methodEnglish

Stakeholder, Participation & Ecosystem Mapping Workbench

Make affected people, power, relationships, missing voices, participation choices, and accountability visible before another method relies on the map.

EnglishEspaƱolFranƧais

Documentation follows the current system language. Missing translations fall back to the page's canonical language.

Stakeholder, Participation & Ecosystem Mapping

This workbench helps a group ask: who affects or is affected by this work, how are actors connected, whose knowledge and voice are present or missing, and what participation is honest and proportionate?

It is not a directory, a popularity ranking, or an objective map of human importance. Every map is produced from a declared perspective, at a point in time, with incomplete evidence. Formal influence, affectedness, knowledge, vulnerability, interest, legitimacy claims, and willingness to participate remain separate judgments.

Perspective-to-participation cycle

The method moves from a declared viewpoint to a reviewed participation design; the map remains a revisable snapshot.

What the workbench produces

  • a traceable actor register that separates voice mode, actor-level voice status, consent, and shareability;
  • directed or reciprocal relationships for influence, accountability, decision authority, dependency, resource flow, information flow, service delivery, collaboration, conflict or tension, trust, regulation, and other explicitly described links;
  • separate influence, affectedness and interest ratings; stance, confidence, needs, concerns, access requirements and voice attribution remain distinct;
  • participation plans with decision rights, channels, accessibility, safeguards, owners, date windows, status and feedback commitments;
  • missing-voice, concentration-of-power, conflict, burden, privacy, and safeguarding gaps;
  • content-bound human review, limitations, reservations, authorization, next-review timing, and exact public-disclosure confirmation;
  • a structured handoff for another Albarena method;
  • a comprehensive web report and branded PDF with machine-readable data.

Human governance

AI may suggest candidate actors, relationship questions, accessibility checks, or gaps. It must not decide who is legitimate, claim representation, infer consent from silence, disclose sensitive identities, assign a personā€™s stance as fact, or mark engagement complete. Humans review every consequential judgment. Assistant context is minimized by default: restricted identities, contact and access details, private voice text, reservations, reviewer identity, and raw evidence are not sent as the active snapshot.

The provider-bound active snapshot contains only non-identifying labels or opaque references, bounded analytical fields, counts, and relationship structure. It is not a safe channel for manually pasting sensitive personal information. Imported or AI-origin actors, assessments, voices, relationships, participation plans, and safeguards retain their origin and require a named human reviewer plus a review timestamp before publication.

Public HTML, PDF, readable briefs, and JSON use a separate privacy-minimized projection. A named human must confirm a fingerprint of the exact final title, summary, resolved selection, narrative lists, and public identity/consent inventory. Until that confirmation is current, public output is blocked. The projected model emits only approved public labels and confirmed public narrative; public reports may add aggregate readiness, authorization state, and finding counts. Ratings, relationships, assessments, voice records, plans, safeguards, evidence, contacts, source lineage, internal identifiers, reviewer identity, and detailed findings remain withheld.

Continue with Workflow and imports.