ProductSights Documentation
Learn how to collect, triage, and analyze product feedback with ProductSights.
Welcome to the ProductSights documentation. ProductSights transforms scattered customer feedback into structured, actionable insights using AI-powered triage and interactive analytics.
What is ProductSights?
ProductSights is a feedback intelligence platform that helps product teams:
- Collect feedback from 12+ sources — Slack, Intercom, email, app stores, and more
- Triage automatically with AI that classifies, scores, and deduplicates every piece of feedback
- Explore trends through interactive dashboards and natural-language AI queries
- Act by syncing insights to your roadmap tools or querying from your IDE
Quick links
- Creating your account — sign up and set up your workspace
- Adding your first source — connect Slack, Intercom, Zendesk, email, or other channels
- Feedback Widget — embed in-app feedback capture
- Browser Extension — capture feedback from any webpage
- Analytics integration — Amplitude, Mixpanel, PostHog enrichment
- AI Triage — how automatic classification works
- Analytics — dashboards, explorer, and AI querying
- MCP Server — query insights from Claude Code, Cursor, or VS Code
- Feedback Portal — branded public submission pages
How it works
ProductSights follows a four-step pipeline:
- Ingest — Feedback arrives from connected sources or manual entry
- AI Triage — Each item is classified into a category (feature request, bug report, etc.), assigned a sentiment score, and given a priority rating
- Review — Your team can review, approve, or reject triaged insights in the review queue
- Analyze & Act — Use dashboards to spot trends, then sync approved insights to Jira Product Discovery or Productboard
Key concepts
| Concept | Description |
|---|---|
| Insight | A single piece of product feedback after AI processing |
| Category | One of six types: Feature Request, Bug Report, UX Issue, Praise, Question, or Churn Signal |
| Sentiment | Positive, negative, or neutral tone of the feedback |
| Priority Score | A 0–100 score indicating importance (higher = more urgent) |
| Cluster | A group of similar insights identified by AI |
| Source | A connected channel where feedback originates |