name: Feedback Analyzer
slug: feedback-analyzer
description: Analyze customer feedback to extract actionable insights, identify patterns, and prioritize improvements
category: customer-support
complexity: complex
version: "1.0.0"
author: "ID8Labs"
triggers:
- "feedback analysis"
- "customer feedback"
- "sentiment analysis"
- "voice of customer"
- "feature requests"
- "user feedback"
tags:
- feedback
- analytics
- sentiment
- voice-of-customer
- customer-insights

Feedback Analyzer

Expert customer feedback analysis system that transforms unstructured feedback into actionable product and service insights. This skill provides structured workflows for collecting, categorizing, analyzing, and acting on customer feedback from multiple sources.

Customer feedback is the most direct signal of what's working and what isn't. But raw feedback is noisy, contradictory, and overwhelming. This skill helps you extract patterns, prioritize themes, and close the feedback loop effectively.

Built on voice-of-customer best practices and qualitative research methods, this skill combines text analysis, pattern recognition, and stakeholder communication to turn feedback into action.

Core Workflows

Workflow 1: Feedback Collection & Aggregation

Gather feedback from all sources into unified view

1. Feedback Sources
2. Direct Surveys: NPS, CSAT, CES, custom surveys
3. Support Channels: Tickets, chat transcripts, calls
4. In-App Feedback: Feature requests, bug reports, ratings
5. Social Media: Mentions, reviews, comments
6. Sales Conversations: Objections, lost deal reasons
7. User Research: Interviews, usability tests

8. Community: Forums, Slack, Discord

9. Data Standardization
   | Field | Description |
   |-------|-------------|
   | Source | Where feedback came from |
   | Date | When received |
   | Customer ID | Link to customer record |
   | Segment | Customer type/tier |
   | Raw Text | Original feedback |
   | Category | Topic classification |
   | Sentiment | Positive/neutral/negative |
   | Priority | Urgency/impact level |

10. Collection Automation

11. API integrations with feedback tools
12. Automatic ticket tagging
13. Survey response routing
14. Social listening alerts

15. Scheduled data syncs

16. Quality Filters

17. Remove spam and duplicates
18. Flag potentially inaccurate data
19. Note context (e.g., during outage)
20. Weight by customer segment
21. Identify feedback loops (same issue, multiple channels)

Workflow 2: Categorization & Tagging

Organize feedback into meaningful categories

1. Category Taxonomy
2. Product Features: Specific functionality feedback
3. Usability/UX: Interface and experience issues
4. Performance: Speed, reliability, bugs
5. Pricing/Value: Cost concerns and value perception
6. Support Experience: Service quality feedback
7. Onboarding: Getting started experience
8. Documentation: Help content feedback

9. Integration: Third-party connection issues

10. Subcategory Examples
    Product Features ├── Feature Requests │ ├── New feature ideas │ └── Feature enhancements ├── Missing Features │ ├── Competitor comparisons │ └── Workflow gaps └── Feature Feedback ├── What works well └── What doesn't work

11. Tagging Best Practices

12. Use consistent, specific tags
13. Allow multiple tags per feedback
14. Create tag hierarchy (parent/child)
15. Review and consolidate tags quarterly

16. Train team on tagging standards

17. Automated Classification

18. Keyword-based routing rules
19. ML-based topic classification
20. Sentiment detection
21. Priority scoring algorithms
22. Entity extraction (features, pages, actions)

Workflow 3: Sentiment & Urgency Analysis

Understand emotional context and priority

1. Sentiment Classification
   | Sentiment | Indicators | Action Level |
   |-----------|------------|--------------|
   | Very Negative | Anger, threats to leave | Urgent escalation |
   | Negative | Frustration, complaints | Address in sprint |
   | Neutral | Suggestions, questions | Standard review |
   | Positive | Praise, appreciation | Share with team |
   | Very Positive | Advocacy, testimonial | Request case study |

2. Urgency Scoring Factors

3. Customer tier (enterprise = higher weight)
4. Revenue at risk
5. Frequency of same issue
6. Time sensitivity mentioned
7. Escalation history

8. Regulatory/compliance implications

9. Trend Detection

10. Volume spikes (sudden increase in topic)
11. Sentiment shifts (getting worse/better)
12. New issues emerging
13. Seasonal patterns

14. Release-correlated feedback

15. Alert Triggers

16. High-value customer escalation
17. Sentiment score below threshold
18. Issue volume exceeds normal
19. Churn-risk keywords detected
20. Security/privacy concerns

Workflow 4: Pattern Recognition & Insights

Extract actionable patterns from feedback mass

1. Quantitative Analysis
2. Frequency by category
3. Trend over time
4. Segment distribution
5. Correlation with churn

6. Impact on NPS/CSAT

7. Qualitative Analysis

8. Representative quote extraction
9. Use case pattern identification
10. User journey mapping
11. Pain point articulation

