Stakeholder mapping is the exercise of listing everyone who can affect your project or be affected by it, then sorting them by two things: how much power they hold over the outcome, and how much they actually care. The output is usually a four box grid, sometimes called the power interest grid, and a table that says what each person gets from you and who is responsible for giving it to them. It takes about ninety minutes for a normal project. Skipping it costs considerably more than ninety minutes later.
The reason it matters is not diplomatic. It is that attention is finite. A project manager can hold a real relationship with maybe six or seven people. If you spend that budget on the loudest voices rather than the ones who can stop the project, you find out during the approval meeting that the person who can stop the project has not been briefed since March. Mapping is how you decide, deliberately and in advance, who gets the six slots.
Key takeaways
- Two axes do almost all the work: power over the outcome, and interest in it. Everything else is decoration.
- Map to a decision, not to an org chart. The question is "who can say no", not "who is senior".
- The grid is worthless without the second table. The grid sorts people; the matrix says what each one receives and who sends it.
- Score current engagement and desired engagement separately. The gap between the two columns is your actual to do list.
- Redraw the map when somebody changes job, not on a schedule. Reorganizations invalidate stakeholder maps faster than anything else.
What is stakeholder mapping?
Stakeholder mapping is a structured way of identifying the people and groups with a stake in a project and classifying them so you can plan how much effort each one deserves. In practice it means three passes: list everyone, score each of them on power and interest, then place them into a grid that tells you the engagement approach for each quadrant. The map is the classification, not the list.
The vocabulary around this is a mess, and it costs teams time. Stakeholder identification is the listing pass. Stakeholder analysis is the scoring pass, where you work out what each person wants and how much leverage they have. Stakeholder mapping is usually taken to mean the visual output, the grid or diagram. Stakeholder management and stakeholder engagement describe what you do afterwards, over the life of the project. Most people use all five terms interchangeably. When somebody asks you for a stakeholder map, ask whether they want the grid or the whole analysis, because one is an afternoon and the other is a week.
One boundary worth setting early: the map is not the same thing as the register. A stakeholder register is a contact list with roles and details, maintained for the record. A map is a judgment about relative influence, made for a decision, and it goes stale far faster. Plenty of organizations maintain a beautiful register and have never once used it to decide who needs a phone call before a steering committee meeting.
The power interest grid, quadrant by quadrant
The power interest grid places each stakeholder on two axes and produces four groups, each with a different engagement approach. It is often credited to the strategy literature of the 1990s and it has survived because two axes are about as much as anyone can hold in their head during a workshop. Three axes look more rigorous and get abandoned by the second meeting.
| Quadrant | Approach | Who is usually here | What they get | The mistake |
|---|---|---|---|---|
| High power, high interest | Manage closely | Sponsor, budget holder, the operational lead whose team changes most | Direct contact, early sight of bad news, a say in trade off decisions | Letting this group grow past five or six people, at which point nobody is managed closely |
| High power, low interest | Keep satisfied | Finance director, legal counsel, a divisional head whose numbers you touch once | Short, infrequent, exception based updates. No detail unless asked | Flooding them with weekly reports, which trains them to ignore you right up to the day you need them |
| Low power, high interest | Keep informed | End users, support teams, subject matter experts, the people who will live with the result | Regular detail, genuine consultation on how rather than whether, a route to raise problems | Treating consultation as communication. This group notices instantly and stops engaging |
| Low power, low interest | Monitor | Adjacent teams, peripheral suppliers, functions touched only at go live | Access to published material. Nothing pushed | Deleting them from the map entirely, then discovering at cutover that one of them owned an interface |
Swipe to see more →
The most common error is not misplacing someone. It is scoring power by job title. A senior person with no budget authority over your project and no veto is not high power for your purposes, however important they are elsewhere. Conversely the systems administrator who controls the release window is high power on a technology project regardless of grade. Score power over this decision.
How to do a stakeholder analysis in five steps
A stakeholder analysis is the work behind the grid. Five passes, in this order, because doing them out of order produces a map that argues with itself.
- List without filtering. Names, not departments. Go through the org chart, the project charter, the supplier contracts, the systems that will change, and the teams downstream of those systems. Aim for thirty names and cut later. The names people forget are almost always in operations, in compliance, and among the second line of suppliers.
- Record what each one wants. One sentence per person, in their language, not yours. "Wants the migration finished before the audit in October" is useful. "Supports the project" is not. If you cannot write the sentence, you do not know them well enough yet and that is itself a finding.
- Score power and interest. Use one to five on each axis, and force yourself to justify anything you score a five. Do this with two people rather than alone. Scores done alone reflect who you happen to have spoken to recently.
