sboghossian Skill Adoption by Tier — Report
Internal product-analytics report tracking which skills each user tier (Free, Plus, Pro, Enterprise) actually uses, how often, and with what success. The primary output is a tier-packaging signal: which skills create tier-upgrade motivation, and which are underused in a tier that could leverage them more.
Purpose
Tier packaging decisions without adoption data are guesses. This report answers:
- Which skills are primarily used by paid tiers? These are the value drivers — protect them behind tier gates or use them as upgrade prompts.
- Which skills are used across all tiers? These are either correct free-tier features or packaging leaks.
- Which high-value skills have low adoption in the tier that should use them most? These are discoverability problems — fix with onboarding or in-product nudges.
- Are Enterprise users getting distinct value from specialized skills? If not, enterprise packaging needs differentiation.
Data inputs
| Signal | Source | Notes |
|---|---|---|
| Skill invocation events | Product analytics (PostHog / Mixpanel) | Tagged with skill_id, user_id, tier, timestamp |
| User tier | Auth / billing database | Free / Plus / Pro / Enterprise |
| Session outcome | Satisfaction score, explicit feedback, task completion | Proxy for whether the skill delivered value |
| Skill depth invoked | Single-turn vs multi-turn agent workflows | Deep-research workflows signal high-engagement users |
Metrics
For each skill × tier combination, compute:
| Metric | Definition |
|---|---|
| Unique users | Count of distinct users in the tier who invoked the skill in the reporting period |
| Total invocations | Total calls, including re-runs and follow-up turns |
| Adoption rate | Unique users / total users in tier |
| Satisfaction rate | % of invocations that received positive feedback (thumbs up, 4-5 star, or no negative signal) |
| Upgrade conversion | For Free-tier users: % who upgraded within 7 days of invoking the skill |
| Avg invocations per active user | Depth of engagement among those who use the skill |
Output format
Tier usage summary table
For each tier (column) and each skill (row):
| Skill | Free | Plus | Pro | Enterprise | Tier concentration |
|---|---|---|---|---|---|
| research-statute-lookup | 12% | 28% | 42% | 18% | Pro-heavy |
| draft-nda-unilateral | 5% | 22% | 54% | 19% | Pro-heavy |
| research-deep-research-orchestrator | 1% | 8% | 35% | 56% | Enterprise-heavy |
| review-contract-redline | 8% | 30% | 48% | 14% | Pro-heavy |
| … | … | … | … | … | … |
Tier concentration classification:
- Free-heavy (> 40% of invocations from Free): potential packaging leak or correct free feature
- Plus-heavy: good Plus value driver
- Pro-heavy: core Pro value driver; protect or use as upgrade prompt
- Enterprise-heavy: strong Enterprise differentiator
Top skills by upgrade conversion (Free → Paid)
| Skill | Free adoption | Upgrade rate within 7 days |
|---|---|---|
| research-deep-research-orchestrator | 2% | 18% |
| review-contract-redline | 9% | 12% |
| … | … | … |
These are the skills to emphasize in Free-tier onboarding to drive upgrade intent.
Underutilized high-value skills
Skills that are available to a tier but adopted by < 10% of that tier's users, where the skill is a core value proposition:
| Skill | Tier | Adoption | Hypothesis | Recommendation |
|---|---|---|---|---|
| research-jurisdiction-comparison | Pro | 6% | Not discoverable in UI | Add to Pro onboarding tour |
| … | … | … | … | … |
Enterprise-specific skill usage
Which skills are Enterprise-only or disproportionately used by Enterprise:
- Multi-jurisdiction workflows
- Deep research orchestrator
- Compliance gap analysis (multi-framework)
- Cap table sanity (VC/M&A use case)
If Enterprise users are NOT using these, that is a customer-success problem, not a packaging problem.
Packaging recommendations
Rules derived from the data:
- Upgrade-gate skills where Free adoption is > 20% but upgrade conversion is high: move behind a paywall with a clear "upgrade to unlock" prompt.
- Highlight in Plus/Pro marketing any skill with satisfaction rate > 85% in that tier.
- Invest in discoverability for skills with adoption < 10% in the tier that should use them most, if satisfaction rate > 80% among users who do invoke them (hidden gem).
- Retire or redesign any skill with adoption < 2% across all tiers and no growth trend over 90 days.
Cadence
- Weekly: top-10 skill invocations by tier (lightweight pulse)
- Monthly: full matrix + packaging recommendations
- Quarterly: deep-dive with cohort analysis (new users vs retained users per tier)
Related skills
- [[report-weekly-ai-quality-trend]]
- [[report-jurisdiction-coverage-matrix]]
- [[report-competitor-output-comparison-weekly]]
- [[report-hallucination-rate-tracker]]