Skip to content

deanpeters/Product-Manager-Skills / skills/usage-data-investigation

usage-data-investigation

Verified

Investigate what customers actually do from product analytics, with every claim labeled Fact, Inference, or Assumption. Use when you need behavior evidence, not opinions, before deciding.

FileKindSize
SKILL.mdskill md12.5 KB
examples/sample-industrial.mdmarkdown2.1 KB
examples/sample.mdmarkdown2.1 KB
template.mdmarkdown1.6 KB

examples/sample.md

View on GitHub
# Usage Data Investigation Example: SaaS

**Question:** trials of the Route Optimizer start, but few reach a first optimized route. Where do they stall, and who?
**Unit:** trial account · **Window:** March trials · **Decision:** where to focus onboarding work

**Why this example exists:** the funnel showed one step, but the segment cut showed it was really
one kind of account, and the data had nothing to say about why.

---

```markdown
## Data check
- Tracking: "first optimized route" fires on route save; verified on three test accounts.
- Definition: unchanged in window.
- Missing: 6 internal test accounts removed; 3 trials from a partner API not tracked (noted).
- Sample: 212 trial accounts. Enough to read the funnel; thin for segments under 20.

## Funnel (March trials, N = 212)
| Step | Count | To next |
|---|---|---|
| Trial started | 212 | 78% |
| Dispatcher invited | 165 | 71% |
| Day loaded | 117 | 42% |
| First route optimized | 49 | n/a |
**Biggest drop:** day loaded → first optimized (117 → 49).

## Segments
| Segment | N | Reached first route |
|---|---|---|
| 50-100 technicians | 88 | 24 (27%) |
| 101-300 technicians | 84 | 22 (26%) |
| Over 300 | 40 | 3 (8%) |

## Findings
1. **Fact:** 49 of 212 trials (23%) reached a first optimized route within 7 days.
2. **Fact:** the biggest drop is between loading a day and optimizing it (117 to 49).
3. **Fact:** accounts over 300 technicians reached it at 8% (3 of 40) vs about 27% elsewhere.
4. **Inference:** large accounts may need a configuration step the smaller ones skip. (Not shown.)
5. **Assumption:** dispatchers don't see value. (The data can't show this.)

## What people must answer
| Question | Who | Ask |
|---|---|---|
| Why do large accounts stop after loading a day? | Dispatchers at 5 large trial accounts | "Walk me through what you did after the day loaded." |
```

## What to learn from this

- **The average (23%) hid a segment (8%).** The finding was in the cut.
- **Finding 4 is labeled an Inference,** and finding 5 an Assumption nobody should act on.
- **The investigation ends with five phone calls,** which is what the data pointed toward.
deprecation-and-migrationaddyosmani/agent-skillsManages deprecation and migration. Use when removing old systems, APIs, or features. Use when migrating users from one implementation to another. Use when migrating a database schema in production, such as renaming or dropping a column without downtime (expand/contract). Use when deciding whether to maintain or sunset existing code.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
performance-optimizationaddyosmani/agent-skillsOptimizes application performance across frontend, backend, queries, and databases. Use when performance requirements exist, when you suspect performance regressions, when Core Web Vitals or load times need improvement, when N+1 query patterns need fixing, or when profiling reveals bottlenecks.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
agent-cost-reportthedotmack/claude-memBelievable agent cost report for any period, default the last 7 full days PT, not counting today. Measured tokens from Claude Code transcripts priced at OpenRouter list prices (ESTIMATED), measured provider spend when a sanctioned source exists, note-taker cost separate, Timing-style HTML/PDF plus report.json, line-items.csv, evidence.json.Stars: 118kUpdated Claude CodeCodexVerifiedInstall
agent-cost-reportthedotmack/claude-memBelievable agent cost report for any period, default the last 7 full days PT, not counting today. Measured tokens from Claude Code transcripts priced at OpenRouter list prices (ESTIMATED), measured provider spend when a sanctioned source exists, note-taker cost separate, Timing-style HTML/PDF plus report.json, line-items.csv, evidence.json.Stars: 118kUpdated Claude CodeCodexVerifiedInstall