Stakeholder analysis template: one decision, one table
Most stakeholder analysis templates are built for project management: a list of names, a power score, an interest score. For a decision that will be read by investors, regulators, staff and the press, the table needs to say more. Here is one that does.
In short
- Build the table around one decision, written in one sentence at the top.
- Go past power and interest. Record what each actor stands to lose, where they stand now, where you need them, and who they listen to.
- Every judgement gets a source and a confidence rating. A guess with no evidence should look like a guess.
- The table records positions before you act. How actors react to the move, and to each other, has to be tested another way.
What a stakeholder analysis is for
A stakeholder analysis is the working record of who a decision affects, what they want, what they can do about it, and what you plan to do about them. It is the table behind the stakeholder map: the map gives the picture, the analysis holds the reasoning.
For a senior team, its job is to make three things explicit before anything is announced: whose reaction can change the outcome, how far each of them is from where you need them, and who is responsible for closing that gap.
The template
Copy these columns into a spreadsheet. Put the decision in one sentence above the table, and date it. One row per named actor: "the sector regulator's supervision team", not "regulators".
| Column | What to write |
|---|---|
| Stakeholder | Group and named actor, e.g. "Employees: claims staff at the two closing offices" |
| Stake | What they gain or lose if the decision goes ahead, in their terms rather than yours |
| Power | 1–5. Can they stop, delay, fine, reverse or publicly damage the decision? |
| Interest | 1–5. How much it affects them, and whether they know yet |
| Legitimacy / urgency | Yes / no for each, one line of reasoning. Drawn from the salience model |
| Current position | Support, neutral or oppose, and how firmly, as of today |
| Needed position | The least you need from them for the decision to land: active support, acceptance, or silence |
| Gap | The distance between the two positions. This column sets the priority more than power does |
| Listens to | The other actors, outlets or people whose view shapes theirs |
| Risk if mishandled | What happens if they hear it late, or hear it badly |
| Action and owner | The engagement step, who owns it, and when it happens relative to others |
| Evidence | Past statements, filings, coverage, survey data or conversations behind the row |
| Confidence | High, medium or low. Low-confidence rows with high power are the ones to research first |
If the table is too wide for your purpose, the columns to keep are stake, power, current and needed position, listens to, owner and evidence. Cut the others before you cut those.
How to fill it in
- Write the decision first. What changes, for whom, from when. A vague decision produces a table where every row says "neutral".
- List actors, not categories. Start with the broad groups (investors, regulators, employees, customers, media, government, partners, communities), then replace each with the named actors who carry weight inside it.
- Draft alone, score together. One person drafts the rows. The scores are agreed in a session with the people who own each relationship. Where two people disagree on a score, record both views in the evidence column.
- Fill the gap column before the action column. It is tempting to plan engagement for everyone. The gap tells you where engagement will change something.
- Trace the "listens to" column. If the regulator reads one trade reporter, and that reporter talks to your staff, the order in which you brief those three matters more than any single message.
- Sort. High power with a large gap goes to the top. Low confidence with high power gets a research task before it gets a message.
- Set a review date. Revisit the table when the decision changes, when a new actor appears, and after any leak or public comment.
Worked example (hypothetical)
Decision: a mid-sized insurer will move claims handling from two regional offices into one central site over the next year, offering relocation or redundancy to affected staff.
| Stakeholder | Stake | Now → needed | Listens to | Action |
|---|---|---|---|---|
| Claims staff, closing offices | Jobs, commute, family ties to the town | Oppose → acceptance | Union rep, local press, each other | Told first, in person, with the support package written down |
| Union | Members' terms; its standing with members | Oppose → negotiate | Members, national office | Briefed before staff hear it elsewhere |
| Insurance regulator | Claims service continuity for policyholders | Neutral → satisfied | Complaints data, trade press | Transition plan showing service levels during the move |
| Local MP, closing towns | Constituents' jobs | Unknown → not campaigning | Local press, constituents | Called on the day, before the press release |
| Policyholders with open claims | Their claim not stalling | Unaware → reassured | Their handler, review sites | Letter naming their handler through the transition |
The example's lesson sits in the "listens to" column. Staff, the union, local press and the MP read each other. Brief them out of order and the story becomes "insurer abandons town" before the regulator has seen the service plan.