- Place them and check the shape. Plot the scores. If more than seven people land in the manage closely quadrant, your scoring was generous and you need to re-score, because you cannot manage twelve people closely. If the quadrant is empty, you have missed the real decision maker.
- Turn the map into commitments. Every stakeholder gets a named owner on your side, a channel, and a frequency. That table is the deliverable. A grid with no owner column is a picture, and pictures do not get followed.
Step two is where most of the value is and where most teams rush. The point of writing down what each person wants is that it surfaces the conflicts before they become escalations. If the finance director wants the cost booked this fiscal year and the operations lead wants the cutover after the peak season, you have a scheduling problem in week one rather than a crisis in month six. That kind of early conflict discovery is exactly what a good project kickoff meeting exists to force into the open.
The stakeholder matrix template, column by column
The matrix is the table the grid feeds. This is the artifact people actually reuse, and it fits on one page for a normal project. Here is each column, what belongs in it, and the test for whether an entry is any good.
| Column | What goes in it | Test for a good entry |
|---|---|---|
| Name | The individual, plus their role in one phrase | A person, never a department. "Finance" cannot be phoned |
| Stake | What they gain or lose if the project succeeds or fails | Written from their side of the table, in a full sentence |
| Power (1 to 5) | Their ability to change, delay, or stop this project | You can name the specific mechanism: a budget, a sign off, a resource, a veto |
| Interest (1 to 5) | How much the outcome affects their day to day work or targets | Based on something they have said or done, not on what you assume they should care about |
| Quadrant | Manage closely, keep satisfied, keep informed, or monitor | Derived from the two scores, not chosen first and justified afterwards |
| Current engagement | Unaware, resistant, neutral, supportive, or leading | Honest. Marking a known blocker as neutral to avoid an awkward conversation makes the whole table useless |
| Desired engagement | The level you actually need from them, which is often not "leading" | Differs from current for no more than a handful of people. If every row has a gap, you are describing a wish, not a plan |
| Relationship owner | The named person on the project side who holds this relationship | Not always the project manager. Sponsors are often the right owner for peer level executives |
| Channel and frequency | How and how often, in specifics | "Monthly one to one, thirty minutes" rather than "regular contact" |
| Last contact | The date somebody on the project last actually spoke to them | Filled in. This one column exposes more neglected relationships than the rest of the table combined |
Swipe to see more →
The last contact column is the cheapest diagnostic in project management. Sort the table by it before a steering meeting and you will find, reliably, that at least one high power stakeholder has not heard from anyone in two months.
Stakeholder mapping example: a finance system replacement
Abstract templates are easy to agree with and hard to copy. Here is a filled map for a mid sized project, replacing a finance system across four business units over nine months, with a project manager, a business analyst, and a systems integrator.
| Stakeholder | Stake | Power | Interest | Quadrant | Current | Desired | Owner and cadence |
|---|---|---|---|---|---|---|---|
| CFO (sponsor) | Owns the business case and the savings target | 5 | 5 | Manage closely | Supportive | Leading | PM, weekly 20 minute call |
| Financial controller | Her team does the parallel run and the reconciliation | 4 | 5 | Manage closely | Neutral | Supportive | PM, twice weekly during parallel run |
| Head of IT operations | Owns the release window and the infrastructure | 5 | 3 | Manage closely | Resistant | Supportive | PM, weekly, plus a standing agenda item at the IT change board |
| Group legal counsel | Data processing terms in the integrator contract | 4 | 1 | Keep satisfied | Unaware | Neutral | BA, exception based only |
| Divisional MD, Business Unit 3 | His unit absorbs the biggest process change | 4 | 2 | Keep satisfied | Neutral | Supportive | Sponsor, monthly, peer to peer |
| Accounts payable team lead | Her team's daily work changes most | 2 | 5 | Keep informed | Resistant | Supportive | BA, fortnightly working session |
| Internal audit manager | Needs the control environment evidenced before year end | 3 | 4 | Keep informed | Neutral | Supportive | BA, monthly, plus sight of the control mapping |
| Integrator engagement lead | Contract margin and reference customer | 3 | 5 | Keep informed | Supportive | Supportive | PM, weekly delivery call |
| Payroll manager | One interface, otherwise unaffected | 2 | 2 | Monitor | Unaware | Neutral | BA, notify at interface testing |
| Works council representative | Consultation on role changes in shared services | 4 | 3 | Manage closely | Unaware | Neutral | Sponsor, brief before any headcount discussion |
Swipe to see more →
Three things are worth noticing in that table, because they are the things a template will never teach you. The head of IT operations is scored resistant and high power, which makes him the single most important row on the page; that row is a project risk and belongs in the project risk register as well as here. The works council representative is currently unaware and holds a genuine veto, which is the classic late discovery that pushes go live by a quarter. And the CFO's desired engagement is set to leading rather than supportive, because a sponsor who merely supports a project does not defend its budget in October. If your project sponsor is sitting at supportive, moving them one level is usually the highest value engagement work available to you.
