Stakeholder Conflict

Overview

The Stakeholder Conflict Framework (T8) is the territory of operations that take a situation involving multiple parties with divergent interests or values and characterize the conflict structure — mapping the parties, surfacing their stakes and salience, and identifying integrative possibilities or fundamental incompatibilities. T8 is descriptive multi-party work; it precedes and grounds the more action-oriented operations of negotiation guidance (T13) and decision-making (T3). Without the conflict structure mapped, those downstream operations run on assumptions about the parties the user has not tested.

The framework currently runs as a single mode at Tier-2 — stakeholder-mapping — with one expansion candidate (conflict-structure, for systemic / wicked / shifting-coalition conflicts) deferred per CR-6. The mode produces the landscape of parties, their stakes named at concrete-interest level (not role-labels), salience assessed multi-dimensionally per Mitchell-Agle-Wood (power, legitimacy, urgency), positioning on Bryson’s power-interest grid, the relationships among parties, and explicit identification of absent or marginalized parties. The inventory deliberately reaches outside the user’s initial frame — the central guard against the central failure mode of stakeholder analysis (mapping what you already know rather than what you do not know).

The framework’s load-bearing intellectual content is the multi-dimensional salience model, the frame-bounded-inventory guard, and the absent-party requirement. The multi-dimensional salience model treats power, legitimacy, and urgency as three independent attributes — a high-power-low-legitimacy party (Dangerous class) is structurally different from a high-legitimacy-low-power party (Dependent class), and the engagement strategy differs. Single-dimensional salience collapses these distinct cases into a single “important” rating that hides the structural difference. The frame-bounded-inventory guard counteracts the failure where stakeholder mapping silently mirrors the user’s existing mental model. The absent-party requirement makes the silent-power-mirroring failure visible — explicit identification breaks the pattern of reproducing existing power asymmetries.

The framework deliberately resists four patterns: frame-bounded inventory, role-as-stake (treating “the customer” as a stake when the actual stake is concrete loss or gain), single-axis salience, and silent power-mirroring.

The framework answers questions like: I need to map out who’s involved here before I can think clearly about what to do. We keep getting blindsided by people we forgot to consider. Who has standing in this and how much?

Systemic context

Stakeholder Conflict is a Tier-3 cognitive territory in the Ora reasoning lattice, paired with three sibling territories at canonical disambiguation points. T8 ↔ T2 (Interest and Power) — power-and-benefit analysis without conflict structure is T2 (Cui Bono / Boundary Critique); conflict-structure work with multi-party landscape is T8. When both fire, T2 typically runs first because interest analysis grounds the descriptive conflict mapping that T8 produces. T8 ↔ T13 (Negotiation) — descriptive mapping is T8; active negotiation guidance is T13. When negotiation guidance is needed but the conflict structure is not yet mapped, T8 runs first; T13 follows. T8 ↔ T3 (Decision-Making) — when the user is the decision-maker and the parties are inputs to the decision, T3 handles it; when the conflict among parties must be characterized before the decision can even be framed, T8 runs first. T8 also has the deferred conflict-structure mode held in reserve for systemic conflicts where the parties themselves are sub-coalitions with shifting alignments — the deferred-mode flag surfaces rather than substituting wicked-problems from T2 (different operations: T2’s wicked-problems mode handles the problem-structure side; the deferred T8 mode would handle the conflict-structure side). The framework is a building block for downstream Decision Clarity Analysis when the stakeholder landscape reveals fundamental value conflicts among parties.

Ora’s output

Example question: Our nonprofit board is debating whether to accept a major gift from a corporate donor whose business practices have drawn public criticism. Some board members say the gift would let us double our programs; others say accepting it would compromise our mission. Can you map the stakeholder landscape so we can have a more grounded conversation?

Ora’s response (running T8 in stakeholder-mapping):

The situation has multiple parties with divergent interests; T8 is the right territory and stakeholder-mapping is the singleton mode at Tier-2. Before mapping, I’ll do the outside-the-frame pass — the parties you’ve named are the board, the corporate donor, and (implicitly) the program beneficiaries. Let me reach outside that frame.

Stakeholder inventory (outside-the-frame pass).

Parties you named or implied: (1) the nonprofit board; (2) the executive director and senior staff; (3) the corporate donor; (4) the program beneficiaries.

