Chapter 9: Map the Stakeholder System

Six publics, one room

The forum is held in the main market hall in month 14, two weeks after construction reached the market end of the corridor. Lena Voss and Aisha Bello have arranged the chairs in a square, and the square does not hold. A fabric trader named Yusuf speaks first, and the room turns out to contain six different projects.

“You say the corridor will bring customers,” Yusuf says. “It will bring them in eighteen months. My stall dies in three. Nobody here has told me where I trade while the works eat the street in front of me.”

A woman from the estate behind the station site interrupts. “And the buses on our street at five in the morning, because the diversion counts say we take fifty-five percent of the traffic, will shake the plaster off my walls. The last time the city promised us something, it cancelled.”

Grace Njoroge, who chairs the city’s disability advocacy federation, holds up the render of the new station. “This drawing has stairs at every entrance. You have drawn a station I cannot use, and you show it to me as good news.”

Ibrahim, the transit operators’ union representative, is calmer. “The drivers will learn the new depot two weeks before opening. That is how strikes start, and it is also how passengers get stranded. We are not your enemy. We are telling you what happens to a plan that does not include the people who run it.”

Councillor Diana Kamau has been quiet. When she speaks, it is to Lena, not to the room. “You keep saying the community is engaged. These are my constituents. You have been talking to the city, and I have been listening to them. The slide says community engagement is on track. On track for whom?”

Lena’s slide really does say that. Behind that scene sits the problem this chapter exists to solve. A stakeholder map that names “the community” as a single box with a favorable rating is not a map of the human system around the work; it is a map of the project’s own hope. The people in this room are not one stakeholder. They are a trader whose income has a deadline, a resident whose walls and memory carry the last broken promise, an advocate whose rights the design forgot, a union whose members will run the system, and an elected politician whose political survival depends on all of them. Their interests overlap, diverge, and shift as the project changes the world around it.

Chapter 8 ended with the governance seats: the named people who decide. The people those decisions land on are a far larger system, and the decisions will only survive contact with them if someone can see the system as it actually is: who is affected, what they want and fear, who holds power and who holds none, whom they trust, and which channels carry news. That seeing is the Alignment term of the Mastery equation: Project Mastery = Judgment × Alignment × Delivery × Learning. Alignment is not a slogan; it is a map with the right people on it, kept current. The sections that follow are how the map gets drawn: the boundary, the interests, the attributes, the network, the ledger, and the depth of engagement each need deserves.

The boundary question

Every stakeholder map begins with a boundary decision: who is inside it. The boundary is a judgment call, and it deserves an argument, because a map is only as good as the people it includes, and the people it excludes are the ones most likely to ambush the project later. The concept that anchors the practice comes from R. Edward Freeman’s 1984 book, which popularized the stakeholder idea in management. A stakeholder is anyone who can influence the work or is influenced by it. That definition is deliberately wide, and the boundary question is where its width becomes workable.

Three tests separate the groups worth mapping from the noise. Who can affect the work: funders, regulators, sponsors, unions, suppliers, officials whose approval gates the plan. Who is affected by it: users, neighbors, traders, patients, commuters, the staff who will run the result, the people whose street the traffic lands on. Who would hold the project accountable if harmed: the groups with legal standing, moral claims, or enough political weight to make harm expensive. A stakeholder belongs on the map when it answers at least one of the three.

The distinctions within the boundary matter more than the umbrella term. An affected party is touched by the work without necessarily influencing it; a resident whose street takes 55 percent of the diverted traffic may have no vote and no way to stop the project, and the map must carry her anyway. A stakeholder is a party with standing to influence or be influenced, which includes the affected party and adds the ones with leverage. A rights holder is a party with a legal or moral claim the project may not violate: workers’ safety and pay, residents’ quiet enjoyment, patients’ privacy. The strongest case is indigenous peoples’ rights over lands and territories. There, the UN Declaration on the Rights of Indigenous Peoples (2007) requires states to consult and cooperate in good faith with the peoples concerned, to obtain their free, prior and informed consent before measures or projects that affect them. Rights holders do not negotiate from a position of mere interest; the project carries an obligation toward them, and mapping them is how the obligation gets a face and a date.

