Build branches that are MECE at every level

Ensure each set of branches is mutually exclusive (no overlap) and collectively exhaustive (nothing important missing).

Why it works

MECE branches prevent double-counting in analysis and ensure coverage. When branches overlap, evidence shared by two branches inflates their apparent weight — the overlapping factor appears to be supported by two reasons when it is actually one. When branches omit a dimension, the analysis may find a solution within the explored space while missing the actual driver. MECE is a logical error-correction discipline, not a stylistic preference.

How to do it

1. For each set of branches, ask: "Could an item fit in more than one branch?" If yes, redefine to eliminate the overlap.
2. Ask: "Is there any dimension of this question that none of these branches address?" If yes, add a branch.
3. Use standard MECE decompositions where they fit: by geography, by customer segment, by cost/revenue, by process step.
4. Check that the branches at each level could be combined to fully answer the parent question.

Evidence

MECE as a structuring principle is McKinsey methodology, widely adopted in strategy consulting and analytical roles. Its logical basis is sound; empirical studies comparing MECE to non-MECE analytical approaches in controlled settings are not available. (anecdotal)

Perfect MECE is rarely achievable in complex real problems; the principle is an aspiration and a diagnostic tool, not a binary standard.

Common mistake

Treating a non-MECE tree as "good enough" when it is visually clean — a well-designed non-MECE tree will silently bias the analysis toward the dimensions it covers and away from the ones it misses.