Why Most Organizational Change Fails — The 5-Phase Science-Backed Method That Makes It Stick

Roughly 70% of organizational change initiatives fail to achieve their stated goals — not because the strategy was wrong, but because of how the change was designed and executed. A 2026 BCG research study published in Harvard Business Review offers a detailed diagnosis: most organizations treat change as a single event rather than a five-phase process, and they skip the phases where success is actually built. Here is the framework that separates organizations that make change stick from those that keep restarting the same transformation.

Why Most Organizational Change Initiatives Fail

The 70% failure rate has been cited in organizational research since McKinsey first documented it in the 1990s. What is striking is how little that number has moved despite decades of change management methodologies, consulting frameworks, and leadership development programs. The reason, according to BCG’s research, is that organizations continue to misdiagnose the problem.

Most failed change efforts are not failures of strategy. The vision is often correct. The competitive logic is sound. The failure is in execution — and more specifically, it occurs at predictable points in the change journey that leaders routinely underinvest in or skip entirely. BCG’s research identified five distinct phases where failure clusters, and it built a framework around addressing each one with intention.

The most common failure patterns include: leadership teams that publicly endorse a change but privately resist it; project plans that assume rational adoption and ignore emotional resistance; launch events that generate enthusiasm but no reinforcement; and initiatives that never formally close, leaving employees uncertain whether the new behavior is permanent or temporary.

The Five Phases of Organizational Transformation

BCG’s 5-phase model — Decide, Plan, Start, Persist, End — differs from earlier frameworks like Kotter’s 8-step model or Prosci’s ADKAR model in one important way: it treats each phase as having its own distinct failure mode. Most change frameworks describe what good looks like; BCG’s model also describes what failure looks like at each stage, giving leaders a diagnostic tool rather than just a prescription.

Each phase requires different actions, different leadership behaviors, and different measures of progress. Treating the five phases as interchangeable steps in a linear project plan is itself one of the most common errors in organizational change management.

Phase 1 — Decide: Why Alignment Is Not the Same as Agreement

The first phase is the decision to change — and the most important insight BCG offers here is the distinction between alignment and genuine agreement. When a leadership team nods its way through a change initiative kick-off, what often appears to be alignment is actually a collection of silent reservations. Each leader has interpreted the change differently, prioritized it differently, and made different implicit promises to their teams about how disruptive it will be.

This surface-level alignment collapses the moment the change gets difficult — which it always does. Leaders who were never truly committed begin to hedge, slow-walk decisions, and send mixed signals to their organizations. By the time the misalignment is visible, months of momentum have been lost.

What genuine executive sponsorship looks like in practice:

• Leaders can articulate not just what is changing but why it matters personally to them
• Leaders are willing to publicly acknowledge what they themselves will need to change
• Leaders have explicitly negotiated trade-offs — what will be deprioritized to make room for this
• Leaders have agreed on what success looks like in concrete, measurable terms
• Leaders have committed to a communication cadence, not just a launch message

Phase 2 — Plan: The Planning Fallacy That Derails Most Change Efforts

Psychologists Daniel Kahneman and Amos Tversky coined the term planning fallacy to describe humans’ systematic tendency to underestimate how long tasks will take, even when they have accurate past experience to draw on. Organizational change is particularly vulnerable to this bias because the plan is created by the people most motivated to believe it will succeed quickly.

BCG’s research shows that change timelines are typically underestimated by 40–60% in complex organizations. The plan doesn’t account for the energy dip that follows the initial launch. It doesn’t account for the parallel workload employees carry while also being asked to change. It doesn’t account for the inevitable false starts that accompany any significant behavioral shift.

Effective change planning incorporates several realities that optimistic plans omit:

• A realistic assessment of the organization’s change capacity — how many other initiatives are already running
• Explicit buffers for the valley of despair that follows the initial launch
• Milestones that measure adoption, not just activity — not “training completed” but “behavior observed”
• A clear owner for the change (not the project) — someone accountable for the human side, not just the deliverables

Phase 3 — Start: Building Employee Support Through Participation

The most well-documented finding in organizational change research is also the most consistently ignored: employees who help design a change are far more likely to adopt it. McKinsey’s research shows that involving frontline workers in the design phase increases change success rates by 3.4 times. The reason is not that their input always improves the design — it is that participation creates psychological ownership. People defend what they helped build.