The category that reveals the quality of a map is the silent stakeholder: the parties who are affected but have no voice, no seat, and no attendance record. The night-shift worker who cannot leave work for an evening forum, the resident who does not read the city’s website, the future residents who will live with the corridor’s design choices for forty years, the environment that carries the drainage and emissions, the non-citizens who work in the market but cannot vote. They do not appear in attendance lists or satisfaction surveys, and they are exactly the ones the map exists to force into view. A map that contains only the people who attend meetings has quietly adopted the loudest possible definition of who counts.

The tell of a bad boundary: the stakeholder list reads like a distribution list of the people the project already talks to. Sponsors, funders, regulators, the council, the operator, the contractor. Every name has an email address on the project’s mailing system. Chapter 3 showed the boundary failure at BlueLine: the model measured everything inside the construction boundary and nothing outside it, where the traffic, the income, and the trust actually landed. The stakeholder map has the same failure mode, and the fix is the same: argue about the boundary before you draw the boxes, and let the argument include the people the first draft left out.

Positions are answers; interests are questions

Once the boundary is set, the next discipline is separating what people say from why they say it. The distinction has been a working tool of negotiation since Roger Fisher and William Ury’s Getting to Yes described it in 1981: people come to the table with positions, the specific things they demand, and beneath every position sit interests, the needs, fears, and values the position exists to serve. Positions are answers. Interests are questions. A project that negotiates positions trades at the surface and is perpetually surprised when agreements collapse; a project that maps interests can design options the surface never offered.

Yusuf’s position is “my stall stays where it is.” His interests are narrower and far more addressable: his family’s income survives the construction period, his regular customers can still find him, and he has some say in when and how the move happens. “No buses past the school before eight” is a position; “my children cross this street” is the interest, and it can be met by a crossing guard, a signal, a lower speed limit, or a bus schedule, none of which is in the original demand. The resident’s position rests on interests of sleep, safety, and a city that keeps promises; the last is the one the map must not miss, because no technical fix will satisfy her until the trust is addressed.

The working instrument is the question ladder: take the stated position and ask “why does that matter to you?” five times, each answer becoming the next question. “Keep my stall.” Why? “The stall feeds my family.” Why does that matter now? “Construction lasts eighteen months and I have no savings.” Why? “The market’s footfall dies with the first fence.” Why? “Nobody has told us how customers will reach the stalls that stay.” The fifth answer is the interest, and it names the design problem the project can actually work on: visibility and access during construction, not the relocation itself. The ladder has a boundary: past a certain depth the questions stop being about the project, and the project’s role is not amateur therapy. Go deep enough to find the interest the project can act on, and record what remains as a constraint rather than a failure.

Interests have a property that makes the map perishable: they change as the project changes. The trader who opposes the corridor because his stall will be buried wants its customers when it opens; the resident who wants a shorter commute also wants a quiet street, and the balance shifts as the works move closer to her door. This is why the map is a living artifact: every material change to the project changes somebody’s interests, and the map must be re-read at the same cadence as the risk register, not archived with the charter.

Power, legitimacy, urgency, and what the grid misses

The third discipline is seeing why a stakeholder matters, which decides how much attention and engagement they deserve. The classic account, from Ronald Mitchell, Bradley Agle, and Donna Wood’s 1997 theory of stakeholder salience, names three attributes. Power is the ability to get the project to do something it would not otherwise do. Its deep source is control over resources the project needs: money, approvals, votes, labor, information, or the ability to block. Jeffrey Pfeffer and Gerald Salancik’s 1978 work on organizations’ dependence on their environments made the underlying logic plain: whoever controls a critical resource controls the terms. The funder that pays on physical progress, the regulator that holds the environmental approval, the union whose members will run the depot, the councillor whose constituents decide the next election: each holds a different resource. Power follows the resource, not the title.

Legitimacy is a recognized claim: the standing to be taken seriously because the party has a lawful right, a mandate, a constituency, an expertise, or a history of being affected. Urgency is the degree to which the claim presses now: time-sensitive and critical to the claimant. Salience grows with the attributes held; a party with all three cannot be safely ignored, while a party with one is easily deferred. The model’s own contribution is the uncomfortable cases. A party with power and urgency but little legitimacy: a politically connected contractor, a media outlet with a vendetta, a lobby with money and no mandate. Ignoring them is not a moral strategy, it is a plan for an ambush; the response is to engage without legitimizing, to protect the process rather than to grant the claim. A party with legitimacy and urgency but no power: the residents, the disabled users, the future tenants. They have the claim and the deadline and none of the leverage, and they are the ones the project must protect, because the protection is the project’s own integrity.

