Skip to content
How to ThinkIn the Age of AI
Rulev2RL-281

After completing a relationship map, write what…

After drawing a complete relationship map, write three to five sentences describing the structural story—focusing specifically on what was invisible before you drew the map rather than summarizing what you already knew.

Why This Is a Rule

A relationship map's value lies in what it reveals that you didn't already know — the emergent structural properties that become visible only when individual relationships are laid out spatially. But without a deliberate extraction step, these insights evaporate: you look at the map, nod, and move on without capturing what the mapping process produced.

The 3-5 sentence extraction focused on "what was invisible before" forces you to distinguish between knowledge you brought to the map (inputs) and insights the map generated (outputs). Summarizing what you already knew is not useful — that's just restating your inputs. Describing what was invisible before captures the map's actual contribution: a bottleneck you didn't recognize, a feedback loop you hadn't connected, a missing relationship that explains a persistent problem.

This extraction makes the map-drawing investment worthwhile. A map without extraction is a visualization exercise. A map with extraction is a thinking tool that produces documented insights.

When This Fires

Common Failure Mode

Writing a summary of the map instead of an extraction of new insights: "The map shows that A connects to B and B connects to C." That describes the map's contents — it doesn't capture what the map revealed. The extraction should surprise you: "I didn't realize until I drew this that the team lead is a single point of failure connecting three otherwise independent workstreams."

The Protocol

After completing a relationship map: (1) Put down the drawing tool. (2) Write 3-5 sentences answering: "What do I now understand that I did NOT understand before drawing this?" (3) Focus on emergent properties: unexpected connections, missing links, bottlenecks, feedback loops, structural vulnerabilities. (4) If you can't identify anything new → the map confirmed existing knowledge (still useful) but didn't generate new insight (less useful). Consider whether a different mapping approach would reveal more.