Sector Overview

A landscape report on an industry: how it earns money, who is winning, what it is worth, and what would change that. The quality of the report is decided in the first step, not the last: a set classified into the wrong archetype produces the right-looking report with the wrong metrics, the wrong multiple and the wrong risks.

Evidence labels, source tiers, staleness, the readiness posture and the intake limits: .agents/skills/research-conventions/SKILL.md, read before the first deliverable.

Workflow

1. Scope it

• Sector or subsector, and how narrowly the boundary is drawn.
• Purpose: a client report, a research packet, pitch material, idea generation.
• Angle: a neutral landscape, or a thematic thesis the report is testing.
• Universe: public names only, or private included.
• Depth: default working analysis; a sector screen is a first pass. Bands and what to cut first: .agents/skills/research-conventions/references/depth.md.

2. Classify into an archetype before analysing

Write down how each company in the set actually earns money, and put it in an archetype. Business model beats label: two names carrying the same sector tag and different revenue mechanics share no metric set, no valuation lens and no risk list. Classify from the revenue recognition note and the segment note of the largest names in the set, not from the sector tag.