Consulting & AdvisoryFree
Stakeholder Analysis and Org Chart Check
Your client's org chart reconciled against their active headcount list, then an interview list ranked by what only each person can answer.
River reads the org chart, the headcount export and whatever record exists of who decided things, then builds the stakeholder register from all three rather than from the chart alone. Every row carries the person's current role as the headcount list has it, the document that evidences their influence, and the engagement questions they can answer. The interview list comes back ranked by unique coverage, so no slot goes to the fourth person who can describe something three people already covered.
Search this and you get Mendelow's power and interest grid, eight times over, with a register of name, role, power, interest and quadrant sitting behind it. Every version opens on step one, identify your stakeholders, and not one says where that list comes from or whether the people on it still work there. A grid is a sorting tool. It takes a list as its input and it cannot tell you the list is wrong.
Built for the consultant with sixteen slots and three weeks, the fractional operator walking into a business they are about to run, and the diligence lead who has to know who signs what. Run it in week one, once the reading order for an unindexed data room has said which files hold people data. The ranked list is then the input to the interview programme that schedules the round. Every row traces back to a file the client sent, so the sponsor can check any of it.
Influence is a document, not a rating
The published tests for who matters in a company ignore job titles. The SEC defines an executive officer as any vice president in charge of a principal business unit, division or function, plus anyone else performing a policy making function. It then adds that executive officers of subsidiaries may be deemed executive officers of the registrant where they perform that function for it. Read as a stakeholder test, that reaches the group pricing director whose box sits in a different company and who sets your client's discount floor.
Calderbeck Packaging, six corrugated plants and 1,180 people, sent a chart exported fourteen months earlier carrying 58 boxes. Joined to the headcount export on employee id, 41 boxes name a person still in that role and seven name somebody who has left. Six are held by a different person than the chart says, and four describe a role that no longer exists. Eleven people at that level sit on no box, two of them employed by the listed parent. The chart correctly names 41 of the 63 people who hold those roles.
Then rank the list rather than working down it. Calderbeck's engagement carried six questions and sixteen slots. Seven interviews reached three independent sources on five questions and both of the only two sources on the sixth, leaving nine slots for whatever those seven turned up. A chart-driven list of the top sixteen by seniority filled every slot, spent four on function heads the pricing question never touches, and left the parent company question with nobody at all.
How it works
Send the material
The chart in any format, the headcount export, the approval records, the delegation schedule.
Name the questions
What the engagement has to answer, since that is what decides who is worth an hour.
Read the reconciliation
Every box classified against a real person, and every person the chart never had.
Book from the ranking
Work down the ranked list in chat, moving names as access and politics change.
What you get
- A stakeholder register built from the chart, the headcount export and the approval records together
- Every box classified: correct, departed, held by somebody else, or a role that closed
- The people on no box at all, each with the reason the chart never carried them
- An influence rating on every row that names the document standing behind it
- Span of control read off manager ids rather than off the boxes somebody drew
- The interview list ranked by unique coverage, with the unanswerable questions named
Common questions
What does it need from me?
The org chart in whatever form it arrived, a headcount export with employee ids and manager ids, and the engagement questions. Add the delegation of authority schedule, any forum minutes and any approval log and the influence column fills itself from documents rather than from your read of the room after two meetings.
The chart is a PDF from two years ago. Is that usable?
Yes, and it is the normal case. A chart is read as a claim about a point in time, dated from what it says about itself, then tested box by box against the current headcount list. The output is the classification: which boxes survived, which name somebody who left, and which are held by a successor nobody wrote down.
How is this different from a power and interest grid?
A grid sorts a list you already have. This produces the list and tests it. You can still plot the result on any two axes you like, and those axes will be worth plotting. The influence score on each row names the delegation entry, the signature or the forum seat behind it rather than a number somebody chose.
How can it tell who really has authority?
By reading what the client already wrote down. Federal contracting takes the strict version of this: a contracting officer binds the government only as far as the delegation goes, and the appointment is in writing with its limits stated. Signature limits, forum charters and approval logs are the same evidence in a private company.
I cannot get HR data. What then?
The reconciliation runs against whatever names real activity: distribution lists, calendar invites, approval logs, document authors, signature blocks in the contract file. It is a weaker join than employee ids and the register says which rows rest on it, so nothing is presented as confirmed when it was inferred.
Does it tell me who not to interview?
It tells you which slots buy nothing, which is the same information with the argument attached. A correctly named director who answers none of your questions is a courtesy call, not an interview, and the ranked list says so with the coverage count that makes the case. Overrule any row. The reason is on the sheet.
Where does this fit in the engagement?
Week one, alongside the rest of engagement intake and the review of the previous firm's work, whose sponsor has usually left. The headcount export is personal data, so every copy of it gets a row in the data handling register. Once the interviews are done, synthesizing the transcripts needs the real role headcounts this produces.
Stakeholder Analysis and Org Chart Check
Fill in the form and your workspace opens with the work already underway.