CORE GUIDE
Research to outline
Research-to-outline turns collected evidence into a structure of claims, questions and sections so long-form generation follows what the evidence can support instead of inventing a narrative first and searching for citations afterward.
Mental model
An outline is a claim architecture. Each section should have a purpose, supporting evidence and unresolved questions; prose comes later, after the evidence structure is coherent enough to guide drafting.
Why it matters
AI makes drafting cheap, which tempts writers to generate a polished structure before research has revealed what is true or important. Evidence-first outlining keeps the narrative flexible, exposes gaps early and reduces the risk of forcing sources into a preselected story.
01
Build the outline from claim and evidence clusters
Group research notes by the decisions or claims they can support, preserve disagreements, and identify which clusters deserve a section. For each section, record its question, provisional claim, strongest evidence and missing evidence. Order sections by reader dependency and argument flow, then allow the structure to change when new sources weaken or split a claim.
02
Example: a market report is outlined before interviews are read
An author asks an AI to create ten chapters from the project brief, then tries to fit interview evidence into those headings. Several findings contradict the original story. Rebuilding the outline around evidence clusters produces fewer sections, adds a disagreement section and marks one claim as unresolved instead of hiding the contradiction.
Common failure modes
- Creating the final outline before reviewing the evidence landscape.
- Using topic labels as sections without defining the claim or question each one resolves.
- Discarding contradictory evidence because it disrupts a cleaner narrative.
Engineering heuristics
- Give every section an evidence-backed purpose before drafting prose.
- Keep unresolved questions visible in the outline instead of filling them with generated text.
- Reorder or split sections when evidence structure changes.
Takeaways
- 01A strong outline is a map of supported claims, not a table of contents generated from keywords.
- 02Evidence should shape narrative structure before prose generation.
- 03Visible gaps in an outline are cheaper to fix than unsupported claims in a finished draft.
Reading evidence
This records actions you actually took; it does not claim mastery, proficiency, or certification.
Used in
This Concept is reused across these canonical learning paths.
Related concepts from the Knowledge Graph
These relationships come from the canonical graph, not a separate Guide taxonomy.