Yet the vast majority of change initiatives are designed by senior leaders or consultants and then communicated downward. The “influencing” of the workforce happens after the design is finalized, at which point it is no longer influence but sales. Employees who were not consulted during the design are correct to be skeptical.

Participation mechanisms that BCG’s research identifies as effective include:

• Design sprints: Cross-functional teams that co-design specific elements of the change, particularly those that affect their daily work
• Pilot groups: Early adopter cohorts that test the change, provide feedback, and become internal advocates
• Advisory forums: Regular touchpoints where frontline employees can flag friction and see that it is acted upon
• Co-creation workshops: Structured sessions where affected employees help design implementation at the team level

Resistance that surfaces during these participatory processes is not a problem — it is intelligence. Organizations that treat early resistance as an obstacle to manage miss the feedback that would have prevented larger failures later.

Phase 4 — Persist: How to Sustain Momentum When Energy Drops

Every significant organizational change follows a predictable energy curve: an initial spike of enthusiasm at the launch, followed by a valley of despair approximately three to six months in. This is when the novelty has worn off but the results are not yet visible. The old way still feels faster. The new way still requires conscious effort. Competing priorities have re-emerged. And leaders, convinced the change is “underway,” have moved their attention to the next initiative.

This is the phase where most change initiatives die quietly. There is no announcement. No one declares failure. The change simply stops being reinforced, and over the following months, the organization drifts back to its previous state.

Sustaining momentum through the valley requires specific actions:

• Visible wins: Identify and publicly celebrate early evidence that the change is working, even if the data is preliminary
• Regular leader check-ins: Not project status meetings, but visible demonstrations of personal commitment from senior leaders
• Adaptive management: Adjusting the how without abandoning the what — showing responsiveness without signaling retreat
• Reinforcement mechanisms: Changing incentives, performance metrics, and meeting agendas to embed the new behaviors into the operating rhythm

Phase 5 — End: How to Close a Change Initiative the Right Way

The final phase is the one most organizations skip entirely — and it is the reason so many changes eventually reverse. When there is no formal closure to a change initiative, the organization never fully processes that the new way is the way. Employees remain in a state of ambiguity, treating the new behavior as a temporary project requirement rather than a permanent shift.

BCG’s research documents a consistent pattern: without explicit closing rituals, teams revert to old behaviors within 12 to 18 months. The change existed in project space but never got embedded in organizational identity.

Closing a change initiative properly involves:

• Formally announcing that the transformation phase is complete and the new behaviors are now the standard
• Acknowledging the effort and difficulty of the change process publicly and specifically
• Updating systems, processes, job descriptions, and performance criteria to reflect the new standard
• Handing the change from the project team to the line organization — making it no one’s special project and everyone’s daily practice

What Leaders Can Do Differently Starting Today

The 5-phase model does not require a transformation program to start applying. Any leader managing a change initiative — large or small — can use these principles immediately:

1. Audit your alignment: Can every leader on your team articulate the same change rationale in their own words? If not, the alignment is surface-level.
2. Add 40% to your timeline: Whatever your current change plan says, it underestimates the adoption curve. Build in explicit buffer for the valley of despair.
3. Identify your pilot group: Who are the early adopters who will help design and test the change? Involve them before the plan is final, not after.
4. Schedule persistence check-ins: Put 90-day, 6-month, and 12-month reinforcement milestones in your calendar now. They will not happen unless they are planned.
5. Plan the ending: Before you launch, decide how you will know the change is complete and how you will formally close the initiative.

Organizational change will always be difficult. But the difficulty is not random — it concentrates in predictable places. The leaders and organizations that understand where the failure points are, and design explicitly for each phase, achieve fundamentally different outcomes from those that treat change as a launch event followed by communication. The framework exists. The question is whether you will use it before the next initiative stalls at month four.

Sources

Designing Organizational Change That Actually Sticks — Harvard Business Review