Two further attributes belong on the map because they decide engagement in practice, and they are this book’s additions to the model rather than its vocabulary. Proximity is how near the party stands to the work: the shop at the station entrance is closer than the shop three streets away. Impact is how much the work changes the party’s circumstances: a station relocation changes the trader’s income dramatically, the councillor’s political exposure, barely the commuter’s life. A party can be proximate without being impacted and vice versa, and the engagement plan should follow the combination.

The attributes are also not fixed. Legitimacy can be manufactured or eroded; power can be assembled. The petition is the cleanest example. In month 8, a petition with 2,300 signatures reached the mayor’s office; by the framing session a week later it carried 3,100. Each signature converted affectedness into a claim, the claim into legitimacy, and the legitimacy, aimed at an election-minded council, into power. The project that maps attributes once and freezes them will be surprised by the petition; the project that re-reads attributes monthly will see it coming in the signature count, the councillor’s changing language, and the market association’s sudden willingness to talk.

The grid is a start; the network is the system

Most engagement planning begins with a power-interest grid, the familiar two-by-two that sorts stakeholders by how much power they hold and how interested they are, usually attributed to Aubrey Mendelow’s 1981 work on environmental scanning. As a starting artifact it has real value: it forces the question of who holds leverage and who cares, and it sorts engagement effort into the classic bands, manage closely, keep satisfied, keep informed, monitor. Used alone, it has three limits: it treats stakeholders as independent boxes with fixed attributes, when their power is a function of their relationships; it freezes them at one point in time; and it cannot show the thing that matters most in practice, how influence travels between the boxes.

The system map for stakeholders is a network, and the network is this chapter’s primary picture. The nodes are the parties, including the silent ones; the ties are the working relationships between them, who talks to whom, who trusts whom, who carries news. The clusters are the dense groups, the traders, the residents, the advocates, that mostly talk to themselves; the brokers are the people whose ties cross cluster boundaries, the association chair, the liaison with field contacts across the estates, the councillor who stands where constituents and the city meet. The weak ties are the thin connections between clusters, and they deserve their own name because they do most of the information work. Mark Granovetter showed in his 1973 study that ties bridging otherwise separate groups carry novel information: the news that does not circulate inside a cluster. Strong ties inside a cluster mostly confirm what everyone already knows. For a project, the weak ties are the early-warning system: the council officer who mentions the petition before it reaches the mayor, the estate rep who hears about the 5 a.m. bus trial before the notice goes out. A map with no weak ties is a map that will be surprised.

Figure 9.1: The BlueLine influence network. Three clusters of
interest, brokers who carry news between them, and affected groups
with no one speaking for them. Solid lines are active working ties;
dashed lines are weak ties that carry novel information between
clusters. The nodes on the far right are affected and connected to
no one: the map's warning lights.

  TRADERS AT THE MARKET              RESIDENTS AND ESTATES
  Yusuf; 1,400 shops                 estate representatives;
  three markets, one school          residents of the diverted streets
        |  solid                           |  solid
        v                                  v
  [MARKET ASSOCIATION]  - - weak - - [RESIDENTS' NETWORK]
  chair: Halima Bakari                 convenor: Mwende Chebet
        |  solid                           |  solid
        v                                  v
  [COUNCILLOR KAMAU] ----------- [AISHA BELLO] -----------
        |  solid                      |                  |
        v                             v                  v
  [MAYOR'S OFFICE]        [LENA VOSS AND THE CONSORTIUM] [TRANSIT OPERATOR]
                                      |                  (wants ridership,
                                      |  solid            no seat)
                                      v                  ^
                              [DISABILITY ADVOCATES] - - -+
                              Grace Njoroge, federation

  [ENVIRONMENT; FUTURE RESIDENTS; SHIFT WORKERS WHO CANNOT ATTEND]
  affected, connected to no one: the map's warning lights

Reading the network answers questions the grid cannot. Where does information travel, and who slows it? The market association is the traders’ broker, so engagement aimed at individual traders will not reach the cluster as a group; the association is the channel, and bypassing it will be read as an insult. Who is connected but underserved? The transit operator holds a weak tie and no seat at the steering table, yet its staff will run the system; the map shows the gap between stake and channel. Who is affected and unconnected? The night-shift workers, the future residents, the environment: no broker, no tie, no voice, which is why the map places them at the edge as warnings rather than inside any cluster. And where could coalitions form? The map is also the coalition map. The residents’ network and the disability advocates share an interest in station access and street safety, so one well-designed access commitment could satisfy two clusters at once; a poor one would hand them a shared grievance. The coalition map is not a conspiracy chart; it predicts where interests will align and where the project can design for shared gain rather than defend against shared loss.

