Stakeholder Map
When To Use
You know the CEO well and nobody else, and the work depends on four other people. The finance lead holds the numbers, a team lead carries the function on your off days, a board member asks the questions, and you have met each once. This answers: who does the outcome depend on, what has each said they need, and where do I have no line at all?
When Not To Use
If the person who hired you is leaving, run Sponsor Change Plan. If one working relationship has already gone cold over a missed commitment, run Relationship Repair Plan. This map does not rate, rank or profile anyone.
Inputs
- The people you deal with or will need, by name and role, and when you last spoke to each
- What each has said they need, with where they said it (meeting, message, document)
- The outcome and the plan, so dependencies can be tied to dated work
If you have none of this, I start from the org chart or your memory of the first meetings, mark every stated need "unknown" and mark the map as a first draft.
Approach
A two-way dependency map by role and stated need, a practitioner convention. Each person is placed by their role in the work, never on a grid of power or attitude; grids of that kind are left out on purpose, because they turn impressions into labels. The judgment is to write only what someone said and to turn every blank into a question you ask in person. The failure this prevents is the single thread: the whole engagement hangs on the CEO's diary and nobody else can say why your seat matters. Works well in Projects (beta), one Project per client, or a plain chat.
Workflow
- Ask at most three questions: who does the outcome depend on besides the CEO; whom have you not yet met; where has work already waited on someone?
- Place each person by role in the work: decides, funds, uses the result, can block, advises. One person may hold two roles; write both.
- Record what each has said they need, with the source and date. A need you inferred is written "inferred" and becomes a question. Nothing is recorded about personality, mood or loyalty.
- Write the dependency both ways, with dates: what you need from them (data, a decision, cover on your off days) and what they need from you. Late dependencies on either side are where the work stalls.
- Mark the gaps: not met, reached only through the CEO, no stated need, no contact since [date]. Each gap becomes a question or an introduction to ask for.
- Choose the first three conversations and what you will ask in each. Update the map after every conversation; it is yours, and it is not shared.
Output Format
# Stakeholder Map
Client: [name] · Outcome: [one line] · Updated: [date]
## People by role in the work
| Person | Role in the work | What they said they need (source, date) | I need from them, by when | They need from me, by when | Last contact | Gap |
|---|---|---|---|---|---|---|
| [name, title] | [decides / funds / uses / can block / advises] | [need, source] or unknown | [item, date] | [item, date] | [date] | [none / not met / only through the CEO / no stated need] |
## Unknowns turned into questions
| Unknown | Question | Ask whom | By when |
|---|---|---|---|
| [what you do not know] | [open question] | [person] | [date] |
## First conversations
1. [person]: [what you will ask] · [date]
## Decision
[You] decide which three gaps to close first, and ask [CEO] for the introductions you need by [date].
Done When
- Every person appears by role and stated need, with no rating of any kind
- Every stated need has a source, or reads "unknown" or "inferred"
- Dependencies run both ways and carry dates
- Each gap has a question and a person to ask
Quality Bar
- No power, influence, attitude or sentiment label on anyone, in any column
- Unknown stays unknown; I never guess what someone wants or thinks
- Nothing private, and nothing a person told you in confidence, goes on the map
- The map is your working page; it is not sent to the client
- Only what you give me is recorded, nothing is invented, and the map holds one client's people and never another's
Next
Run frac-first-30-days-diagnostic (First 30 Days Diagnostic) to turn those conversations into evidence and a written assessment.
About the makers
This pack is made by 90n (90n.ai): AI-native ways of working and tools for Product, Marketing and HR teams, built by ex-McKinsey founders. We map where time and money leak, redesign the work around AI and stay until your people run it themselves, and we run our own company on Claude. If your team has outgrown the self-serve version, message Pauline (linkedin.com/in/paulinebertry).