Mistakes that make the table useless
- Rows for groups, not actors. "Media: medium interest" cannot be acted on. "The two reporters who cover the sector" can.
- Scores without evidence. A power score of 4 with nothing in the evidence column is an opinion with a number attached.
- Positions in your words. The stake column should read the way the stakeholder would say it. "Efficiency programme" is your phrase; "losing my job" is theirs.
- A table drawn once. Positions move after every public moment. A table dated three months ago describes a different situation.
- No owner. An action without an owner does not happen.
What the table cannot tell you
A stakeholder analysis records how you understand each actor before you act. It is a strong planning tool and a weak predictor, for three reasons:
- Positions are judged one row at a time. Reactions are not. The union's stance after the announcement depends partly on what the local press and the MP say first.
- The move changes the field. Wording, timing and who hears it first can turn a neutral row hostile. The table holds the position, not the reaction to the move.
- It cannot show which story wins. The same decision can be read as cost-cutting, as service improvement, or as a company leaving a town. The table does not say which reading spreads.
Teams test this with war-gaming sessions or red teams who argue the hardest rows. Both help, and both are limited to who is in the room.
Testing the analysis before you act
Isaiah picks up where the table stops. You bring the decision and the documents that frame it. Isaiah builds the stakeholder field from that material and fresh research, simulates those stakeholders responding to your move and to each other in simulated public discourse, and tests candidate counter-moves. The board-ready report shows where coalitions form, which narratives spread and which moves backfire, with each finding traced to quotes from named simulated stakeholders. The simulation stays open so you can question any of them.
It is a stress-test, not a forecast. Simulated stakeholders are not survey respondents, and the output does not replace your advisers or your judgement. It shows which rows of your table are most likely to move once the decision is public.
Frequently asked questions
What should a stakeholder analysis template include?
At minimum: the stakeholder and a named actor, what they stand to gain or lose, their power over the outcome, their current and needed position, who they listen to, the risk if they are handled badly, the engagement action and its owner, and the evidence behind each judgement. A confidence rating on each row stops guesses from reading like facts.
Is a stakeholder analysis the same as a stakeholder map?
No. The map is the picture: a grid or network that shows priorities at a glance. The analysis is the table behind it, with the reasoning, evidence and actions for each actor. Most teams build the table first and draw the map from it.
How many stakeholders should the template list?
As many as the decision touches, but rows should be named actors rather than broad groups once you pass the first draft. A long list is fine if it is sorted; the top rows, the ones with high power and a position gap, are where the team's time goes.
Who should fill in a stakeholder analysis?
The people who own the relationships: corporate affairs, investor relations, HR, regulatory, legal and the business lead for the decision. One person can draft it, but the scores should be agreed in a working session where disagreements are written down.
Have a decision coming up?
Tell us what it is. We will set up a private stress-test on it and show you how the stakeholder field reacts before anything goes public.
Request a demoRelated
- Stakeholder mapping: how to map a decision before you act
- Power interest grid: how to use it for a real decision
- Holding statement: what to say before you know everything
- Crisis communication plan: what goes in and how to test it
Sources
- Mitchell, R. K., Agle, B. R., & Wood, D. J. (1997). Toward a Theory of Stakeholder Identification and Salience: Defining the Principle of Who and What Really Counts. Academy of Management Review, 22(4), 853–886. doi.org/10.2307/259247 (checked 8 November 2026)
- Government Analysis Function (UK). Stakeholder mapping: assessing power and interest. analysisfunction.civilservice.gov.uk/policy-store/stakeholder-mapping (checked 8 November 2026)