The stakeholder engagement assessment matrix
The engagement assessment matrix is the second grid, and it is the one that turns analysis into work. You score each stakeholder's current engagement level, then mark the level you need, and the gaps become your engagement plan. Five levels are standard.
| Level | What it looks like in practice | What moves someone up a level |
|---|---|---|
| Unaware | Does not know the project exists, or knows the name and nothing else | A single direct conversation. Not an email, not a newsletter |
| Resistant | Knows about it and is against it, openly or quietly | Finding the actual objection. It is almost never the one stated first, and it is usually about workload, timing, or a previous project that went badly |
| Neutral | Aware, neither helping nor blocking, will not spend their own capital on it | Giving them something concrete they want from the project |
| Supportive | Speaks well of it, cooperates when asked, does not initiate | Asking them to own something visible, not just to approve it |
| Leading | Actively drives it, spends their own credibility on it, chases others | Nothing. Protect this and do not waste it on low value asks |
Swipe to see more →
The discipline here is that "desired" is rarely "leading". You need one or two people at leading, several at supportive, and it is perfectly fine to leave a peripheral stakeholder at neutral forever. A matrix where every row needs to move two levels is a matrix nobody will act on. Pick the three rows that matter and work those.
Moving somebody out of resistant is the only genuinely difficult move on the table, and it is worth being blunt about how it is done. You ask them, in private, what would have to be true for them to be comfortable, and then you either fix it or you tell them honestly that you cannot. Most resistance in project settings is not opposition to the goal. It is a well founded belief that the project will land extra work on an already stretched team, at a bad time, without extra resource. That is a resourcing conversation, and it belongs with your capacity planning, not with a persuasion campaign.
Stakeholder mapping vs the register, the RACI, and the communication plan
Four artifacts in this space overlap and teams routinely maintain three versions of the same information. Here is the seam this site uses, and it is worth adopting one explicitly rather than letting each project invent its own.
| Artifact | Question it answers | Changes when | Owned by |
|---|---|---|---|
| Stakeholder map or matrix | Who matters, how much, and what does each one need from us | People change roles, or the project's scope moves into a new area | Project manager |
| Stakeholder register | Who are they, formally: name, role, contact, category | Joiners and leavers | Project administrator or PMO |
| RACI matrix | Who is accountable for each piece of work | The work breakdown changes | Project manager, signed off by the sponsor |
| Communication plan | Who receives which information, how often, from whom | The stakeholder map changes, which is why the map comes first | Project manager |
Swipe to see more →
The dependency runs one way and it is worth stating plainly: the map feeds the communication plan. If you write the communication plan first, you get a distribution list. If you map first, you get a plan where the frequency for each audience is justified by their power and their interest rather than by habit. Where the two meet is the "I" in RACI, the people who must be informed, and the map is what tells you how informed is enough.
How a PMO runs stakeholder mapping across a portfolio
At single project level, mapping is a project manager's private tool. At portfolio level it becomes something more useful, and most PMOs never exploit it. Three things change.
First, the same forty people appear on twenty project maps. The head of IT operations is high power on every technology project in the portfolio. Once you can see that, you can stop twenty project managers independently competing for his attention and instead give him one consolidated view at one cadence, which is both cheaper for him and more effective for you. This is one of the quieter arguments for running a real portfolio review meeting rather than twenty separate steering committees.
Second, the current engagement scores aggregate into something a portfolio director can act on. If four projects all score the same divisional MD as resistant, that is not four communication problems. That is one unresolved disagreement about strategy, and it needs handling at portfolio level, probably at the steering committee, not by four project managers writing better emails.
Third, a stakeholder map is an early warning system for a specific and expensive failure mode: the project whose sponsor has quietly disengaged. Track the last contact date and the sponsor engagement level for every project in the portfolio, and review them monthly. A project whose sponsor has dropped from leading to supportive and has not been spoken to in six weeks is at risk long before that shows up in the schedule or the RAG status. Of all the soft signals available to a PMO, this one has the best ratio of effort to warning time.
A standing PMO rule worth adopting: require a refreshed stakeholder map at every stage gate rather than annually. Gates are when scope moves and when new functions get pulled in, so gates are when maps go stale. Wiring it into the stage gate process costs a project manager an hour and catches the newly relevant stakeholder before they become a late objection.
