What the latest research reveals about why transformations stall, and how a new generation of AI tools is changing what leaders can actually see.

For thirty years, one statistic has refused to move. McKinsey has found repeatedly that roughly 70% of large-scale transformations fail to meet their goals, a figure echoed in John Kotter’s foundational work on change and largely unchanged across a decade of follow-up research (McKinsey). Organizations have gotten better at strategy, better at technology, better at project management. The success rate has barely budged. The interesting question is why.

The answer that emerges from the research is consistent, and it has little to do with the quality of the strategy itself. Transformations fail in execution, specifically in the human part of execution, where a plan has to become the changed daily behavior of thousands of people. As McKinsey’s transformation practice puts it, transformational change requires individuals to behave differently, and a persistent blind spot is the failure to engage frontline employees and their managers in the effort.

The execution gap is widening, not closing

If anything, the human side of change has gotten harder. Gartner’s workforce research found that employee willingness to support enterprise change fell from 74% in 2016 to just 38% in 2022, roughly half, in six years (Gartner). The cause is not stubbornness. The average employee now faces around ten planned enterprise changes a year, up from two in 2016, and the result is widespread change fatigue.

The trust picture is just as stark. An April 2025 Gartner survey of more than 2,800 employees found that 79% report low trust in organizational change, and that only 32% of leaders globally succeed in getting employees to adopt change in a healthy way (Gartner). More than a third of leaders say their teams now hesitate and wait to see whether a change will stick before adopting it. The workforce has learned to be skeptical, and skepticism is invisible on a project plan.

Why the conventional toolkit can’t see the problem

Here is the deeper issue. The tools most organizations use to manage change are not built to detect any of this in time to act. Engagement surveys arrive quarterly and compress lived experience into a number. Learning platforms track course completions, not whether behavior actually changed. Status reports reflect what managers believe is happening, filtered upward through layers that tend to smooth out bad news.

Each of these is a snapshot of the past. None of them shows a leader where resistance is building right now, which teams are clear and which are confused, or where frontline managers are too overloaded to carry the change forward. By the time a problem surfaces in a survey or a slipped milestone, the window to fix it cheaply has usually closed.

Gartner’s own conclusion points in a telling direction: the inspirational, top-down model of driving change (announce it, train on it, hope it lands) “falls flat in a low-change-trust environment.” Establishing change as a routine, supported and measured continuously, is roughly three times more effective (Gartner). That shift, from broadcasting change to instrumenting it, is what a new category of software is being built to enable.

What “execution-centric” actually means

Most enterprise software in this space has historically been learning-centric: its core question is whether people received the knowledge. An execution-centric platform asks a harder question: are people actually working differently, and if not, where exactly is the change breaking down? The first measures input. The second measures movement, which is the only thing that correlates with outcomes.

Practically, these platforms work less like a survey and more like a conversation at scale. Rather than watching activity logs, they engage each employee directly through short, structured, often private interactions focused on that person’s specific role, blockers, and readiness to change. The behavioral science behind this is well established: people adopt change more reliably when they can reflect, voice concerns in a psychologically safe setting, and connect a change to their own work, the same conditions Amy Edmondson’s research on psychological safety has linked to learning and adaptation for years.

The organizational intelligence is a byproduct of those interactions. In aggregate, thousands of individual conversations become something no quarterly survey can produce: a continuous, evolving map of where readiness, resistance, and misalignment actually live, visible early enough to act on. Strategy stops being something a leader hopes is landing and becomes something they can watch land, week over week.

Addressing the obvious questions

Two concerns surface immediately for any serious buyer, and both are reasonable. The first is trust: employees already skeptical of change are unlikely to speak candidly to a tool they think is surveillance. This is why credible platforms in this category emphasize privacy by design: individual conversations kept confidential, insights surfaced only in aggregate, and data handled under enterprise-grade security and regional compliance standards such as GDPR. Candor depends on it.

The second is proof. Leaders have been promised transformation results before. What distinguishes an execution-centric approach is that it is measurable on its own terms, through indices that track movement in adaptability, openness to change, and perceived support across a workforce over time, rather than anecdotes or a single end-of-program survey. The point is not a better feeling about change. It is a visible, defensible trend line tied to adoption.

Where this is heading

Gartner notes that organizations with better-than-average healthy change adoption report roughly twice the year-over-year revenue growth of those without it (Gartner). The prize for closing the execution gap is real and measurable. For decades it stayed out of reach mainly because the human side of execution was impossible to see at scale. That constraint is lifting.

Execution-centric AI platforms, Pandatron among them, represent one practical answer: systems that engage every employee individually while giving leaders a real-time view of how a strategy is actually being adopted. The technology will keep evolving. But the underlying shift is already clear. The organizations that can see their execution will outpace the ones still guessing at it, and in transformation, visibility is quickly becoming the whole game.

Sources: McKinsey & Company; Gartner; Harvard Business Review (A. Edmondson); J. Kotter, Leading Change. Figures as reported in the cited research.