Building the network is fieldwork, not deskwork. Name the nodes from the boundary exercise. Then ask, in short conversations with people who know the ground, who talks to whom, who is trusted, who carries news between groups, and who is affected but not connected. Observation verifies the answers: whose invitations are actually accepted, whose name comes up in other people’s sentences. The map’s shape should be challenged, not admired: a broker no one on the team can verify, a cluster drawn from one person’s account, a silent node that turns out to be represented. The network is a hypothesis about how influence moves, and it earns its place by being tested against events and revised when events disagree.

Neutrality is a position; ambivalence is not hypocrisy

With the network in view, the next discipline is naming each stakeholder’s stance, and the naming must be finer than the familiar binary. Support is declared and observable: the councillor’s public advocacy, the association’s cooperation on the phasing trials. Resistance is declared or observable opposition: the petition, the union’s warning about the depot, the advocates’ refusal to bless the stairs. Neutrality looks like neither, and the map must treat it as a stance rather than an absence: “no objection” frequently means nobody asked, and “we have no position” means it has not been tested because nothing has touched them yet. A neutral stakeholder is an open question waiting to be asked, and the engagement plan should say who asks and when.

Ambivalence is the category most often misread as hypocrisy, and it deserves the map’s respect. Ambivalent stakeholders hold genuinely conflicting legitimate interests: the trader who wants the corridor’s customers and needs his stall to survive the construction, the resident who wants the shorter commute and the quiet street, the councillor who wants the flagship delivered and her constituents’ trust. The project that names them “flip-floppers” has stopped thinking. The project that maps the two interests separately can design for both: stage the relocation so the stall survives, route the buses so the street keeps its sleep, sequence the works so the councillor can deliver wins before she must ask for patience. The discipline is per-issue stance, not per-person verdicts: the same trader supports the corridor and resists the relocation, and both rows belong on the map.

The deepest rule in this territory: resistance is information before it is an obstacle. Nobody resists a project they believe helps them; the resistance is a report about what the project is doing to something they value, or about what history taught them to expect. The residents’ resistance was not to buses; it was to a city that cancelled the last promise, and no amount of bus engineering fixes a cancelled promise. The union’s warning was not obstruction; it was operational knowledge about how drivers learn, delivered before the incident, the cheapest form of learning a project can buy. The mapping question is never “who is with us,” which flatters the project, but “what would have to be true for each stance to shift,” which disciplines it. A map that cannot answer that question has drawn faces without reading them.

The trust ledger

Every engagement strategy runs on trust, and trust is not an attitude, it is a history. The working definition that has held since Robert Mayer, James Davis, and F. David Schoorman’s 1995 integrative model: trust is the willingness to make oneself vulnerable to another party. That willingness rests on expectations of the other’s ability, benevolence, and integrity. Ability is the belief that they can do what they claim; benevolence, that they will consider the other’s interests; integrity, that they mean what they say. The three dimensions explain failures that power analysis cannot: the project may have the ability (competent engineers), and the benevolence (genuine care), and still fail on integrity, because the last promise was cancelled, and cancelled promises are the currency of a ledger in deficit.

The trust ledger is the map’s third register, beside the network and the stance notes: for each key stakeholder, what they have seen the project do, what the project promised, and what the project owes. The entries are dated, and they accumulate. The residents’ ledger carried a deficit before the project began: the prior cancellation. Every credible commitment kept adds a credit, and the first credits are the expensive ones, because a ledger in deficit charges interest. The traders’ ledger is being written now: the relocation was approved conditionally, and every week the phasing plan stays unsigned is a debit with the association. The advocates’ ledger began with the stairs render, a debit handed to them as a gift. The ledger explains the behavior the grid cannot: why the powerful supporter stalls in private and why the same promise means different things to different groups. The field signal for a trust deficit hiding behind a supportive public face: the support continues in speeches while the cooperation stalls in practice, and every follow-up is met with warmth and no date.

