| name | product-feedback-loop |
| description | Designs product feedback loops connecting customer input to product decisions with transparency and communication. Use when systematizing how feedback drives your roadmap. |
| allowed-tools | Read Write Glob |
| metadata | {"author":"matthewhitcham","version":"1.0"} |
Product Feedback Loop
When to Use This Skill
Use this skill when you need to:
- Design a system that connects customer feedback to product decisions
- Build transparency around how user input shapes the roadmap
- Create communication workflows that close the loop with customers
- Establish a repeatable process for collecting, prioritizing, and acting on feedback
DO NOT use this skill for feature request boards (use feature-request-system), user research plans, or NPS survey design. This is for the end-to-end feedback-to-decision loop.
Core Principle
A FEEDBACK LOOP IS ONLY A LOOP IF INFORMATION FLOWS BACK TO THE CUSTOMER — COLLECTING FEEDBACK WITHOUT CLOSING THE LOOP TRAINS USERS TO STOP SHARING.
Phase 1: Brief
Required Inputs
| Input | What to Ask | Default |
|---|
| Product | "What product is this feedback loop for?" | No default — must be provided |
| Feedback volume | "How many pieces of feedback do you receive monthly?" | Under 100 |
| Current process | "How do you handle feedback today?" | Ad hoc / no system |
| Decision makers | "Who decides what gets built?" | Solo founder |
| Communication gap | "Do customers know when their feedback influences a decision?" | No |
GATE: Confirm the brief before designing the loop.
Phase 2: Design the Loop
The Four-Stage Feedback Loop
COLLECT → ANALYZE → ACT → COMMUNICATE
↑ |
└────────────────────────────────────┘
Stage 1: Collect — Gather feedback from all channels into one system
Stage 2: Analyze — Tag, categorize, and identify patterns
Stage 3: Act — Prioritize and decide what to build, improve, or ignore
Stage 4: Communicate — Tell customers what happened with their feedback
Collection Channels Map
| Channel | Type | Capture Method |
|---|
| In-app widget | Reactive | Auto-logged to central system |
| Support tickets | Reactive | CS team tags feedback items |
| NPS/CSAT surveys | Proactive | Scheduled quarterly |
| User interviews | Proactive | Monthly with 3-5 users |
| Social media | Reactive | Manual capture, weekly review |
| Sales calls | Reactive | CRM notes tagged as feedback |
| Review sites | Reactive | Weekly monitoring |
GATE: Confirm loop design and channels before building the system.
Phase 3: Build the System
Tagging Taxonomy
Every piece of feedback gets tagged with:
- Category: Feature area (billing, onboarding, reporting, etc.)
- Type: Bug, feature request, UX issue, praise, complaint
- Sentiment: Positive, neutral, negative
- Impact: Revenue risk, retention risk, growth opportunity
- Source: Channel it came from
Analysis Cadence
| Frequency | Activity |
|---|
| Daily | Triage incoming feedback (tag and categorize) |
| Weekly | Review patterns and rising themes |
| Monthly | Prioritization session — decide what to act on |
| Quarterly | Trend analysis and roadmap alignment |
Decision Framework
For each feedback theme, answer:
- How many users mentioned this? (Volume)
- How does it impact revenue or retention? (Business impact)
- How hard is it to address? (Effort)
- Does it align with product vision? (Strategy fit)
Score and prioritize using a simple matrix:
- High volume + high impact + low effort = Do now
- High volume + high impact + high effort = Plan for next quarter
- Low volume + low impact = Acknowledge and defer
- Conflicts with vision = Decline with explanation
Communication Templates
Feedback received:
"Thanks for sharing this. We have logged it and it will be reviewed in our next prioritization session."
Feedback being worked on:
"You asked for [feature]. We are building it now and expect to ship it by [timeframe]. We will let you know when it is live."
Feedback shipped:
"You asked, we built it. [Feature] is now live. Try it here: [link]. Thanks for helping us make [Product] better."
Feedback declined:
"We reviewed your suggestion for [feature]. After careful consideration, we decided not to pursue it because [reason]. We appreciate you sharing and encourage you to keep the ideas coming."
Phase 4: Polish
1. Transparency Practices
- Publish a monthly "What we heard" summary (blog or email)
- Show a public roadmap with "Inspired by your feedback" labels
- Tag changelog entries that came from user feedback
- Share aggregate feedback stats: "We reviewed 147 pieces of feedback this month"
2. Feedback Health Metrics
## Loop Health Metrics
- **Collection rate:** Pieces of feedback per month (trending up = good)
- **Response time:** Average time to acknowledge feedback
- **Close rate:** % of feedback items with a final status (built, declined, deferred)
- **Action rate:** % of feedback that influenced a product decision
- **Customer satisfaction with feedback process:** Survey annually
3. Quality Checklist
## Feedback Loop Checklist
- [ ] All feedback channels feed into one central system
- [ ] Tagging taxonomy is documented and consistently applied
- [ ] Weekly pattern review is on the calendar
- [ ] Monthly prioritization session is scheduled
- [ ] Communication templates exist for received, building, shipped, and declined
- [ ] Customers hear back on their feedback within one business day
- [ ] Public transparency mechanism exists (roadmap, blog, changelog labels)
- [ ] Loop health metrics are tracked monthly
- [ ] Declined feedback includes a reason (never just silence)
- [ ] Feedback that ships is celebrated and attributed
Example
Monthly "What We Heard" email excerpt:
"This month, we reviewed 83 pieces of feedback from 47 unique users. The top theme was invoice customization — 12 of you asked for custom branding on invoices. We are adding it to the March release. The second most-requested item was bulk actions on the dashboard. That is now in our Q2 plan. Three requests for Xero integration were declined for now — we are focusing on QuickBooks depth before adding new integrations."
Anti-Patterns
- Collecting without responding — the fastest way to kill feedback is to ignore it. Acknowledge everything, even if briefly.
- Building everything requested — feedback informs decisions; it does not make them. You still need strategic direction.
- Single-channel bias — if you only read support tickets, you only hear from frustrated users. Diversify collection channels.
- No decline path — saying nothing to declined requests is worse than saying no. Always close the loop.
- Feedback hoarding — sitting on months of unreviewed feedback creates a backlog that becomes useless. Process weekly.
Recovery
- Overwhelmed by feedback volume: Automate tagging where possible. Focus the weekly review on the top 10 themes, not every individual item.
- Users do not give feedback: Make it easier — reduce the form to one field. Proactively ask during support interactions.
- Feedback is mostly complaints: Actively solicit positive feedback too. Ask what users love, not just what is broken.
- Team ignores the feedback data: Tie feedback themes to revenue and retention metrics. Business impact gets attention.
