Index of every page on this site: llms.txt
All skills
deanpeters avatar

/usage-data-investigation

@476ae12
Run it yourself

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.

Use this Skill: https://skilld.dev/gh/deanpeters/product-manager-skills/usage-data-investigation

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

examplessample-industrial.md

≈535 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Usage Data Investigation Example: Industrial

Question: what is the early-life fault pattern on NFA-500 units, and does it differ by configuration? Unit: commissioned unit · Window: first 90 days in the field · Decision: whether to issue a field bulletin

Why this example exists: the "usage" data is service and warranty records, small counts, and the pattern depended on a field the system recorded inconsistently.


## Data check
- Source: service system (fault tickets) joined to ERP (units commissioned).
- Tracking: fault cause is free text in 40% of tickets; unusable for those.
- Definition: "fault" includes two ticket types that changed on Aug 1 (flagged).
- Missing: units at partner sites reported by partners monthly (lag up to 5 weeks).
- Sample: 64 commissioned units, 11 fault tickets. Small: read counts, not rates.

## Findings
1. **Fact:** 11 fault tickets on 64 units in first 90 days (17%; 11 of 64).
2. **Fact:** 7 of the 11 involve the bracket-mounted configuration; 4 involve the standard mount. The bracket-mounted configuration is 28 of 64 units.
3. **Inference:** bracket-mounted units fault at a higher rate (7 of 28 vs 4 of 36). With these counts, the gap could be noise.
4. **Fact:** 5 of the 7 bracket-mounted faults sit with one partner crew.
5. **Assumption:** the torque spec is the cause. (Free-text causes can't confirm it.)

## What people must answer
| Question | Who | Ask |
|---|---|---|
| Is it the bracket, or the crew? | Service lead; that partner's installers | "Show me how you torque the fasteners. What does the guide say?" |
| What is the real cause on the 40% unlabeled tickets? | Service technicians | Review the 4 open tickets by hand |

**Do not issue a bulletin yet.** Do re-check at 120 days with the cause field made mandatory.

What to learn from this

  • Finding 3 is an Inference with a warning. Seven of 28 versus four of 36 is not a result.
  • Finding 4 changed the question from "the bracket" to "the crew," which no chart would have said.
  • The recommendation is to wait and fix the data, which is a legitimate outcome.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 warning6d3 checks · Risk SAFE
  • Gen Agent Trust Hub6d

    This skill is a collection of markdown-based workflow instructions for analyzing product usage data. It contains no executable code, scripts, or network-capable operations and is safe to use.

  • Socket6d

    No alerts

  • Snyk6d

    Risk: MEDIUM · 1 issue

Signed by skilld at 476ae12. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub last week.

Activeupdated last week
argument-hint
[the question about customer behavior, and what data you can reach]
type
workflow
theme
discovery-research
Other metadata
intent
Turn a question about customer behavior into a disciplined read of product usage data: define the question and the unit of analysis, check the data's trustworthiness, look at funnels, cohorts, and segments, separate what the data shows from what you conclude, and name the follow-up that qualitative research must answer. Quantitative discovery that points to the "why" without claiming it.
best_for
[
  "Finding where customers drop off, and who, before deciding what to build",
  "Checking a belief about behavior against what the product data shows",
  "Reading a launch or trial funnel, cohorts, and segments honestly",
  "Knowing which question the numbers can't answer and a conversation must"
]
scenarios
[
  "Trials start but few reach a first optimized route. Where do they stall?",
  "Leadership says feature X is popular. What does the usage data actually show?",
  "Do customers who onboarded after the fix retain better than those before?",
  "Commissioned units report faults early. Which unit configurations, and when?"
]
estimated_time
45-90 min

README badge

README badge for deanpeters/product-manager-skills/usage-data-investigation