Parties surfaced from outside the frame: (5) other current donors (chose your nonprofit partly for mission integrity); (6) potential future donors; (7) peer nonprofits in your sector; (8) the corporate donor’s critics (advocacy groups, journalists, affected communities); (9) program staff and front-line workers; (10) sector funders and intermediaries; (11) the communities the corporate donor’s business practices have affected (often unrepresented in donor-recipient negotiations); (12) board members themselves as individuals.

Twelve identified parties; your initial frame named four. Several of the next eight have been load-bearing in similar past cases.

Stake per party (concrete interests, not role-labels). Other current donors: may withdraw if the gift signals mission drift, stay if it signals capacity expansion. Future donors: entry decision depends on the same signal. Peer nonprofits: share of philanthropic attention shifts; gift may normalize corporate-money-with-tradeoffs sector-wide. Critics: gift either gives the corporate donor a reputational asset they oppose or fails to. Program staff: doubled programs mean more work and possible strain; concrete interest is sustainable growth. Sector funders: evaluation of your strategic judgment shifts; future grant relationships are affected. Affected communities: depends on whether the gift comes with any commitment to address externalities or is purely transactional. Board members as individuals: each member’s professional and reputational standing is partly tied to the decision; rarely named but operates.

Salience classification (Mitchell-Agle-Wood three-attribute).

Party	Power	Legitimacy	Urgency	Class
Board	High	High	High	Definitive
Corporate donor	High	Medium	High	Dominant (legitimacy contested by external critics)
Executive director	High	High	High	Definitive
Program beneficiaries	Low	High	High	Dependent
Other current donors	Medium	High	Medium	Dominant
Future donors	Low	Medium	Low	Discretionary
Peer nonprofits	Low	Medium	Low	Discretionary
Critics of corporate donor	Medium	Medium	High	Dependent / Dangerous (depending on tactics)
Program staff	Low	High	High	Dependent
Sector funders	High	High	Medium	Dominant
Affected communities	Low	High	High	Dependent
Board members as individuals	High	High	Medium	Dominant

The board, executive director, and program beneficiaries are the highest-salience parties (Definitive or Dependent with all three attributes weighted). Note the Dependent class — program beneficiaries, program staff, affected communities — carries high legitimacy and urgency but low power; the framework’s discipline names these parties even though their structural position would otherwise let them be omitted from the conversation.

Power-interest grid positioning (Bryson). Players: board, executive director, corporate donor. Subjects: program beneficiaries, program staff, affected communities, mission-driven current donors. Context-setters: sector funders, peer nonprofit boards, the donor’s critics if mobilized. Crowd: general public.

Relationships among parties. The corporate donor and affected communities are in active conflict; the board and executive director are typically aligned but may diverge here; current donors and the corporate donor are in a reputational zero-sum game on the gift’s symbolism; the donor’s critics and affected communities are usually allied; program beneficiaries and staff are aligned on sustainable growth.

Absent or marginalized parties (explicit). The communities affected by the corporate donor’s business practices are the most marginalized party in this conversation. They have legitimacy and urgency but no power in the donor-recipient negotiation. The framework names them explicitly so the board’s deliberation includes their stake even though they have no representation at the table.

Confidence per finding. High on inventory and stake-naming. Medium-high on salience classification (Mitchell-Agle-Wood attributes inferred from typical positions; specific salience may vary by history). Relationships depend on local detail; what I named are typical patterns.

The conflict structure is now mapped. The board can have a grounded conversation — not about whether to accept the gift in the abstract, but about which parties’ stakes the decision honors and which it subordinates. DCA is the next step if the board determines the conflict involves fundamental value tensions; T13 (negotiation) is the next step if the board wants to negotiate terms with the donor.

Commercial AI comparison

Comparison content auto-populates when the comparison-refresh framework runs against this question. Drafters do not author this section.

Brief comparison commentary

Auto-populates with the comparison content above.

How to use this framework

You can run the Stakeholder Conflict pattern with any AI of your choice. The composition is single-pass through the stakeholder-mapping mode.

The prompt:

[Paste the framework specification.]

Run stakeholder-mapping on this situation.

Situation: [Plain prose description of the multi-party situation.]

Parties you’ve already named (optional): [If you have a starting list, share it — the AI will reach outside this frame, but the named list speeds up the inventory pass.]

Your role (optional, if you are a stakeholder): [Declared up front so the AI doesn’t accidentally treat your perspective as the neutral one.]

