A power-interest grid puts each stakeholder in a quadrant and implies an engagement strategy. Most teams start writing the plan. The step worth taking first is testing whether those placements are facts or assumptions. Skip it, and the engagement plan looks confident in the project charter but collapses the moment a stakeholder moves.
I watched a project team at an infrastructure company map their CFO into the "keep satisfied" quadrant: high power, low interest. The engagement plan specified quarterly briefings and a standing invitation to steering committee meetings that everyone knew the CFO would decline. The team treated the placement as settled and moved to communication schedules.
Six months later, the project crossed a cost threshold that pulled it into the CFO's capital approval process. Interest went from low to high in one meeting. The team had no working relationship, no established reporting cadence, and no sponsor willing to bridge the gap. The project stalled for eleven weeks while they rebuilt access from scratch.
The grid placement was a snapshot taken in a workshop. The engagement plan treated it as permanent. Nobody had asked what would change the CFO's interest level, because the power-interest grid does not require that question.
- Rewrite each stakeholder placement as a testable claim. "CFO: high power, low interest" becomes "We believe the CFO has the authority to block this project but will not take active interest in it."
- Score each claim on influence and confidence. High influence + low confidence = the placement that needs investigation before the engagement plan builds on it.
- Investigate the high-risk placements. One conversation or one calendar check per claim.
- Build the engagement plan only on verified placements. The Walk provides a structured way to test which placements rest on evidence before the engagement plan operationalises them.
- Assign ownership and a review trigger. Each stakeholder relationship gets a named owner and the condition under which the placement should be reassessed.
After a stakeholder analysis, the standard next step is an engagement plan based on the power-interest grid. The step worth taking first is to test whether each placement rests on evidence or assumption.
The standard next step: build an engagement plan
Freeman's original stakeholder framework defined stakeholders as "any group or individual who can affect or is affected by the achievement of the organisation's objectives." The 1984 formulation established the principle that organisations must manage relationships with these groups, not merely identify them.
The power-interest grid, sometimes called the Mendelow matrix, sorts stakeholders into four quadrants based on two dimensions: their power to influence the project and their level of interest in it. Each quadrant prescribes a management approach. "Manage closely" for high power, high interest. "Keep satisfied" for high power, low interest. "Keep informed" for low power, high interest. "Monitor" for low power, low interest. A worked power-interest matrix example shows how teams assign stakeholders across the four quadrants.
The standard next step after completing this analysis is to build an engagement plan that assigns communication frequency, message type, and escalation paths based on where each stakeholder sits. Newcombe applied the approach to construction projects and showed how stakeholder mapping could extend the concept of "client" into a richer stakeholder landscape, one where contractors, regulators, and community groups each required distinct engagement strategies based on their grid position.
What the engagement plan adds
Stakeholder analysis on its own is an identification exercise: it asks who the stakeholders are and where they sit on the grid. The engagement plan is the formulation step: it asks what to do about each one. If you have completed a stakeholder analysis and want to move beyond the grid, an engagement plan is what most project management frameworks recommend.
The contribution is real. It connects analysis to action, assigns responsibility, and creates a communication rhythm. Without it, the power-interest grid stays on a slide and never reaches the project team's operating cadence. A steering committee that is not making decisions often has exactly this problem: the right people are in the room but the engagement was never matched to their actual level of interest.
But the plan inherits whatever the grid gives it. If the grid says a stakeholder has low interest, the plan assigns light-touch communication. If the grid says a community group has low power, the plan assigns monitoring. The engagement plan does not test those inputs. It operationalises them.
Rewrite the stakeholder placement your engagement plan rests on as a claim and test it before the grid locks it into a quadrant. Start the Walk →
Where the standard playbook breaks down
The engagement plan introduces a problem of its own. It treats every grid placement as settled. "High power, low interest" becomes a quarterly briefing. "Low power, low interest" becomes a newsletter. If the placement is an assumption rather than a verified position, the engagement strategy inherits the assumption without marking it.
Mitchell, Agle, and Wood's 1997 paper on stakeholder identification and salience introduced a model based on three attributes: power, legitimacy, and urgency. Their central finding was that stakeholder salience is dynamic. A stakeholder who possesses only power (a "dormant" stakeholder) can acquire urgency and legitimacy as conditions change, shifting from the periphery to the centre of management attention. The power-interest grid captures one snapshot. It has no mechanism for signalling when that snapshot expires.
Bourne and Walker's work on visualising and mapping stakeholder influence attempted to address this by introducing the Stakeholder Circle, a tool that tracks proximity, influence, and the direction of change. Their contribution acknowledged what the standard grid does not: that stakeholder positions are not fixed, and any mapping tool that treats them as fixed becomes a liability the moment conditions shift.
The Casement Park stadium redevelopment in Belfast demonstrated the cost of treating a stakeholder map as permanent. The project team positioned local residents as a low-power group to be monitored but not actively engaged. When the proposed 38,000-seat stadium drew objections over traffic, light obstruction, and safety, residents organised, challenged the planning decision in court, and won. The High Court quashed the planning permission, ruling the decision-making process "fundamentally flawed." The project was delayed by years and the redesign cost millions. The residents had not changed. The analysis had been wrong.
When I work with teams on stakeholder buy-in, the first thing I check is which placements were tested and which were guessed. The answer is usually the same: none were tested. The question is not whether to hold more stakeholder meetings. It is whether the placements driving those meetings were ever verified.
I have seen this pattern across nearly fifty years of working with organisations. The grid produces a structure that looks analytical, and the structure itself becomes the justification. Teams stop asking whether the placements are right because the engagement plan has already operationalised them. When disagreement surfaces, it is usually a sign that the original mapping missed something, not that the team is being difficult.
The step to take first
The engagement plan is not the problem. The problem is building the plan before testing the placements it rests on. Before a grid placement drives a communication strategy, it needs to survive a simple question: what are we assuming here?
Take each placement and rewrite it as a claim. "CFO: high power, low interest" becomes "We believe the CFO has the authority to block this project and will not take active interest in it until it reaches the capital approval threshold." Then ask two things about that claim: how much influence does this stakeholder have on the project's success, and how confident are we that the placement is accurate? A placement with high influence and low confidence is the one that needs investigation before the engagement plan builds on it.
A placement you are confident about and that matters to the project can enter the engagement plan. A placement you are guessing at and that carries the project's critical relationship should not. The same test applies to every quadrant. An entry earns the right to drive a communication strategy only after someone tests whether the placement is true.
The infrastructure team from the opening would have placed the CFO in "keep satisfied." Rewritten as a claim: "We believe the CFO will not take active interest in this project." Influence on the project: high, because the CFO controls capital approval. Confidence: low, because nobody had checked whether the project's cost trajectory would trigger the CFO's review threshold. That placement needed verification before the engagement plan assigned quarterly briefings, not after.
When I work with teams after a stakeholder analysis, the first thing I do is separate placements they have evidence for from placements they are assuming. The separation takes ten minutes, and it changes which engagement strategies survive the planning step. Most teams have never been asked to make that distinction, because the grid does not require it.
This is the step between completing a stakeholder analysis and committing to the engagement strategies it generates. Separate evidence from guesses. Test the placements that carry the project. The Universal Decision-Making Method calls this "recognise assumptions" and places it before any commitment, because an engagement plan built on an untested placement is not a plan.
You could plot every stakeholder and still build an engagement plan on placements nobody tested.
Work through your decisionNo sign-up. Just pick your decision and start.
Grant Purdy is the co-author, with Roger Estall, of Deciding (2020), and the architect of the Universal Decision-Making Method.