You can have the right strategy, the right evidence, and the authority to sign it off, and still watch a change quietly die. It dies because the people whose cooperation it needed were never genuinely on board, they nodded in the meeting and then carried on as before. Mobilising stakeholders and coalitions is the unglamorous craft of fixing that: working out who has to back the change, who can block it, and how you turn a room of polite spectators into a group that actually pushes.

The quick version

  • A stakeholder is anyone who can affect your change or is affected by it. Mobilising them means moving people from passive awareness to active support, not just informing them, but recruiting them.
  • A coalition is a small, deliberately assembled group with enough power, credibility and expertise to drive the change. Kotter calls it a guiding coalition, and building one is the second of his eight steps for leading change.
  • Start by mapping stakeholders on two axes, how much power they have and how much interest they take, then match your effort to where each one sits, rather than treating everyone the same.
  • The trap is mistaking sign-off for support. A leader's permission gets a project started; a coalition of committed people is what carries it through the months when the novelty wears off.

The idea in depth: nobody changes anything alone

The most-cited account of why change fails is also the bluntest. John Kotter spent ten years studying more than 100 companies attempting major transformation and reported the results in "Leading Change: Why Transformation Efforts Fail" (Harvard Business Review, 1995), later expanded into the book Leading Change (1996). His finding: efforts rarely founder on the merits of the idea. They founder in the early steps, failing to create urgency, and failing to assemble what he named a guiding coalition: a group with enough positional power, expertise, credibility and leadership to drive the effort and stop the resisters blocking it.

Two of Kotter's specifics are worth holding onto. First, he argues a change has enough urgency behind it only when roughly 75% of a company's management is honestly convinced the status quo is unacceptable, anything less stores up trouble later. Second, he warns that the coalition almost never includes the whole senior team at first, and that is fine; what matters is reaching a critical mass of committed, credible people who pull together rather than a complete org chart who merely comply. So the move is to stop treating your sponsor's signature as the milestone. Before you launch, ask a harder question: who are the five to fifteen people whose visible backing would make this feel inevitable, and have I actually recruited them, or just told them?

A leader's permission starts a project. A coalition is what finishes one.

An honest limitation. Kotter's model is a distillation of what he observed, not a controlled experiment, and the "75%" figure is a rule of thumb he offers, not a measured threshold, treat it as a vivid way of saying "more buy-in than you think you need," not a number to audit. The eight steps are also drawn largely from large Western corporations of the 1990s; in a flatter team, a regulated body, or a volunteer-run organisation, the same principles apply but the mechanics differ. Use it to ask the right questions about coalition and urgency, not as a checklist that guarantees the outcome.

Map before you mobilise: power and interest

You cannot give every stakeholder the same attention, and you shouldn't try. The discipline of working out who matters most begins with R. Edward Freeman's foundational stakeholder theory (Strategic Management: A Stakeholder Approach, 1984), which made the simple but consequential argument that an organisation answers to far more groups than its shareholders. The practical tool most people reach for came a little later: a grid that plots stakeholders by their power (how much they can help or hinder you) against their interest (how much they care about this particular change).

A note on provenance, because it is usually told wrong. The grid is widely credited to Aubrey Mendelow, whose 1981 conference paper, "Environmental Scanning, The Impact of the Stakeholder Concept" (ICIS), actually proposed a power/dynamism matrix. The familiar power/interest version that most courses teach as "Mendelow's matrix" was the adaptation popularised by Gerry Johnson and Kevan Scholes in Exploring Corporate Strategy. The lineage matters less than the discipline it imposes: four quadrants, four different jobs.

flowchart TD
  Q(["Map each stakeholder
on power × interest"]) --> A(["High power, high interest
MANAGE CLOSELY, your coalition"])
  Q --> B(["High power, low interest
KEEP SATISFIED, don't let them block"])
  Q --> C(["Low power, high interest
KEEP INFORMED, your volunteers"])
  Q --> D(["Low power, low interest
MONITOR, light touch"])
The power–interest grid (Johnson & Scholes' adaptation of Mendelow). Four quadrants, four engagement strategies. Leaders Loop

Draw the grid before you write the comms plan, not after. The high-power, high-interest quadrant is your candidate coalition, the people to manage closely and recruit first. The high-power, low-interest group is the quiet danger: they don't care yet, but they can kill the change if it starts to threaten them, so your job is to keep them satisfied enough not to mobilise against you. The low-power, high-interest people are gold often wasted, they care, and while they can't sign anything off, they make brilliant advocates, pilot users and early evidence-gatherers. The low-power, low-interest majority needs a light touch, not a roadshow.

An honest limitation. A static two-by-two flatters its own neatness. Power and interest both move, a low-interest executive becomes very interested the moment your change touches their budget, so the map is a snapshot, not a fixture. Redraw it at every major phase. And "power" is slippery: the person with no title who everyone privately consults has more of it than the grid will show. Use the map to force the conversation about who you might be neglecting, not to file people away.

