methodEnglish

Build factors with evidence and traceability

Keep the external observation, supporting record, interpretation, and strategic implication distinct.

EnglishEspaƱolFranƧais

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

Build factors with evidence and traceability

A useful factor is specific enough to test and consequential enough to matter. Give it a short title, then write four distinct parts:

  1. Observation: what external condition or change is being noticed, bounded by place and time.
  2. Evidence: the retrievable source, data, expert input, or user-supplied record supporting or contradicting it.
  3. Interpretation: what the evidence may mean, including alternative readings.
  4. Strategic implication: how the condition changes the decision, objectives, stakeholders, options, timing, or evidence needs.

Link factor evidence rather than pasting unsupported claims into the description. Evidence raises readiness only when it has a substantive title and summary, a retrievable reference, citation, or safe web URL, and a human user supplied or verified state. ā€œUser suppliedā€ means the team provided it; ā€œverifiedā€ requires a human to confirm the retrievable record and its relevance. Blank placeholders and AI drafts never raise evidence readiness. Contradictory evidence should remain visible.

For every factor, add relevant geographies, horizon, stakeholders, objectives, assumptions, dependencies, uncertainty, and recommended response. Link interacting forces to expose second-order effectsā€”for example, a legal privacy requirement can slow technological adoption, change economic cost, and influence social trust.

AI-drafted factors and evidence remain drafts. The assistant must not invent a source, convert an inference into an observation, or mark its own claim verified. If current external facts matter, collect and attach evidence with a relevant date, jurisdiction, and version before relying on them.