1---2name: frac-stakeholder-map3description: Maps the people a fractional engagement depends on by their role in the work (decides, funds, uses, can block, advises), what each has said they need, what you depend on them for and they on you, last contact, and the gaps and unknowns turned into questions, with roles and stated needs only and no ratings. Use for "run frac-stakeholder-map", "stakeholder map for my client", "stakeholder mapping", "who do I need to know at this client", "I only know the CEO", "map who the work depends on", "who else should know why I am here", part of the Claude for Fractional Leaders Pack by 90n.4metadata:5 source: https://skillmd.com/skills/polar-bear-org/frac-stakeholder-map6---78# Stakeholder Map910## When To Use11You know the CEO well and nobody else, and the work depends on four other people. The finance lead holds the numbers, a team lead carries the function on your off days, a board member asks the questions, and you have met each once. This answers: who does the outcome depend on, what has each said they need, and where do I have no line at all?1213## When Not To Use14If the person who hired you is leaving, run Sponsor Change Plan. If one working relationship has already gone cold over a missed commitment, run Relationship Repair Plan. This map does not rate, rank or profile anyone.1516## Inputs17- The people you deal with or will need, by name and role, and when you last spoke to each18- What each has said they need, with where they said it (meeting, message, document)19- The outcome and the plan, so dependencies can be tied to dated work20If you have none of this, I start from the org chart or your memory of the first meetings, mark every stated need "unknown" and mark the map as a first draft.2122## Approach23A two-way dependency map by role and stated need, a practitioner convention. Each person is placed by their role in the work, never on a grid of power or attitude; grids of that kind are left out on purpose, because they turn impressions into labels. The judgment is to write only what someone said and to turn every blank into a question you ask in person. The failure this prevents is the single thread: the whole engagement hangs on the CEO's diary and nobody else can say why your seat matters. Works well in Projects (beta), one Project per client, or a plain chat.2425## Workflow261. Ask at most three questions: who does the outcome depend on besides the CEO; whom have you not yet met; where has work already waited on someone?272. Place each person by role in the work: decides, funds, uses the result, can block, advises. One person may hold two roles; write both.283. Record what each has said they need, with the source and date. A need you inferred is written "inferred" and becomes a question. Nothing is recorded about personality, mood or loyalty.294. Write the dependency both ways, with dates: what you need from them (data, a decision, cover on your off days) and what they need from you. Late dependencies on either side are where the work stalls.305. Mark the gaps: not met, reached only through the CEO, no stated need, no contact since [date]. Each gap becomes a question or an introduction to ask for.316. Choose the first three conversations and what you will ask in each. Update the map after every conversation; it is yours, and it is not shared.3233## Output Format34```markdown35# Stakeholder Map36Client: [name] · Outcome: [one line] · Updated: [date]37## People by role in the work38| Person | Role in the work | What they said they need (source, date) | I need from them, by when | They need from me, by when | Last contact | Gap |39|---|---|---|---|---|---|---|40| [name, title] | [decides / funds / uses / can block / advises] | [need, source] or unknown | [item, date] | [item, date] | [date] | [none / not met / only through the CEO / no stated need] |41## Unknowns turned into questions42| Unknown | Question | Ask whom | By when |43|---|---|---|---|44| [what you do not know] | [open question] | [person] | [date] |45## First conversations461. [person]: [what you will ask] · [date]47## Decision48[You] decide which three gaps to close first, and ask [CEO] for the introductions you need by [date].49```5051## Done When52- Every person appears by role and stated need, with no rating of any kind53- Every stated need has a source, or reads "unknown" or "inferred"54- Dependencies run both ways and carry dates55- Each gap has a question and a person to ask5657## Quality Bar58- No power, influence, attitude or sentiment label on anyone, in any column59- Unknown stays unknown; I never guess what someone wants or thinks60- Nothing private, and nothing a person told you in confidence, goes on the map61- The map is your working page; it is not sent to the client62- Only what you give me is recorded, nothing is invented, and the map holds one client's people and never another's6364## Next65Run frac-first-30-days-diagnostic (First 30 Days Diagnostic) to turn those conversations into evidence and a written assessment.6667## About the makers6869This pack is made by 90n ([90n.ai](https://90n.ai)): AI-native ways of working and tools for Product, Marketing and HR teams, built by ex-McKinsey founders. We map where time and money leak, redesign the work around AI and stay until your people run it themselves, and we run our own company on Claude. If your team has outgrown the self-serve version, message Pauline (linkedin.com/in/paulinebertry).