Why 70% of Organizational Transformations Fail Before They Even Begin
There is a statistic that has been circulating in boardrooms for decades without provoking the discomfort it deserves: between 60 and 75 percent of major organizational transformation processes fail or fall well short of their stated objectives. The data is not new. What is new — or should be — is starting to take it seriously as a symptom of something structural in the way leadership conceives of change.
Julia Dhar, Managing Director and Partner at Boston Consulting Group and founder of the firm's Behavioral Science Lab, published alongside Kristy Ellmer and Philip Jameson the book How Change Really Works: Seven Science-Based Principles for Transforming Your Organization. In a recent interview for the HBR IdeaCast podcast, Dhar articulates with precision the diagnosis that many executives intuit but rarely formulate with such clarity: the problem with transformation is not in the strategy, it is in the behavior. And more specifically, it lies in the distance between those who design the change and those who must live it.
This distinction is not a nuance. It is the difference between an initiative that leaves a mark and one that consumes millions in consulting fees, PowerPoint presentations, and alignment meetings, only to quietly dissolve into the inertia of the system.
Design and Adoption Are Not the Same Thing
There is a hidden logic behind the way executive teams manage transformations. A disproportionate amount of time and money is invested in strategic diagnosis, in designing the new operating model, in vision workshops, and in the architecture of change. Once that phase concludes, the implicit assumption is that implementation will naturally follow. That if the strategy is sound, the organization will adopt it.
That assumption is the mother of most failures.
Dhar introduces the concept of take-up — the actual probability that people will do what the program expects them to do — as a variable that is almost never rigorously audited. The question that few organizations ask themselves before launching a transformation process is this: Is it likely that the people we are asking to change their behavior will actually do so? And if the honest answer is "we don't know," then the entire business case rests on an unverified assumption.
This is not a problem of internal communications or change management in the cosmetic sense of the term. It is a problem of decisional architecture. When an executive team approves a transformation process, they are implicitly assuming that adoption capacity exists within the organization. If that capacity has not been assessed — if no one has asked what specific behaviors are required, from whom, with what incentives, and against what real barriers — then the transformation is overbuilt from the very beginning.
Dhar's research involving 6,000 people across twelve countries reveals something that contradicts the managerial cliché that "people don't want to change." Forty-five percent of employees in non-executive positions feel instinctively a positive or very positive predisposition toward change. Among executives, that number rises to 70 percent. The gap is real, but the critical point is this: the majority of people are not allergic to change. They are allergic to changes that are poorly explained, poorly designed, or that they perceive as disconnected from their own interests or capabilities.
When a transformation encounters resistance, the most common executive diagnosis is "people don't want to change." That diagnosis is convenient because it shifts responsibility downward. But it is almost never accurate. What lies beneath the resistance, in the majority of cases, is a mixture of genuine anxiety, a lack of clarity about what is expected, the absence of coherent incentives, or the perception that those leading the process are bearing no personal cost in it.
False Alignment as a Silent Risk
One of the most costly patterns I observe in medium and large organizations is what we might call performative alignment: that state in which members of the executive team verbally express their commitment to an initiative while internally maintaining reservations they never verbalize. Not because they are being dishonest. But because the cost of dissenting, in many organizational contexts, is perceived as higher than the cost of staying silent.
Dhar frames this directly: before scaling an initiative, the first step is to ask whether what exists is genuine agreement or false alignment. Her tactical proposal is deliberately low-cost: ask core team members to write down on paper exactly what it is they agreed to do and how it will work. If the answers do not converge, there is no transformation program. There is a collective fiction with assigned funding.
This exercise is uncomfortable precisely because it exposes what group dynamics had buried. In most executive meetings, the silence at the end of a presentation is interpreted as consensus. But silence can be many things: doubt, fatigue, political calculation, uncertainty about what the implications will be for one's own role. None of those things is alignment.
The cost of operating on false alignment is not immediately visible. It appears three or six months later, when the messages reaching the organization are inconsistent because each member of the executive team is interpreting the mandate in their own way. When resource allocation decisions contradict the declared priorities. When the executive sponsor of the initiative stops attending follow-up meetings because their attention has already migrated toward another urgency.
Dhar mentions this last point as an early warning signal that is systematically ignored: an executive who stops showing up at program meetings not because of an emergency but out of disinterest. It is perhaps the most honest signal about the true state of a transformation, and also the most politically uncomfortable one to name.
Momentum Is Not an Accident, It Is a Decision
The second variable that is chronically underestimated in transformation planning is momentum. Not as a motivational metaphor, but as an organizational resource that can be deliberately managed.
Dhar makes a distinction worth underlining: the problem with many transformations is not that they become chaotic in the intermediate stage. It is that they become boring. The executives who designed the change have already cognitively internalized it and their energy has shifted toward the next challenge. Employees, on the other hand, are still in the middle of an adoption process that requires sustained effort. And what leadership diagnoses as "change fatigue in the organization" is frequently, in more precise terms, executive abandonment of the process.
Momentum is not sustained by enthusiasm at launch alone. It is built through a deliberate cadence of signals: early wins communicated with more emphasis than seems necessary, follow-up rituals that do not depend on the availability of a particular sponsor, visible recognition of the behaviors that the transformation seeks to install. None of these things are sophisticated. But all of them require discipline, which is exactly what is in short supply when leadership has shifted its attention to the next priority.
The Delta Air Lines case that Dhar cites illustrates this principle well. The airline, emerging from its bankruptcy period with more than 100,000 employees in operation, institutionalized a practice that persists to this day: executives and organizational leaders regularly dedicate time to being physically present with frontline employees — cabin crew, mechanics, check-in agents, baggage handlers — to listen to them, recognize them, and be visible. The logic behind this is not sentimental. It is an explicit theory of change: if leadership takes care of employees, employees take care of customers, and satisfied customers produce sustainable financial results.
What is notable about this example is not the practice itself. It is that it was designed as a structure, not as an initiative. It does not depend on the goodwill of any particular CEO or on a season of high organizational culture. It was built to generate its own momentum, with aligned incentives — such as profit-sharing — that make the message coherent with the lived experience.
The Conversation Nobody Has in Time
Beneath every transformation that fails there is, invariably, a conversation that did not happen when it should have. Sometimes it is the conversation between the CEO and their team about whether there is genuine agreement on the objectives or only on the narrative. Sometimes it is the conversation someone should have had with the organization about the real cost that the requested change implies, rather than selling a softened version to avoid initial resistance. Sometimes it is the internal conversation that a leader avoids having with themselves about whether what they call "organizational resistance" is not, in part, a reflection of their own inconsistencies.
The behavioral science that underpins Dhar's work is not a set of tricks to make change more palatable. It is a diagnostic system that forces the formulation of uncomfortable questions before the program gains momentum: Who specifically must change their behavior? What behaviors, with what degree of precision? Do they have the right incentives to do so? Are there concrete barriers that no one has named because doing so would mean questioning decisions that have already been made?
When an executive team can answer those questions with specificity and honesty, success is not guaranteed, but the probability rises in a measurable way. When they cannot answer them, what they have is not a transformation in progress. They have a well-intentioned spending process with a built-in expiration date.
The statistic of 70 percent failure has been cited for decades in conferences and academic articles without the underlying patterns changing in any substantial way. The most plausible hypothesis for that persistence is not that leaders are incompetent or that strategies are flawed. It is that the system of incentives surrounding transformations — including the consultants who design them, the executives who sponsor them, and the boards that approve them — consistently favors elegant design over honest adoption. And as long as that distortion is not named with clarity, the number is not going to move.
