# Firm Stakeholder Interview Plan

> Plans whom to hear about the client's problem by role and what each knows that others do not, with a neutral brief giving everyone the same facts, the same core questions asked separately before any group meeting, and a synthesis that keeps unique information apart from repeated opinion and records disagreements and missing voices. Use for "run firm-stakeholder-interview-plan", "plan the stakeholder interviews", "who should I talk to at the client", "everyone agreed after the boss spoke", "hear people separately before the workshop", "synthesise the interviews", "whose view am I missing", part of the Claude for Consultants · Team in Claude Pack by 90n.

- Skill: `polar-bear-org/firm-stakeholder-interview-plan` (Agent Skill)
- Install (CLI): `npx get-skill add polar-bear-org/firm-stakeholder-interview-plan`
- Raw SKILL.md: https://api.skillmd.com/api/skills/polar-bear-org/firm-stakeholder-interview-plan/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Product & Planning
- Author: polar-bear-org (https://skillmd.com/u/polar-bear-org)
- Updated: 2026-11-14
- Page: https://skillmd.com/skills/polar-bear-org/firm-stakeholder-interview-plan

---


# Stakeholder Interview Plan

## When To Use
In the first meeting the most senior person gave their reading and five heads nodded. You left with one view said six times. Run this when the diagnosis depends on what several people know and you want each account separately, before a group conversation flattens them.

## When Not To Use
To map who signs and who funds the purchase, use Buying Committee Map; that is about the buying decision, this is about the problem. Before signature, with access to one or two people only, a short list of questions from Hypothesis Test Plan may be enough.

## Inputs
- The problem statement and your current hypotheses
- The roles your contact can give you access to, and how much of their time
- What is sensitive, and what your contact has asked you not to raise
If you have none of this, I start from the root question and a list of roles to hear, and mark the plan as a first draft for your contact to correct.

## Approach
The rule comes from nominal group practice, described generically: collect independent first views before any discussion, because once a senior voice has spoken the others converge on it. Independence is a starting condition, not a ban on discussing things later. The failure it prevents is counting one opinion repeated by five people as five pieces of evidence. Nothing here rates the people interviewed; the plan weighs information, never individuals.

## Workflow
1. Ask, in one go: what do you think is missing from the picture so far? Which roles know something the others do not? How will your contact introduce the conversations?
2. List whom to hear by role, with what each can know first-hand: the people who do the work, those who receive its output, those who fund it, those who tried to fix it before. Include at least one role outside the senior circle.
3. Write a neutral brief of five or six lines: the same facts and the same open question for everyone, with no hint of what the sponsor or you believe.
4. Write four to six core questions asked of everyone in the same words (what happens today, what you have seen first-hand, what was tried, what you would check), then one or two per role. Each conversation happens separately, before any group session, in writing or by voice as the person prefers.
5. Synthesise in three passes. Unique information: who reported something no one else did, and on what basis. Repeated opinion: the same claim from several people, traced to its origin where you can. Disagreements: who sees it differently and what evidence each side has. Promise no anonymity you cannot keep.
6. List the missing voices and what you could not ask. Hand your contact the evidence map and the open questions; a show of hands does not settle what the sponsor must decide.

## Output Format
```markdown
# Stakeholder Interview Plan
Problem: [root question] · Access agreed with: [contact role] · Date: [date]
## Whom to hear
| Role | Knows first-hand | Format | When (before any group session) |
|---|---|---|---|
| [role] | [what only this role sees] | [call / written] | [date] |
## Neutral brief (same for everyone)
[Five or six lines: shared facts, the open question, how the answers will be used]
## Core questions (same words for everyone)
1. [question]
## Synthesis
| Type | What was reported | Reported by (role) | Basis |
|---|---|---|---|
| Unique information | [fact] | [role] | [seen first-hand / record] |
| Repeated opinion | [claim] | [roles; origin if known] | [basis] |
| Disagreement | [claim against claim] | [roles] | [evidence on each side] |
## Missing voices and open questions
- [Role not heard, or question not asked, and why]
## Decision
[You] agree the list and the brief with [contact role] by [date]; [client decider role] receives the evidence map and decides what to do with what remains open.
```

## Done When
- Every interview happens before any group discussion of the question
- The neutral brief gives no hint of the sponsor's view or yours
- The synthesis separates unique information from repeated opinion and records each disagreement
- Missing voices are listed, not left out

## Quality Bar
- The same core questions in the same words for every role
- Nothing rates, ranks or characterises the people interviewed
- No anonymity is promised that the set-up cannot protect
- A majority view is reported as a majority view, never as the answer
- I plan the questions and sort what you bring back; I invent no stakeholder view, and you hold every conversation yourself.

## Next
Run firm-strategic-options (Strategic Options) to lay out the real ways the client could tackle the problem.

## About the makers

This 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).

