The Real Reason Change Programmes Stall: What a Meta-Analysis of 5,651 Leaders Reveals
If you have ever rolled out a change leadership programme that worked brilliantly in one part of the business and barely made a dent in another, you have already discovered the finding at the centre of a major 2026 meta-analysis before the researchers published it.
Ferede and Getahun (2026) analysed 14 studies covering 5,651 participants on the relationship between strategic leadership and change management success. The headline finding is reassuring: strategic leadership has a statistically significant positive effect on change success. The more important finding is what came next.
The Number That Actually Matters
The heterogeneity statistic across the studies was I² = 97.84%. In plain terms, the relationship between leadership and change success varied enormously from one context to the next — from almost no effect in some organisations to a very strong effect in others. The paper’s conclusion is direct: a one-size-fits-all leadership model is inadequate for the diverse challenges of organisational change.
This is not an argument against leadership development. It is an argument against generic leadership development. The paper found that the transformational dimension of leadership — building trust, fostering psychological safety, mobilising commitment — was the primary mechanism through which change succeeded. Not resource allocation. Not the strategic plan itself. The relational, human-centred work.
“A leadership approach that worked in the last organisation, or the last role, is not guaranteed to work in this one. Context is not a footnote. It is the finding.”
Why Diagnosis Has to Come Before Design
This is precisely why I do not start any organisational engagement with a template. Two businesses with almost identical change challenges can need entirely different approaches, because their history, their trust levels, their existing psychological safety and their leaders’ starting points are different.
The paper also makes a point that is easy to miss: measuring change success purely through financial outcomes understates what is actually happening. Employees’ willingness to embrace change, and behavioural shifts in line with new goals, are part of the success picture — and they are exactly the outcomes that transformational, trust-building leadership produces.
Diagnosis is key for tailoring learning. Photo Getty Images Unsplash
What This Means in Practice
• Diagnose before you design — the same programme will not produce the same result in every part of your organisation
• Prioritise the relational capabilities — trust, psychological safety, genuine communication — over process and planning alone
• Measure behavioural and cultural change alongside financial outcomes, not instead of them
• Expect and plan for variation across teams and functions rather than treating inconsistent results as a delivery failure
For L&D professionals
When a change programme delivers strongly in one division and weakly in another, the instinct is often to blame delivery quality. This research suggests the more useful question is contextual: what was different about the leadership relationship, the trust level, and the psychological safety in each environment before the programme began.
Ready to have a conversation?
Start with the Leadership Resilience Diagnostic — a clear, structured snapshot of where you are across all seven dimensions of The Resilience Wheel.
Leadership Resilience Diagnostic:
theresiliencecoach.co.uk/resilient-leadership-diagnostic
Book an introductory call:
zcal.co/russellharvey/intro-call
russell@theresiliencecoach.co.uk | theresiliencecoach.co.uk