Conciseness

The body is token-efficient: dense decision tables, specific benchmarks (3%-5% conversion, CAC 50%-70% lower), and named tools with almost no prose padding or explanation of concepts Claude already knows. Minor trimmable content remains (the 'When invoking' meta-preamble and the awkward aside 'Referral is necessity in overseas markets, not alternative'), keeping it below anchor 5.

Actionability

For an instruction-only strategy skill the guidance is concrete: quantified reward benchmarks (10%-30% of price), a worked tiered example ($10/$15/$20), specific attribution windows (30-90 days, 180 for subscription), named tools per use case, and a defined Output Format. It is not fully copy-paste ready (no templates or examples of finished output), which keeps it at anchor 4 rather than 5.

Workflow Clarity

There is a coherent sequence (Initial Assessment -> 5-step Design Framework -> Output Format) and the tables provide per-option decision guidance. No validation checkpoints exist, but the skill is advisory rather than destructive/batch, so the feedback-loop cap does not apply; the missing checkpoints are the minor gap that holds it at 4.

Progressive Disclosure

The single SKILL.md is well-organized with clear section headers and self-contained at one level (no nested references). However, at ~130 lines it exceeds the simple-skill threshold, and detail like the Tools table and benchmark data could plausibly live in reference files, which is the minor organization gap of anchor 4.

Specificity

The description names the domain ('referral program strategy') and three actions ('plan, implement, or optimize'), but these verbs are generic and do not convey sub-capabilities like reward design, tracking, or fraud prevention. It sits between anchor 3 (1-2 concrete actions, not comprehensive) and anchor 4 (several specific actions), and the actions are too abstract to justify 4.

Completeness

It explicitly answers both 'what' ('plan, implement, or optimize referral program strategy') and 'when' ('When the user wants to...' and 'Also use when the user mentions...') with concrete trigger phrases, matching the anchor-5 example structure exactly. A 4 would require the 'when' to be less explicit, which is not the case here.

Trigger Term Quality

The trigger list is comprehensive with natural synonyms users would actually say: 'referral program,' 'referral marketing,' 'user referral,' 'refer-a-friend,' 'word-of-mouth growth,' 'referral rewards,' 'referral tracking,' 'referral code,' 'referral incentives,' and 'viral loop.' This matches the anchor-5 example of comprehensive natural-term coverage including synonyms.

Distinctiveness Conflict Risk

The referral niche is clearly staked with distinct trigger terms, and it even includes explicit disambiguation ('For referral landing copy, use landing-page-generator'), reducing conflict risk with adjacent skills. It does not trigger on generic marketing language, matching anchor 5.