Two registers of trust coexist, and both belong in the ledger. Institutional trust attaches to the organizations: the city, the consortium, the funder, and it is the register the prior cancellation poisoned. Personal trust attaches to named people: Aisha Bello’s field relationships, the association chair’s standing with her members. A project that cannot borrow institutional trust can still earn personal trust through the people the stakeholders already trust. The liaison who spans the clusters is not a nice-to-have; she is the project’s credit line, and the map should show which ties run through named people rather than letterheads.

Engagement by need, not by title

The map exists to decide engagement, and engagement depth should be chosen by what the stakeholder needs from the project and what the project needs from them, never by their rank. The scale that has organized this thinking since Sherry Arnstein’s 1969 ladder of citizen participation runs from information at the shallow end to real power at the deep end: inform, consult, involve, partner, delegate. Arnstein’s warning remains the sharpest lesson in the field: the middle rungs, informing and consultation, become tokenism when they carry no influence. The point is not that every project must climb to the top. It is that a project must be honest about which rung it occupies, and must not sell a lower rung as a higher one. Telling people about a decision after it is made and calling it “engagement” is not a communication failure; it is a betrayal of the word.

Choosing the depth is a decision with four inputs. Who must decide? Those stakeholders get partnership or delegation: the association co-designing the phasing plan, the advocates co-authoring the access standard. Who must change behavior for the value to exist? They get involvement, early and continuous: the transit operators who will run the depot, the merchants who must stay open through construction. Who is affected and wants a voice? They get consultation with a visible response. For the residents and the diversion mitigations, the consultation must close the loop, publishing what changed and why, because a consultation that absorbs and ignores manufactures cynicism. Who must not be surprised? They get information, timed to their need: the funder, the neighboring jurisdictions, the mayor’s office, informed before the public is, because surprise is how allies become liabilities. The four inputs produce four depths, and the map records each depth with the evidence that the engagement worked, because engagement measured by attendance or satisfaction is engagement measured by the wrong number.

The ethical boundaries deserve their own paragraph, because influence is a craft and its misuse is a professional failure. Never promise a seat that will not be given: the deepest damage a project can do to its own map is to invite people to a table that has no power, because the discovery of the empty table erases every credit the ledger had. Engagement is not information extraction: mining stakeholders for data and giving nothing back is exploitation. Presenting a finished decision as an open question is manipulation, and it reads as manipulation. The strongest boundary is the rights floor: free, prior and informed consent is not a consultation technique, it is a right. And the audit rule: the engagement record must show what changed because of engagement, and a stakeholder who cannot point to a single change they caused has been processed, not engaged.

The engagement assessment at BlueLine takes the shape of a compact table. The table is the map’s decision surface: the place where network, stances, and depths become one artifact.

The map that earns its keep

The worked application is the forum itself, read back through the map. The evening looked like noise: six publics, one room, six grievances. Read through the network, it was a set of design inputs, and the map changed three decisions in the weeks after.

The relocation. The petition had already forced the conditional acceptance at the gate, and the engineers’ preference was to move the stalls in one operation, fast and complete. The map said the association was the traders’ broker, that the traders’ core interest was income continuity rather than location loyalty, and that the phasing plan was the actual object of negotiation. The redesigned approach staged the relocation in three moves keyed to the construction sequence, each planned with the association, with the footfall-mitigation measures agreed before the first fence: wayfinding to the stalls that stayed, the market’s busiest entrances kept open. The association’s chair carried the plan to her members rather than the project’s liaison carrying it from outside. The decision that changed was not whether to relocate; it was who designed the how, and the answer moved from the engineer’s office to the co-design table.

The access standard. Grace Njoroge’s federation had been treated as a consultation audience: shown renders, invited to comment. The map showed the advocates as a legitimate, urgent, right-adjacent cluster with a federation structure that gave them a broker, and the engagement depth moved from consult to partner. The access standard was co-authored before the station went to tender, and the stairs disappeared from the render at the design stage, where the fix costs a fraction of what the same fix costs after the station is contracted. The project cannot prove the counterfactual, but the comparison is not subtle: an access failure found by the people who will use the building, in the drawing phase, is the cheapest defect a project ever fixes.

