated.

Keith Driver — most change is not failed, it is partially absorbed.

Driver’s 2025 meta-analysis of two decades of global transformation data found a remarkably stable distribution: roughly 30% of large change efforts succeed fully against original ambition, around 50% deliver partial value, and around 20% collapse outright.

His more interesting argument is about why the middle 50% is the middle 50%. It is not, in most cases, that the change failed in any meaningful sense. The organisation could not fully absorb what was being built. The new capability outpaced the governance designed to use it. The new process bumped against incentives still pulling people the old way. The new structure existed on the org chart, but the decision rights stayed where they were. The work was real. The absorption was not.

This connects directly to Julie’s people-centric point. The structural conditions for absorption — governance clarity, decision rights, slack capacity, interface design — are the conditions under which people can actually adopt and own the change. Before adding ambition, check whether the organisation can digest what it already has.

Deborah Rowland — the inner state of the leader matters more than most leadership programmes assume.

This is where I want to flag a finding that I plan to explore properly in a future Arkaro Insights podcast with Deborah Rowland, whose two decades of research at Sherpa Consulting are among the most rigorous I have come across on this question.

Rowland’s team coded leadership story transcripts from large complex change programmes and ran variance analysis on the outcomes. Their headline finding: a particular bundle of leadership qualities explained more than 52% of the variance in whether the change succeeded or failed. The single biggest contributor within that bundle was the leader’s inner state — their capacity to stay present under pressure, choose their responses rather than react, read the system honestly rather than read into it, and acknowledge difficulty rather than perform certainty.

This connects to Julie’s point about sense-making, but at a different level. Julie is talking about the meaning-making that happens in teams, and the manager’s role in facilitating it. Rowland is talking about the leader’s own internal capacity to perceive what is actually happening, before they act on it. The two reinforce each other: a leader who cannot read the system honestly will struggle to create the conditions for honest sense-making in others.

I want to be careful not to overclaim her conclusions before we have had the conversation properly. But the basic shape of the finding fits with something many of us have observed in practice — that change programmes track the inner condition of the people running them in ways the official methodology does not capture, and that the leadership work most easily abandoned under pressure is precisely the reflective work that makes the rest of it possible.

These four pointers do not add up to a complete framework. They are not meant to. They are four lenses on the same uncomfortable question — what does it actually take to do change well? — and each of them, in my experience, makes the conversation more useful than the headline statistic did.

What they share is a shift away from change as something done to an organisation by its leaders, toward change as something the organisation does with itself, in conditions designed for absorption, led by people whose inner state can stay steady under the pressure the work generates. None of which is novel. All of which is harder than it looks.

What is missing from this list will be obvious to anyone who has led significant change. Energy. Trust. The political economy of decision rights. The specific role of the manager versus the leader. The question of when to push and when to wait. Each of these deserves its own conversation, and I would be interested in the ones readers think matter most.

A question for this community

Julie’s line about change being “done with people, not to people” sounds simple, and it is. The interesting part is the textures of how it actually looks in practice — and how it fails to look that way when the pressure rises.

What does absorbed change look like in your own experience? Have you been part of a programme that genuinely stuck — where the new way of working became, after a while, just the way the organisation works — and if so, what made that possible? Equally, have you been part of one that delivered the project but somehow never became part of the organisation? What was missing?

The fuller article will publish here shortly. Your reflections will keep shaping the thinking, which is exactly what this series has tried to be.

About this series

This three-part series grew out of a conversation with Professor Julie Hodges on the Arkaro Insights podcast — the conversation that gave the series its starting point and most of its constructive material. The empirical foundations come from Mark Hughes (whose 2011 paper remains the definitive provenance critique), Keith Driver (whose 2025 meta-analysis provided much of the contemporary scaffolding), and Dr David Wilkinson (whose independent investigation surfaced one of the McKinsey cases discussed in article 2). Deborah Rowland’s research on leadership inner state will be the subject of a future podcast. The community responses to the earlier articles have refined the argument throughout.

Listen to the full conversation with Julie Hodges: People-Centric Change: The End of Linear Thinking (also available on LinkedIn and on YouTube)