mece-decomposition

Breaks a problem, question, metric, or topic into a MECE structure (a McKinsey staple, popularized by Barbara Minto): branches that are Mutually Exclusive (no overlaps) and Collectively Exhaustive (no gaps). Catches double-counting and blind spots so the output divides cleanly into workstreams.

The principle, the decomposition archetypes, and the ME and CE tests live in references/mece.md. Load that file before decomposing.

When to use

The user wants structure: a problem to break down, a metric to decompose into drivers, a topic to map completely, or an issue tree to build. It pairs downstream of problem definition (define the problem, then decompose it) and upstream of solving (decompose, then work each branch). If the user only needs a quick informal list, this is heavier than necessary; reach for it when the breakdown must be rigorous.

Language

Run the session and write the output in the language the user is writing in. The reference file is English; keep the framework keyed to it regardless of output language.

Inputs

• SUBJECT (required): what to decompose. A problem statement, a question, a metric, or a topic.
• PURPOSE (optional): what the breakdown is for (diagnose a drop, divide work, map options, structure an analysis). The purpose often decides which cut is best, so ask if it is unclear.