The silent stakeholders. The night-shift workers did not attend the forum; the map’s warning nodes flagged them when the 5 a.m. bus trial was planned, and the engagement design changed accordingly: not another evening forum, which would have excluded them again, but the union and the estate representatives, their actual brokers, reached at the times and places where night-shift workers exist. The trial went out through Ibrahim’s noticeboards and Mwende Chebet’s estate network before the public notice, and the objections arrived before the trial instead of after it, which is the difference between consultation and ambush.

One number carries the whole lesson. The petition did not grow from 2,300 to 3,100 signatures because the project communicated badly. It grew because 3,100 households each received a private, credible report from their own networks about what the corridor was doing to their streets and their livelihoods, and the project’s map did not contain those networks. The petition is the price of a map that ends at the distribution list. The corrected map does not guarantee harmony; it guarantees that the next report of this kind arrives through the project’s own channels first, with enough lead time to act on it. That is the difference between a stakeholder system that is managed and one that is known.

The map as theater

The failure patterns deserve to be named, because competent people produce every one of them in good faith.

The poster map. A stakeholder map is drawn beautifully at kickoff, printed, and never touched again. The tell: the same map appears in every steering pack for a year, engagement decisions are justified by reference to it, and nobody in the room can say who owns it or when it was last changed. It fails twice at once: it freezes the parties at the moment it was drawn, and it is consulted after the work has changed everything, the relocation changing the traders’ interests, the opening the commuters’, the policy shift the regulator’s. A map that is not updated is not a map of the stakeholder system; it is a decoration of the project’s initial guess.

The map of the loud. It is built from the people who attend meetings, join the working groups, and reply to the emails, and it is accurate about exactly those people. The tell is the one the boundary question already named: the list matches the project’s distribution list, and the affected who cannot attend, who do not read, who cannot leave work, are absent, and their absence is the map’s quietest lie.

The consultant’s map. An outside firm produces a polished analysis, presents it once, and leaves; the team that must live with it recognizes none of its detail and owns none of its decisions. The tell: the names on the map are not names anyone on the team can place in a real relationship. A map not built by the people who must use it is a report about someone else’s system.

The classified map. It lives in the project leader’s head, drawn from private conversations and kept from the team and the sponsor, so two team members can make contradictory promises to the same stakeholder in the same week. The tell: the contradiction surfaces in the field as a surprise. A map that cannot be seen cannot be disagreed with, and a map that cannot be disagreed with cannot be corrected.

Different cadences, the same practice

The practice does not change with the delivery approach; the cadence does. Predictive work refreshes the map at the gates, with a formal engagement plan and evidence that each depth was delivered, because long horizons and the high cost of change make the map a control artifact. Adaptive work refreshes it continuously, because every iteration and every release is a stakeholder event: the release itself changes interests, and the retrospective should include a map check as naturally as it includes a quality check. Hybrid work, which is BlueLine’s shape, refreshes it at the seams: formal engagement for the construction and regulatory work, continuous engagement for the co-design and the trials, with the map reviewed wherever the two cadences meet. The owner differs too: a named engagement lead or the project leader holds the map, never “everyone,” but the map must be visible to the whole team, because a map the field workers cannot see will be contradicted by the field.

What the machine can draft, and what it cannot, deserves its own honesty. An assistant can produce a first catalogue from project documents, propose interview question banks, and cluster forum transcripts into recurring themes; those drafts are genuinely useful. What the machine cannot do is the map’s core work: it cannot sense who is silent, cannot weigh trust, cannot read which broker is trusted and which is bypassed. It will cheerfully invent plausible stakeholders, because a fabricated stakeholder is indistinguishable from a real one in fluent text. The map is a political artifact before it is a document, and political artifacts require human judgment to draw and human names to sign. The boundaries hold here as elsewhere: names, complaint records, and anything regulated or personal stay out of any tool the organization has not approved. Every node, tie, and stance on the machine’s draft is verified against named people who know the ground before it earns a place on the map.

Practice

One. A quick classification. Each of these comes from a real engagement. Is it a position or an interest? (a) “Move the station and we will block the works.” (b) “My stall must survive the construction.” (c) “Buses must not pass the school before eight.” (d) “We want a share of the corridor’s maintenance contracts.” (e) “The night shift cannot lose the car park.”

