Power/Interest Grid vs Salience Model
A power/interest grid takes about twenty minutes and is almost always worth doing. Two axes, four boxes, a room full of people who can now agree on who needs managing closely and who needs an occasional update. Its weakness is not that it is crude. Its weakness is that it is crude in one particular direction, and the direction is predictable enough to plan around.
The salience model sorts on a different basis, and the practical question is not which model is better but which question you are trying to answer. A grid answers how much engagement effort to spend on whom. Salience answers whose claim on the project deserves attention and why, which is a slower question and occasionally a much more important one.
What each one is actually sorting
A power/interest grid places people by how much influence they hold over the project and how much they care about its outcome. The output is a plan for effort: close management for the powerful and interested, keeping the powerful but uninterested satisfied, keeping the interested but powerless informed, monitoring everybody else. It is quick, it produces a shared picture, and it survives being drawn on a whiteboard in a meeting, which is a genuine virtue.
The salience model sorts on three attributes instead of two: power, meaning the ability to impose one's will; legitimacy, meaning a claim that is recognised as proper; and urgency, meaning a claim that is time-sensitive and being pressed. Stakeholders holding one attribute are treated differently from those holding two, and those holding all three are where the project's attention belongs. The extra attribute costs more to assess and returns a more nuanced ordering.
Section 2.5 of the PMBOK® Guide is where stakeholder practice sits, and it is emphatic that a person's position on any chart is a starting hypothesis with a short shelf life. Both models produce a snapshot. Neither produces a classification anybody should still be working from six months later.
The blind spot in a power/interest grid
Look at the box marked keep informed. In most analyses it contains the people who will live with the result and have no authority over it: end users, frontline staff, the operations team inheriting the service, sometimes a community or a customer group. The grid is telling you, correctly, that they cannot stop you. What it cannot tell you is whether they should be able to, or whether their eventual refusal to adopt will stop you anyway by a slower route.
This is the recurring pattern behind projects that deliver on time and fail on adoption. Nothing was mismanaged. The analysis said these people had low power, the engagement plan gave them a newsletter, and the newsletter was honest. What the analysis never asked was whether their claim on the design was legitimate, which it almost always is, because they know things about the work that nobody in the steering group knows.
A general insurer replacing its claims handling system ran a careful power/interest analysis. The claims directors and the finance director sat top right. The handlers themselves, four hundred of them, sat in keep informed, and were consulted through a user group of three that met monthly. The build went well against its requirements. At go-live, average handling time rose by nineteen per cent and stayed there for four months, because the new screens followed the process as documented and the floor had been working to a set of learned sequences that were faster and that nobody had written down. The handlers had no power in any sense the grid recognised. Their claim on the design was entirely legitimate, and the cost of discovering that late ran to the better part of a year's efficiency saving.
When the third attribute earns its place
Legitimacy is the attribute worth adding first, and often the only one you need. Asking who has a proper claim on this decision, separately from who can enforce one, catches the group above before the design is fixed. It also disciplines the opposite error, which is treating a loud stakeholder with no real standing as though their objection carries weight.
Urgency is the attribute that stops you confusing volume with importance. A claim being pressed hard right now and a claim that will matter for the next five years are different things, and a project under pressure will naturally serve the first. Naming urgency explicitly lets you say that something is urgent, legitimate and worth an immediate response, or that it is urgent, marginal and can be acknowledged without being acted on.
Power remains the attribute that determines what happens when the other two are in conflict, which is why it cannot be dropped. Somebody with legitimacy and urgency and no power needs an advocate who has it, and finding them that advocate is often the most useful thing a project manager does in a week.
The cost side is real. Assessing three attributes across forty stakeholders takes an afternoon and a degree of judgement that invites argument, and on a small uncontested project it buys very little. Use the grid where the stakeholder set is small and the project is not contentious. Reach for salience where adoption will decide success, where a regulator or a representative body is involved, or where somebody keeps being described as difficult without anybody explaining what their claim actually is. Recognising which of those situations you are in is the judgement underneath the technique, and a structured PMP exam preparation course concentrates on that judgement more than on the diagrams.
For a PMP® candidate, the distinction to hold is that power and engagement are separate things, and that either model is a hypothesis to be revisited as positions change. Of any stakeholder described in a situation, the question to ask is what their claim is and whether anybody has tested it, because the response that fits depends on the answer.
Half an hour is enough to improve a grid you already have. Go through the keep informed box and ask, for each entry, what they would have a right to expect if this project were done well. Where you can answer that with something specific, you have found somebody whose involvement is not optional, whatever the grid says about their authority. Two or three names usually come out of that exercise, and they are almost always the names that would have appeared in a post-implementation review.
Interested in going further?
Deciding how much analysis a stakeholder picture deserves, and spotting the group whose legitimate claim has been filed under low priority, is a judgement that pays for itself long before go-live. Omega's PMP® Exam Preparation works through stakeholder decisions as they actually present themselves on projects.
The Stakeholders Performance Domain in the PMBOK® Guide Eighth Edition sets out the wider practice these two models sit inside.
Ad · Amazon affiliate link.
References
A132: The PMBOK 8 Stakeholders Performance Domain: What It Really Covers
A137: When Stakeholders Disagree: What Should the PM Do First?
A136: Why Hostile Stakeholders Still Need Engagement
A139: Communication Plans That People Actually Use
A135: Stakeholder Power Is Not the Same as Engagement
PMP and PMBOK are registered marks of the Project Management Institute, Inc.
