Organization Deliver

Kotter's 8 Steps

Kotter's 8 steps are a sequence for leading large change, from creating urgency to anchoring the new ways in the culture.

The questionWhich of the conditions for a large change are already in place in this program, and which have we skipped?

Also called the 8-Step Process for Leading Change. The 2014 revision is called the 8 Accelerators.

Process · John P. Kotter, 1995, 1996

1
Establish urgency People see why the status quo will not do
2
Form a guiding coalition A group with enough authority to lead it
3
Create a vision A short picture of where the change goes
4
Communicate the vision Repeatedly, in words and in conduct
5
Empower others to act Remove the structures and people in the way
6
Create short-term wins Visible results, planned in advance
7
Consolidate and keep going Use credibility to change more
8
Institutionalize the new ways Tie them to promotion and succession

The eight steps in the order of the 1995 article, in shortened wording.

When to use it

A sponsor has approved a transformation and the plan is a list of workstreams with no view of how people will be brought along.
A program has stalled after launch and you need a quick audit of which conditions were never built.
The steering committee is about to declare success at go-live and you want a reason to keep the program running until the change is anchored.
The client wants a sequence leaders already recognize, so the discussion can move straight to what is missing.

When to avoid it

The change is small or local. Eight steps of coalition building and vision work are overhead for a process tweak in one team.
Change is continuous and emergent, with no single transformation to lead. Kotter's own later version, or an iterative method, fits better than a fixed sequence.
The barrier is individual adoption, with people who agree and still do not change. ADKAR or Bridges look at that level.
The organization has not yet decided what to change. Use a diagnostic such as 7S or Nadler-Tushman first.

Overview

Kotter’s eight steps lay out the order in which a large organizational change is built: urgency, a guiding coalition, a vision, communication, removing obstacles, short-term wins, consolidation, and anchoring in the culture. Consultants mostly use it to audit a change program and find the conditions that were skipped. It supports a choice about where the next effort should go.

The common failure it describes is stopping early. Plans tend to be strong on vision and communication, and weak on the last two steps, where the new behavior gets built into promotion and succession. The model rests on Kotter’s observation of cases, not on controlled studies. Kotter later revised it into accelerators that run in parallel, so treat the order as a guide, not a rule.

The eight steps

Establish a sense of urgency

Enough people, including a large share of management, have to believe the current way cannot continue. Kotter's point was that most programs underestimate how hard this is. Urgency built on the organization's own evidence holds better than a manufactured crisis.

Form a guiding coalition

A group with the position, expertise, and credibility to lead the change, working as a team outside the normal hierarchy. A coalition made of the people who already run the business, meeting as a steering committee, is the usual substitute and the usual weakness.

Create a vision

A picture of the future that can be explained in a few minutes and that tells people what to do and stop doing. A vision that needs a long deck to explain is usually a list of projects.

Communicate the vision

Use every existing channel, many more times than seems necessary, and make leaders' behavior match the message. Kotter singled out undercommunication as a common failure. People believe what managers do over what they say.

Empower others to act on the vision

Remove the obstacles that stop willing people from acting, such as structures, approval chains, incentives, and managers who block the change. This step is where a program meets the formal organization.

Plan for and create short-term wins

Plan visible improvements that can land early in the program, deliver them, and recognize the people involved. Wins give evidence to the doubters and buy time from the board. They have to be planned, not hoped for.

Consolidate improvements and produce still more change

Use the credibility from early wins to change more systems and structures, and bring in more people. Kotter warned against declaring victory too early, because the old ways return once the program office closes.

Institutionalize new approaches

Connect the new behavior to results so people see it working, and build it into how leaders are chosen, promoted, and succeeded. If the next generation of leaders does not hold the new approach, it fades with the current one.

What it is not

A plan that ends at launch. Steps seven and eight happen after the first results, and many program plans stop before them.
A rigid recipe that has to run one step at a time. Kotter himself revised it into concurrent accelerators, and in practice several steps run together.
A tested causal model. It rests on Kotter's case observation. It is useful as a checklist of conditions, but it does not predict which programs will succeed.