(a), (c), (d), and (e) are positions: each names a specific demand with a threat or a condition attached. (b) is an interest: it names the outcome that matters, the survival of the stall, without prescribing the mechanism. The drill’s lesson is that (b) is the only one the project can act on directly; the others must each be laddered to the interest beneath them. “No buses past the school before eight” might rest on crossing safety, on sleep, or on the memory of a prior accident, and the engagement that follows is different for each.

Two. A field drill: draw the network. Take an initiative you know and build its influence network from five to eight short conversations with people who know the ground. Name the nodes, ask who talks to whom, who is trusted, who carries news between groups, and who is affected but not connected. Draw the network, mark the brokers and the weak ties, and place the silent stakeholders at the edge. Then check your work: is at least one affected group connected to no one, is at least one broker someone you did not expect, and can you name the one weak tie that would warn you earliest?

The drill succeeds when it produces the two discoveries almost every honest network produces: an affected group with no broker, the map’s warning light, and a broker the team had not recognized, the map’s best channel. If the network contains no weak ties, it has probably been drawn from the team’s own relationships rather than from the field’s; the fix is to ask the stakeholders, not to ask each other. The most valuable output is the early-warning tie, because it converts the map from a description into an instrument.

Three. A decision room: the phasing plan. Month 14 at BlueLine, and the evidence is on the table. The relocation is approved conditionally; the engineers propose moving all stalls at once, which protects the schedule and risks the market; the association proposes three staged moves, which costs about six weeks and some rework of the construction sequence; three traders have already closed; the petition stands at 3,100; the councillor has said publicly that her constituents must not be sacrificed to the date. Decide the phasing, the engagement depth for the association, the residents, and the advocates, and the one thing you would measure to know the decision worked.

Staged relocation with the association co-designing the plan is the defensible answer: it matches the traders’ mapped interest (income continuity, not location loyalty), gives the councillor a deliverable she can carry to her constituents, and converts the association from a channel to be managed into a partner whose members carry the plan themselves. The depth table follows from the map: involve for the association, consult with visible response for the residents, partner for the advocates on the access work. The measure that matters is not satisfaction but closure: the count of trader closures, which the plan must stop growing within one construction phase. Unsafe choices: deciding the phasing without the association, which converts a legitimate broker into an opponent; and promising the residents a seat in the phasing design that will not be given, which adds a credit to the trust ledger’s deficit column. Reasonable but risky: accepting the engineers’ single-move plan with strict conditions and the association’s consent, defensible only if the schedule genuinely outweighs the market risk.

Four. The mastery drill: the mayor who delays. Mayor Freya Asante has championed the corridor in every speech, and the relocation approval has sat on her desk for six weeks. The works are at the market end; the association’s chair has asked twice and been told it is “being finalized”; the councillor’s constituents are watching; the funder’s gate is in eight weeks. Your map shows the mayor with high power, high legitimacy, and a public stance of full support, and the evidence says the support is real and the delay is real. Choose your engagement strategy: brief her office with a decision memo that prices the delay; ask the councillor to carry the message as an intermediary; make the delay public; or wait for the gate. Defend the choice in terms of the map, and name what would change your answer.

The defensible strategy is the private decision brief first: a one-page memo to the mayor’s office stating the decision required (sign the relocation approval or formally defer it), the options, the recommendation, and the cost of waiting, with a date by which the answer is needed. This is the escalation discipline of chapter 8 applied to the stakeholder system: the mayor is the decider, and the delay is a trust-ledger event, a deficit accruing interest with the association and the councillor every week. A private brief respects her public support while making the cost of delay visible and dated. The councillor as intermediary is reasonable, and she is the natural channel if the mayor’s office will not receive the brief, but it should carry the same decision structure, not a message of pressure; the councillor can make the mayor’s ownership of the delay politically attractive rather than costly. Making the delay public is the unsafe choice at this point: the mayor’s public support is an asset, and public pressure converts an ally’s stall into an ally’s resentment. Waiting for the gate is the most expensive choice: it lets the deficit compound to the funder’s gate, where the delay surfaces anyway, at the worst moment, attached to the project’s name. What would change the answer: evidence that the delay is not the mayor’s choice, for example a blocked negotiation inside the coalition, which changes the target of the brief; or evidence of real opposition behind the public support, which changes the diagnosis from “stalled” to “hostile” and shifts the strategy from decision support to alliance building.

