agentsclimarketplace

Mece framework

Skill ConrayGambit/Strategy-Consultant-5-Consulting-Frameworks/skills/mece-framework

Tier-1 strategy-consultant frameworks (MECE, Issue Trees, Hypothesis-Driven, Pareto, So What?) packaged as Claude Skills + LLM-agnostic prompts. Drop in any LLM and get whiteboard-style structured analysis. Adapted from Analyst Academy on YouTube.

Install
npx -y skills add ConrayGambit/Strategy-Consultant-5-Consulting-Frameworks --skill mece-framework

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 13 stars13 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Categorize the factors of a problem so they don't overlap (Mutually Exclusive) and nothing is missed (Collectively Exhaustive). Outputs nested Markdown bullets. Use for problem structuring, "make sure I've thought of everything", or step 1 of a strategic analysis.

SKILL.md

3.3 KB, as published. Nobody here has run it

MECE Framework

Concept

MECE — Mutually Exclusive, Collectively Exhaustive — is the foundation of structured problem solving:

  • Mutually Exclusive: No factor belongs in two categories.
  • Collectively Exhaustive: Every relevant factor is captured.

Non-MECE breakdowns look rigorous but aren't — overlapping categories double-count, and missing categories blind you to root causes.

Required output format

Nested Markdown bullets. Top-level bullets are categories in bold; nested bullets are the specific factors under each category.

- **Category 1**
  - Sub-factor A
  - Sub-factor B
- **Category 2**
  - Sub-factor C
  - Sub-factor D

Defaults & flex points

DefaultWhen to flex
3–6 top-level categoriesIf the problem has only 2 real categories, use 2 — don't pad. If it has 7, use 7.
2–5 sub-factors per categoryUse as many as the problem actually contains.
Categories specific to the problemGeneric "People/Process/Technology" buckets are a smell — only use them if they genuinely fit.

Rule that doesn't flex: the categories must be MECE. Test each one — could a factor reasonably belong to two? Then merge or split. Could relevant factors not fit anywhere? Then add a category.

Example — logistics

Problem: A regional courier company's on-time delivery rate fell from 94% to 81% in six months.

- **Network capacity**
  - Vehicle fleet utilization (peak hour saturation)
  - Driver headcount vs. demand
  - Hub sortation throughput
- **Routing & dispatch**
  - Algorithm changes after the Q2 system migration
  - Address-data quality (geocoding errors)
  - Manual route overrides by dispatchers
- **Demand pattern shift**
  - New high-volume customer added in Q1
  - Seasonal swing into peak retail months
  - Same-day delivery share rising
- **Workforce**
  - Driver turnover and onboarding ramp time
  - Overtime caps from new labor agreement
  - Sick days / absenteeism trend
- **External**
  - Construction zones in two metro areas
  - Two consecutive months of unusual weather

Notice: every category is specific to a courier business. No "People/Process/Technology." Every sub-factor is concrete enough to investigate.

Common mistakes

  • Overlap. A category like "Operational issues" overlaps with everything else; rename to something specific.
  • Generic buckets. "External factors" is a placeholder, not a category. Name what's external.
  • List-as-categorization. A flat list of 12 factors with no grouping isn't MECE — it's just inventory.

When to use this skill alone vs. as part of the full analysis

  • Standalone: when the user just wants to make sure they're thinking comprehensively before going off to investigate.
  • As part of full analysis: invoke strategy-consultant — MECE is step 1 of 5.

Acknowledgment & License

Adapted from Analyst Academy on YouTube — 5 Consulting Frameworks to Solve Any Problem. MIT-licensed; see LICENSE.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.