The steps

Audit the plan against the eight

Lay the program's current plan beside the eight steps and mark which have real activity, an owner, and a date. Most plans are strong on vision and communication and thin on wins and anchoring. What comes out is a gap list.
Test urgency outside the coalition

Interview people two or three levels down and ask what problem the change solves and whether it is theirs. If they repeat the sponsor's slide without examples, urgency has not spread. What comes out is a short read of where belief exists and where it does not.
Check the coalition's makeup and authority

List the members and ask whether they include the people the front line trusts, whether they can remove obstacles without escalating, and whether they meet as a team. Add or change members before the program goes further.
Plan the wins and the criteria for success

Choose a small number of visible wins with dates and measures agreed in advance. Write down, before launch, what has to be true before anyone declares the change complete.
Build the anchoring work into the plan

Add workstreams for promotion criteria, hiring profiles, onboarding, and the measures in management reviews, with owners who stay after the program office closes. Set a review date at least a year after go-live.

In practice

The sponsor of a seven-hospital network had a plan to standardize discharge planning, and she wanted it laid out in Kotter's steps for the board. The consultant used the steps as an audit of the existing plan. Urgency existed on one slide in the board pack, based on network averages that no individual hospital recognized as its own. The guiding coalition was the operations committee of vice presidents. It had no nurse, no hospitalist, and nobody who had run a discharge. Short-term wins were not planned at all. The plan ended on the go-live date.

The revised plan began with each hospital's own delayed-discharge data, presented by its chief nursing officer to her own staff. Two charge nurses and a hospitalist joined the coalition with time freed from their rosters. Two medical units were chosen to show results within three months, with the measure agreed in advance. A final workstream, owned by the chief nursing officer, changed the charge nurse job description and the criteria for promotion to unit manager. It ran for a year past go-live.

What you leave with

A gap list showing which of the eight steps the current plan covers and which it skips.
A revised guiding coalition, with names and the authority each brings.
Planned short-term wins with dates, measures, and owners.
Written criteria for when the change can be called complete.
Anchoring workstreams that run beyond go-live, with owners in the line organization.

Failure points

Urgency as a burning-platform speech

Leaders announce a crisis that staff do not recognize from their own work. People wait for it to pass, and the next real warning gets the same response.

The steering committee as coalition

The existing hierarchy renames itself the guiding coalition. It holds formal authority but not the trust or knowledge of the people doing the work, and it runs the change like any other project.

Communication as broadcast

The program counts town halls and emails sent. Nobody checks whether people can explain the change, and leaders keep rewarding the old behavior in the meantime.

Victory declared at go-live

The system is live, the program office is closed, and the budget moves on. The new behavior has not settled, and the old way returns within a year.

The steps as a reporting template

Each workstream reports status against the eight steps, which turns them into milestones to be ticked. Some steps need repeated effort for the whole program and cannot be marked done.

The failures above are usually Presence: the framework was sound, but the attention of the stakeholders was not held.

Origin

John Kotter, then a professor at Harvard Business School, published "Leading Change: Why Transformation Efforts Fail" in Harvard Business Review in March-April 1995. He based it on watching more than 100 companies attempt large changes, and he framed each of the eight steps around the mistake he had seen at that point. The 1996 book Leading Change expanded the article, and the wording of several steps differs slightly between the two. The evidence is Kotter's own observation of cases, not controlled comparison.

In a 2012 Harvard Business Review article and the 2014 book Accelerate, Kotter revised the model into eight accelerators. They run at the same time instead of in sequence, they are driven by a large volunteer network of employees as well as a small core group, and that network works alongside the normal hierarchy. His firm, Kotter Inc., presents the 8 steps and the accelerators as its methodology and sells training on them; this note describes the ideas in its own words. A failure rate of about 70 percent for change programs is often quoted next to the model. Mark Hughes (Journal of Change Management, 2011) traced the published sources of that figure, Kotter's 2008 book A Sense of Urgency among them, and found no valid empirical evidence for it.