Five. The transfer question. In the last project you were part of, who was affected but never in the room, who carried news between the project and the groups it touched, and what did the trust ledger look like before the work began? Which group’s resistance was a report about the project, and what was it reporting?

The durable principle: the stakeholder system is a living network of interests, power, and trust, and mapping it is the practice of keeping that network visible. The practice names the boundary that decides who counts, the interests beneath the positions, the attributes that make a stakeholder matter, the brokers and weak ties through which influence actually moves, the ledger that records trust, and the engagement depth each need deserves. The most common next failure is the one the forum exposed: the map gets drawn and then treated as finished, a poster of the project’s initial guess. The system beneath it keeps moving, and the project discovers the movement in the form of a petition, a strike, or a cancelled approval. The map’s gift is lead time, and lead time is only as good as the next step: converting what the mapped stakeholders need into outcomes, requirements, and scope the project can actually build to. That is the traceability chain of the next chapter, “Turn Needs into Outcomes, Requirements, and Scope.”

Notes

• BlueLine Urban Mobility Program is a composite case created for this book; no real city, corridor, company, or people are depicted, and all named characters are composite. The corridor numbers cited here (the field’s 55/45 diversion split, the petition growing from 2,300 to 3,100 signatures, the 1,400 shops, the prior cancellation) are author-created illustrative figures consistent with the earlier BlueLine chapters.
• The stakeholder concept as used in management traces its mainstream origin to R. Edward Freeman, Strategic Management: A Stakeholder Approach (Boston: Pitman, 1984). The three tests in this chapter (who can affect, who is affected, who would hold the project accountable) are the author’s working version of the “can affect or is affected” idea, not a quotation.
• The positions-versus-interests distinction is taught in Roger Fisher and William Ury, Getting to Yes: Negotiating Agreement Without Giving In (Boston: Houghton Mifflin, 1981; second edition with Bruce Patton, New York: Penguin, 1991). The question ladder is the author’s practical instrument, not the book’s vocabulary.
• The salience model of power, legitimacy, and urgency is from Ronald K. Mitchell, Bradley R. Agle, and Donna J. Wood, “Toward a Theory of Stakeholder Identification and Salience: Defining the Principle of Who and What Really Counts,” Academy of Management Review 22, no. 4 (1997): 853-886. Proximity and impact are this book’s additions for engagement practice, not terms from the model.
• The resource-based view of power follows Jeffrey Pfeffer and Gerald R. Salancik, The External Control of Organizations: A Resource Dependence Perspective (New York: Harper & Row, 1978).
• The power-interest grid is usually attributed to Aubrey Mendelow’s paper “Environmental Scanning: The Impact of the Stakeholder Concept,” Proceedings of the Second International Conference on Information Systems (1981). The chapter’s criticism of the grid and its network replacement are the author’s synthesis.
• The role of weak ties in carrying novel information between groups is from Mark S. Granovetter, “The Strength of Weak Ties,” American Journal of Sociology 78, no. 6 (1973): 1360-1380.
• The ladder of participation is from Sherry R. Arnstein, “A Ladder of Citizen Participation,” Journal of the American Institute of Planners 35, no. 4 (1969): 216-224. The chapter’s four-depth scale (inform, consult, involve, partner, delegate) is the author’s working adaptation; Arnstein’s warning about influence-free consultation becoming tokenism is preserved in the chapter’s own words.
• The integrative model of trust as willingness to be vulnerable based on expectations of ability, benevolence, and integrity is from Roger C. Mayer, James H. Davis, and F. David Schoorman, “An Integrative Model of Organizational Trust,” Academy of Management Review 20, no. 3 (1995): 709-734.
• The rights-holder requirement for free, prior and informed consent follows the UN Declaration on the Rights of Indigenous Peoples, adopted by the UN General Assembly in resolution 61/295 on 13 September 2007, especially Articles 19 and 32; readers should consult the current official text for the precise obligations in their jurisdiction.
• ISO 21502:2020, “Project management: Guidance on project management,” the general ISO guidance, speaks of “interested parties” where this book says “stakeholders”; the terms denote the same working idea. The current PMBOK Guide (Eighth Edition, Project Management Institute, 2025) likewise gives stakeholder engagement sustained attention; readers should consult the official publishers for current editions. This book is independent of ISO, PMI, and all named framework owners.