Stakeholder analysis: sorting the people who can stop you

Stakeholder analysis is the step where you work out who is affected by a piece of work, how much influence each of them has over whether it succeeds, and how much they care. The output is not a diagram — it is a communication plan, and if it does not change who you speak to next week, you have drawn a picture rather than done an analysis.

It is worth distinguishing from a stakeholder register, which is the list itself: names, roles, contact details, interest. The register records who they are. The analysis judges what each of them can do to you, and that judgement is what determines your effort.

How stakeholder analysis is usually structured

The common form is a two-by-two grid — power on one axis, interest on the other, sometimes called the Mendelow matrix. Four quadrants, four different treatments.

• High power, high interest: manage closely. Involve them in decisions, brief them before anyone else, and never let them learn something important from a third party.
• High power, low interest: keep satisfied. They can stop the work but do not want detail. Short summaries, no meetings they did not ask for, and immediate escalation of anything that touches their area.
• Low power, high interest: keep informed. Often the people doing the actual work or living with the result. They give the best early warning of practical problems.
• Low power, low interest: monitor. A note in the general update is enough, but check the classification occasionally — circumstances move people.

Power is not the same as seniority, which is where most first attempts go wrong. The person who administers the system you are replacing may have no title and complete practical veto: they know the exceptions nobody documented, and if they decline to help the project slows to nothing.

Doing it properly

1. List everyone affected, including people outside the organisation — customers, suppliers, regulators, an outsourced provider.
2. For each, write what they gain and what they lose. Losses drive behaviour far more reliably than gains.
3. Rate power as the practical ability to stop or delay, not job title.
4. Rate interest as how much the outcome touches their daily work.
5. Place them, then write the treatment for each quadrant as a specific commitment: who contacts them, how often, in what form.
6. Name an owner for each of the high-power relationships.
7. Revisit at every phase boundary. People move quadrants when the work reaches their area.

Stakeholder mapping beyond the grid

The grid is a starting point and it flattens some things worth keeping. Two additions repay the effort. One is attitude: supportive, neutral or opposed, which is orthogonal to power and interest — a high-power supporter and a high-power opponent both sit in the same box and need entirely different handling. The other is relationships between stakeholders: who listens to whom. Influence often travels sideways, and the fastest route to a sceptical executive is frequently a peer rather than a report.

That is also the honest limit of any stakeholder map. It records a judgement about people, made by one person, at one moment. Treat it as a working hypothesis rather than an assessment, keep it out of general circulation, and rewrite it when someone surprises you.

The part people skip

Almost every stakeholder analysis is done once, during initiation, and never touched again. This is the single reason the technique has a reputation for being busywork. Stakeholders change as work progresses: the finance director who was indifferent during design becomes highly interested the moment the invoice arrives, and the team lead who was engaged loses interest once their part ships.

Reviewing it at each phase boundary takes ten minutes and is the difference between a live tool and a document in the project folder.

Where to keep it

Ettex Board works for this because the output is actions rather than a document: each high-power relationship becomes a recurring card with an owner, so keep closely informed turns into a scheduled conversation on the same board as the rest of the work. The classification itself can live on the card.

There is no stakeholder matrix view and no automatic diagram — the grid is something you draw or describe in a card, not a feature. Given how sensitive the underlying judgements are, that suits most teams better than a chart anyone can open.

Two neighbours turn the analysis into something actionable. A competency matrix tells you which capabilities the group actually has rather than which titles it holds, and a swot analysis is the more familiar framing that stakeholder work usually feeds into rather than replaces.

Rate power by who can actually stop the work, add attitude to the grid, and rewrite the whole thing at each phase boundary. If it does not change your calendar, it was not an analysis.