John P Kotter Leading Change

What actually happens when you try to run a change initiative

Most people treat change like it is a project with a start date and an end date. It is not. It is a process that drags, stalls, and quietly eats budget for months. I learned this the hard way about eight years ago when our company was trying to migrate from an on-prem ERP system to a cloud platform. We had the budget. We had executive buy-in on paper. We failed because we skipped the early steps and jumped straight to implementation. The framework most people reference for this kind of work comes from John P Kotter Leading Change and it is useful exactly because it forces you to slow down. The eight-step model is well known, but most organizations get through it like a checklist. That misses the point.

Why the eight steps matter more than you think

Kotter's eight steps are: create urgency, form a powerful coalition, create a vision for change, communicate the vision, remove obstacles, create short-term wins, build on the change, and anchor the changes in corporate culture. The problem is not the model. The problem is that step one through four happen too fast in practice, and step six gets treated as a celebration rather than a strategic momentum builder. I have seen teams compress steps one through three into a single town hall meeting. That does not work. Urgency is not the same as fear, and it is not the same as a crisis narrative. Genuine urgency comes from data that makes the current state feel untenable to the people who have to live with it. My team spent about six weeks just on step one for that ERP migration. We pulled customer churn data, internal cost reports, and competitor benchmarking, then shared it directly with middle management before any leadership announcement. The resistance dropped noticeably after that.

The step most people botch

Step two is forming a guiding coalition. This is where change dies in most organizations. People think this means assembling a steering committee of VPs. It does not. It means identifying the actual influencers across the organization, regardless of title. Technical leads, senior individual contributors, people who informally control knowledge flow. If you skip the informal network, your change hits a wall at step four when you try to communicate it. In my experience, a guiding coalition of six to ten people works better than a large group. Anything above twelve becomes a coordination tax that slows decision making without adding influence. I usually vet potential members by asking two questions: who do people go to when they are confused about how work actually gets done, and who can deliver a coherent explanation of this change to their peers without sounding like they are reading a script.

How vision and communication actually function

A vision statement is not a poster. It is a decision filter. When someone comes to you with a request about the change, the vision tells you whether to say yes or no without needing a committee vote. I wrote a one-page document that mapped our cloud migration vision to specific operational decisions. Anything that did not align with reducing downtime, improving scalability, or cutting infrastructure overhead got pushed back. That saved us roughly forty hours a month in meeting time over the duration of the project. Communication needs to happen through multiple channels at a frequency most leaders consider excessive. One email per week minimum during active change phases, supplemented by team briefings, Q&A sessions, and visible progress tracking. The rule of seven from classic marketing applies here too. People need to hear the same message roughly seven times before they internalize it, and that is if the message is consistent.

Short-term wins are not optional

This is where most mid-sized initiatives stall. You need visible, credible wins within ninety to one-hundred-twenty days. Not big wins. Small ones that prove the direction is correct. In my ERP migration, the first win was decommissioning one legacy reporting module that took two weeks of effort and eliminated a monthly manual reconciliation process for three departments. We celebrated it publicly and tied it directly to the broader vision. Without that kind of concrete proof, skepticism grows faster than support can replace it. The biggest pitfall is treating the steps as linear. They are not. Step six feeds back into step one. A short-term win that lands poorly can kill urgency. Step five, removing obstacles, often uncovers structural issues that require you to loop back to step two and renegotiate the coalition. I learned this when our IT department blocked a key integration workstream because the migration touched systems they considered untouchable. We had to bring that IT director into the guiding coalition before step five could proceed. That added three weeks to the timeline but prevented a six-month blockage later. Another pitfall is underestimating how long step eight takes. Anchoring change in culture is not a one-time event. It is a maintenance cost. Research suggests that cultural anchoring requires sustained reinforcement for eighteen to twenty-four months after implementation. Organizations that stop investing at month twelve see regression rates between thirty and fifty percent depending on industry.

Where this model breaks down

Kotter's model assumes a degree of organizational stability that does not exist in every environment. In companies undergoing rapid layoffs, mergers, or restructuring, the concept of building a stable guiding coalition falls apart. People leave. Reporting lines shift. The model also places heavy emphasis on top-down vision communication, which does not work well in highly decentralized or matrixed organizations where no single leader has enough authority to drive alignment. For those situations, I lean toward an adapted approach that starts with step five instead. Identify the obstacles first, then build the coalition around solving them, then let the vision emerge from the solutions rather than preceding them. It is slower at the beginning but tends to produce more durable outcomes in volatile environments.

Practical timeline expectations

A typical enterprise change initiative using this framework takes eighteen to twenty-four months from initial urgency creation to cultural anchoring. If someone promises you six months, they are skipping steps. I would recommend adding a twenty percent buffer to whatever timeline you estimate, because the unknown resistance always surfaces at step four or step five. Resource allocation should reflect that. Underfunding the communication and training phases is the fastest way to see adoption rates drop below sixty percent within the first quarter after go-live. The model is not a shortcut. It is a map of the terrain. The work itself remains hard and mostly unglamorous. But it beats the alternative, which is spending millions on a change initiative and finding out six months later that nobody actually changed anything.