Discover · Designed
Customer Feedback Intelligence
Customer feedback is fragmented across Sales, CS, support, calls, and feature requests.
Career signal: Product Strategy, Product Operations, Product Management
All projects1. Problem
Customer feedback is fragmented across Sales, CS, support, calls, and feature requests.
Roadmap and GTM cannot see themes with account and revenue context attached.
Without this: Duplicate requests, lost commitments, and no closed loop from request to release.
2. Users
Primary: Product and product operations
Secondary: AEs and CSMs who captured the original signal
Job: See what customers are asking for, how often, and what it is worth — then close the loop.
3. Evidence
Sales–product feedback loops at Truffle, plus release work that needed prior requests, POC blockers, and lost-deal gaps in one place — not a standalone feature tracker.
4. Goals and non-goals
Goals
- Ingest and classify feedback with account context.
- Deduplicate themes and attach revenue / stage context.
Non-goals
- Replacing the product backlog tool.
- Auto-committing roadmap items from sentiment.
5. MVP
A theme board sourced from field notes, with account tags and duplicates collapsed.
Question the MVP tests: Does structured feedback change prioritization faster than a raw request list?
6. Workflow
- Feedback sources
- Ingestion
- Extraction
- Classification
- Deduplication
- Account / revenue context
- Themes
- Prioritization
7. System design
Plain language
- Sources
- Ingestion
- Normalization
- Themes
- Prioritization
- Closed loop
Technical
- CRM / call / ticket sources
- Ingestion workers
- Extraction and classification
- Postgres
- Prioritization UI
8. Data model
- Signal
- Theme
- Account
- Evidence
- Persona
- Outcome
9. Metrics
Operational
- Reduced duplicate work
- Faster theme identification
Behavioral
- Product review of themed feedback
Business
- Hypothesis: better roadmap prioritization and closed-loop customer communication
10. Business impact hypothesis
If feedback is themed with account context, roadmap prioritization and closed-loop communication should improve. No fake revenue attached.
11. Tradeoffs
- Human confirmation of themes before they hit roadmap reviews.
- Account context over volume-only scoring.
12. Prototype
Designed system. The field loop exists; the intelligence layer is not a shipped app.
Intended value
- Faster identification of product themes
- Better roadmap prioritization
- Reduced duplicate work
- Stronger closed-loop customer communication
13. What I would build next
- CRM and call-note ingestion.
- Deduplicated themes.
- Join to release intelligence when a theme ships.