Where stakeholder mapping goes wrong
| The failure | What it looks like | The fix |
|---|---|---|
| Mapping the org chart | Every senior person is high power, everyone junior is low | Score power over this decision, and name the mechanism |
| Doing it once | A map dated the week of kickoff, referenced in month seven | Refresh at each gate and whenever anyone on it changes role |
| No owner column | A tidy grid, and nobody has called the works council | Every row gets a named person on the project side |
| Polite scoring | Nobody is marked resistant because the file might be shared | Keep the engagement columns in a working copy the project team controls |
| Thirty people in "manage closely" | Everyone is a priority, so the map gives no guidance at all | Cap the quadrant at six or seven and force the trade off |
| Confusing consultation with information | Users are asked for input after the decision is made | Be explicit about which decisions are open and which are not |
Swipe to see more →
The polite scoring problem deserves a note, because it quietly destroys the artifact in large organizations. The moment a stakeholder map with an engagement column becomes a document that might be circulated, every entry drifts toward neutral and supportive, and the map stops carrying information. Keep the honest version as a working tool for the project team, and publish only the parts that need publishing. That is not deception; it is the same reason a RAID log is more candid than a status report.
Frequently asked questions
What are the 4 types of stakeholders?
In the power interest model the four types are the four quadrants: high power and high interest, who you manage closely; high power and low interest, who you keep satisfied; low power and high interest, who you keep informed; and low power and low interest, who you monitor. A separate common split is internal versus external and primary versus secondary, which describes where stakeholders sit rather than how to treat them.
What is the difference between stakeholder mapping and stakeholder analysis?
Stakeholder analysis is the whole investigation: identifying everyone with a stake, working out what each of them wants, and judging how much influence they hold. Stakeholder mapping is the classification step within that, where you place people on a grid so you can decide how much attention each deserves. In everyday use the terms are treated as synonyms and little harm comes of it.
What is the power interest grid?
The power interest grid is a four box chart that plots stakeholders by how much power they have over a project on one axis and how interested they are in it on the other. Each box carries a different engagement approach: manage closely, keep satisfied, keep informed, or monitor. It is the most widely used stakeholder mapping technique because two axes are simple enough to actually get used.
How do you write a stakeholder analysis?
Work in five passes. List every person with a stake, by name. Write one sentence per person describing what they want, in their words. Score each on power and interest from one to five. Place them into the grid and sanity check the shape. Then convert the grid into a table with a named relationship owner, a channel, and a frequency for every row. The last step is the one that makes it a plan.
What is a stakeholder engagement assessment matrix?
It is a table that scores each stakeholder's current engagement level against the level the project needs, using five levels: unaware, resistant, neutral, supportive, and leading. The gaps between the current and desired columns become the engagement plan. It works because it forces you to admit that some stakeholders only need to reach neutral, so effort goes where it changes an outcome.
What is the difference between a stakeholder map and a stakeholder register?
A stakeholder register is a formal list: names, roles, contact details, and categories, maintained for the record. A stakeholder map is a judgment about relative influence and interest, made to guide where you spend attention. The register changes when people join or leave. The map changes when power shifts, which happens far more often and catches teams out.
Who should do stakeholder mapping on a project?
The project manager owns it, but should never do it alone. Scores produced by one person reflect who that person happens to have spoken to recently. Do the first pass with the business analyst and the sponsor in the room, and revisit it with the delivery leads once the work is better understood. The sponsor's contribution matters most for executive stakeholders, because they know the politics you cannot see.
How often should you update a stakeholder map?
Update it at every stage gate, and immediately whenever someone on it changes role, leaves, or gains new responsibilities. Calendar based refreshes miss the event that actually invalidates a map, which is a reorganization. On a long program, a quick review at each monthly steering meeting takes ten minutes and catches drift before it turns into a surprise objection.
What are examples of stakeholders in a project?
Typical project stakeholders include the sponsor, the budget holder, end users and the teams whose daily work changes, operational and IT support functions, finance, legal, compliance and internal audit, suppliers and delivery partners, regulators where relevant, and employee representatives where roles are affected. The ones most often missed are second tier suppliers and the teams downstream of a system you are changing.
Can a stakeholder be low power and still matter?
Yes, and treating them as unimportant is a common and expensive mistake. Low power stakeholders with high interest are usually the people who will operate whatever you build, and their cooperation determines whether the benefits actually land after go live. They also talk to each other, so sustained neglect of that quadrant turns into organized resistance surprisingly quickly.
What is the difference between stakeholder management and stakeholder engagement?
Stakeholder management is the older term and implies controlling information flow to keep people onside. Stakeholder engagement implies a two way relationship where stakeholders influence the project as well as receive news from it. The practical difference shows up in consultation: management tells people what has been decided, engagement asks them before the decision is closed.
Last updated August 2026