Mobilising is emotional, not just logical

Mapping tells you who; it doesn't tell you how to move them. The most important correction to the "make the business case and they'll follow" instinct comes from Kotter's later work with Dan Cohen, The Heart of Change (2002), built on a study of real transformations across many organisations. Their finding, summarised as see–feel–change, is that people change behaviour less when they're given analysis that shifts their thinking, and more when they're shown something that shifts their feelings. Coalitions are mobilised by a felt sense that the change matters and is winnable, not by a spreadsheet alone.

This dovetails with how Ronald Heifetz frames leadership itself. In Leadership Without Easy Answers (1994), Heifetz defines leadership as mobilising people to do adaptive work, to face problems that no authority can simply solve for them, where the work involves changing people's own values and habits. That reframing is the heart of this Toolkit topic: for genuinely hard change that meets resistance, you are not pushing a decision through a passive organisation; you are mobilising the people who must do the changing. Which is why a coalition needs a job, not just a briefing, a visible problem to own, a pilot to run, a result to point at. People commit to changes they help build, and stay spectators to changes done to them.

An honest limitation. "Appeal to emotion" is easily abused into manipulation or theatre, and a sceptical audience smells it instantly. See–feel–change works when the thing you show people is true, a real customer's frustration, a real cost, a real early win, and backfires when it's a staged moment with no substance behind it. Emotion earns attention; only delivery keeps it.

A worked example

Take a mid-sized insurer, call it Harbour, rolling out a new claims system that forces every adjuster to change how they work. (Illustrative scenario; the people and figures are invented to teach the method, not drawn from a real company.) The transformation lead, Priya, has the COO's sign-off and a polished plan. By Kotter's lights, that is the most dangerous moment: permission without a coalition.

So before the launch email goes out, Priya maps the stakeholders. The Head of Claims Operations, high power, high interest, goes straight into the coalition; nothing moves without her. The Finance Director, high power, low interest, doesn't care about claims software but will care intensely if the rollout overruns, so Priya keeps him satisfied with a one-page risk summary and no surprises. A cluster of experienced adjusters, low power individually, very high interest, are the people the change is being done to; she recruits three of the most respected as pilot users rather than leaving them to grumble in the tearoom.

flowchart LR
  A(["Sign-off from COO
(permission, not support)"]) --> B(["Map stakeholders
power × interest"])
  B --> C(["Recruit coalition:
Head of Claims + 3 respected adjusters"])
  C --> D(["Run a small pilot
see–feel–change: a visible win"])
  D --> E(["Coalition tells the story
peers believe peers"])
  E --> F(["Adoption spreads on
evidence, not mandate"])
Mobilisation in motion: permission becomes a coalition, a coalition runs a pilot, and the pilot becomes the story that moves everyone else. Leaders Loop

The pilot does the emotional work that no all-staff deck could. When three adjusters everyone respects say the new system saved them an hour a day, and demonstrate it, that is see–feel–change in one sentence, told by peers rather than by management. Priya hasn't avoided resistance; she has built a coalition credible enough to absorb it. Reverse the order, launch first, recruit later, and she'd be defending a mandate against a workforce that had every reason to wait for it to fail.

Frequently asked questions

What's the difference between a stakeholder and a coalition?

Stakeholders are everyone with a stake, a large, mixed group you analyse and engage at different intensities. A coalition is a small, deliberately chosen subset you actively recruit to drive the change. Every coalition member is a stakeholder; very few stakeholders belong in the coalition. The art is choosing the handful with the power, credibility and energy to pull others along.

How big should a guiding coalition be?

Small enough to act, broad enough to be credible. Kotter describes successful coalitions as anything from a handful to fifty people depending on the size of the organisation, growing over time. The test isn't headcount; it's whether the group together holds enough positional power, expertise and trust that the rest of the organisation takes the change seriously. A coalition of three peers with no clout is a club, not a coalition.

What do I do about powerful people who oppose the change?

Don't ignore them and don't try to win every one. The grid helps: a high-power opponent who is low-interest needs to be kept satisfied so they don't engage against you; a high-power, high-interest opponent has to be confronted directly, understood, negotiated with, and where necessary visibly outweighed by the rest of the coalition. The mistake is hoping a determined, powerful blocker will simply lose interest. They rarely do.

Isn't this just managing politics?

Partly, and there's no shame in that. "Politics" is just the reality that organisations run on relationships and competing interests, not on org charts. Mobilising stakeholders is doing that work honestly and in the open, mapping interests, building genuine support, and being straight about trade-offs, rather than pretending the rational merits of your plan should be enough. The dishonest version is manipulation; the honest version is leadership.

How is mobilising stakeholders different from communicating change?

Communication informs; mobilisation recruits. You can communicate flawlessly and still have a passive audience. Mobilising means giving people a role, a pilot to run, a result to own, a story to tell their peers, so they have a stake in the outcome rather than just an awareness of it. Communication during change is a vital tool inside mobilisation, but it isn't the whole job.

Related in the Toolkit

Where to go next