# Customer Insights

> Customer feedback aggregation and health check. Collect feedback across channels, identify recurring themes and sentiment trends, score customer health, and recommend actions. Triggers on: customer feedback, customer sentiment, pulse check, customer health, what are customers saying.

- Skill: `infometa/customer-insights` (Agent Skill)
- Install (CLI): `npx skillmds@latest add infometa/customer-insights`
- Raw SKILL.md: https://api.skillmd.com/api/skills/infometa/customer-insights/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: infometa (https://skillmd.com/u/infometa)
- Updated: 2026-09-09
- Page: https://skillmd.com/skills/infometa/customer-insights

---


# Customer Insights

Collect and analyze customer feedback across channels to identify themes, sentiment trends, and churn risk.

## Workflow

1. **Collect feedback from all channels** — Email, support tickets, reviews, social media mentions. Ask the user to paste or upload if no system integration is available.

2. **Categorize by theme and sentiment** — Group feedback into themes (product, service, pricing, onboarding, etc.). Tag each item as positive, neutral, or negative sentiment.

3. **Score customer health** — For each customer segment or key account:
   - Retention risk (high / medium / low)
   - Satisfaction trend (improving / stable / declining)
   - Engagement level (active / moderate / dormant)

4. **Identify top 3 themes and trends** — Rank themes by volume and severity. Highlight any emerging issues or positive shifts.

5. **Recommend actions per theme** — For each top theme, provide:
   - Summary of what customers are saying
   - Root cause hypothesis
   - Specific action items with urgency
   - Template response for proactive outreach if applicable

## Output Format

```
## Customer Pulse Summary
- Period: [date range]
- Total feedback items: [count]
- Sentiment split: [positive %] / [neutral %] / [negative %]

## Top 3 Themes
1. [Theme] — [volume] items, [sentiment] trend
   - Root cause: ...
   - Action: ...

## At-Risk Accounts
- [Account]: [risk level] — [reason]

## Recommended Actions
1. [Urgent] ...
2. [This week] ...
3. [Ongoing] ...
```