The AI runs the inventory pass (with the explicit outside-the-frame requirement), names stakes at concrete-interest level, classifies salience on all three Mitchell-Agle-Wood dimensions, positions parties on Bryson’s grid, surfaces relationships among parties, and identifies absent or marginalized parties explicitly. The output is matrix-friendly with seven required sections.

For best results:

Provide your role if you’re a party. “We are the board” or “I am a beneficiary” or “I’m an external journalist trying to understand this” — the framework treats your perspective as one of many, not as the neutral observer’s.
Don’t pre-filter the inventory. When the framework surfaces parties you hadn’t named, resist the urge to dismiss them as irrelevant. The frame-bounded-inventory guard exists because the parties one has not named are exactly the ones who tend to appear later as surprises.
Take the absent-parties section seriously. When the framework names a marginalized party, that is the moment the analysis is doing its load-bearing work — surfacing a stakeholder the conventional analysis would have omitted.
Use the output as input, not as decision. Stakeholder mapping is descriptive, not prescriptive. The map shows you who is in play and at what salience; it does not tell you what to do. For action guidance, the output hands off to T13 (negotiation), T3 (decision-making), or DCA (when value conflicts among stakeholders are fundamental).

The framework is deliberately tool-agnostic. The Mitchell-Agle-Wood classification, the Bryson grid, and the outside-the-frame discipline are conceptual disciplines that survive the lift to any environment. The output is a structured matrix with prose explanation.

Other examples

A municipal land-use decision — A planning director’s rezoning case. Outside-the-frame pass surfaces the school district whose enrollment projections the rezoning shifts; the transportation department; environmental-justice advocates representing downwind communities; commercial property owners; and undocumented residents in nearby buildings who fear displacement and have no formal representation. Mitchell-Agle-Wood places the undocumented residents in the Dependent class — named explicitly even though they will not appear at the public hearing. Demonstrates the absent-party requirement on a procedural-democracy decision.
A multi-team product roadmap conflict — Three engineering teams with conflicting feature dependencies. Each team is a stakeholder with concrete interests (team A’s velocity, team B’s reliability, team C’s customer commitments). Outside-the-frame surfaces customers tied to team C’s commitments, on-call engineers bearing team B’s reliability cost, and the executive whose OKR is at risk if team A slips. The dispute repositions as a five-or-six-party negotiation rather than a three-team trade-off. Demonstrates concrete-interests-vs-role-labels in an internal organizational case.
A regulatory-engagement landscape — A startup’s compliance preparation. Outside-the-frame surfaces industry associations whose positions may diverge from the startup’s; peer startups; congressional/parliamentary staff; end users; the regulator’s own internal divisions; media reporters; investors; previous-administration holdover staff. Mitchell-Agle-Wood surfaces the regulator’s internal divisions as a higher-salience subset than the startup had treated as monolithic. Demonstrates the framework’s value where missing a stakeholder class is expensive.

Citations

The Stakeholder Conflict Framework draws on the foundational stakeholder-theory literature. Freeman’s Strategic Management: A Stakeholder Approach (1984) re-cast organizational analysis as multi-party rather than shareholder-only. Bryson’s Strategic Planning for Public and Nonprofit Organizations (1995/2018) provides the power-interest grid. Mitchell, Agle, and Wood’s “Toward a Theory of Stakeholder Identification and Salience” (Academy of Management Review, 1997) provides the three-attribute salience model (power, legitimacy, urgency) and the eight-class taxonomy — the framework’s central methodological commitment, because single-dimensional salience collapses structurally distinct cases. Phillips’s Stakeholder Theory and Organizational Ethics (2003) and Eden and Ackermann’s Making Strategy (1998) extend the apparatus.

The outside-the-frame discipline draws on Reed et al.’s 2009 typology (Journal of Environmental Management) which surveys recurring failure modes — most commonly mapping the user’s mental model rather than the actual stakeholder landscape. The absent-party requirement draws on Ulrich’s critical-systems-heuristics and environmental-justice scholarship documenting how stakeholder analyses in regulatory and planning contexts systematically omit marginalized parties whose stakes are highest precisely because they have no formal standing.

The framework is single-author and originated 2026-05-01 from the T8 territory consolidation. Population: one resident mode (stakeholder-mapping); one expansion candidate (conflict-structure) deferred per CR-6.

Downloads

Framework specification (PDF) — link to ora-ai.org canonical artifact when published
Framework specification (plain text) — link to ora-ai.org canonical artifact when published
Full white paper (PDF) — link when published