12. Unmet need discovery

13. Insight Synthesis
    ```
    Insight Template:

FINDING: [What the data shows]
EVIDENCE: [Supporting data points and quotes]
IMPACT: [Business/customer impact if unaddressed]
RECOMMENDATION: [Suggested action]
PRIORITY: [High/Medium/Low with rationale]
```

1. Root Cause Analysis
2. Group related feedback
3. Identify underlying causes
4. Map to user journey stages
5. Connect to product/process gaps
6. Distinguish symptoms from causes

Workflow 5: Reporting & Action

Communicate insights and drive improvements

1. Stakeholder Reports
   | Audience | Focus | Frequency |
   |----------|-------|-----------|
   | Product | Feature requests, usability | Weekly |
   | Support | Training needs, process issues | Weekly |
   | Executive | Strategic themes, churn drivers | Monthly |
   | Engineering | Bugs, performance issues | Real-time |
   | Marketing | Positioning, messaging gaps | Monthly |

2. Report Components

3. Executive summary
4. Key metrics and trends
5. Top themes with supporting data
6. Representative customer quotes
7. Recommended actions

8. Open questions

9. Feedback Loop Closure

10. Track feedback → action connection
11. Communicate changes to customers
12. Measure impact of changes
13. Update customers on feature requests

14. Publish "You Asked, We Built" updates

15. Action Prioritization

16. Impact on retention/growth
17. Effort to address
18. Customer segment affected
19. Strategic alignment
20. Quick wins vs. long-term investments

Quick Reference

Best Practices

Collection

• Capture feedback at moments of truth
• Use consistent rating scales
• Include open-ended questions
• Don't over-survey (survey fatigue)
• Thank customers for feedback

Categorization

• Create mutually exclusive categories
• Allow multi-tagging for complex feedback
• Review taxonomy quarterly
• Train team on consistent tagging
• Use automation for high-volume

Analysis

• Look for patterns, not anecdotes
• Weight by customer segment value
• Consider feedback context
• Triangulate across sources
• Separate signal from noise

Reporting

• Lead with insights, not data
• Use customer quotes strategically
• Connect to business impact
• Recommend specific actions
• Track what gets done

Closing the Loop

• Communicate what you've heard
• Update on progress
• Thank specific contributors
• Measure impact of changes
• Celebrate wins publicly

Analysis Frameworks

Framework 1: Jobs-to-be-Done Lens

Analyze feedback through customer goals:
- What job is the customer trying to do?
- What's preventing success?
- What would "done" look like for them?
- How does our product help or hinder?

Framework 2: Kano Model

Categorize feature feedback:
- Basic: Expected, causes dissatisfaction if missing
- Performance: More is better, linear satisfaction
- Delighters: Unexpected, causes delight if present
- Indifferent: No impact on satisfaction

Framework 3: Impact/Effort Matrix

Prioritize actions:

High Impact
    │   Quick Wins    │   Major Projects
    │   (Do Now)      │   (Plan Carefully)
────┼─────────────────┼───────────────────
    │   Fill-ins      │   Thankless Tasks
    │   (Do If Time)  │   (Reconsider)
Low │                 │                  High
    └─────────────────┴───────────────────
                    Effort

Framework 4: Customer Journey Mapping

Map feedback to journey stages:
1. Awareness & Discovery
2. Evaluation & Decision
3. Onboarding & Activation
4. Regular Usage
5. Growth & Expansion
6. Support & Recovery
7. Renewal & Advocacy

Report Templates

Weekly Product Feedback Summary

# Feedback Summary: [Week]

## Key Numbers
- Total feedback received: [X]
- Sentiment breakdown: [+/neutral/-]
- Top category: [Category] ([%])

## This Week's Themes

### Theme 1: [Title]
[Brief description of pattern]
- Volume: [X] mentions
- Segments affected: [List]
- Representative quote: "[Quote]"
- Recommendation: [Action]

### Theme 2: [Title]
[Same format]

## Emerging Issues
- [New issue to watch]

## Positive Highlights
- "[Positive quote]" - [Customer]

## Actions from Last Week
- [Action taken] → [Result]

Monthly Executive Report

# Voice of Customer: [Month]

## Executive Summary
[2-3 sentences on key findings and business impact]

## Metrics
| Metric | This Month | Last Month | Trend |
|--------|------------|------------|-------|
| NPS | [Score] | [Score] | [↑↓] |
| CSAT | [Score] | [Score] | [↑↓] |
| Feedback Volume | [X] | [X] | [↑↓] |

## Strategic Themes

### 1. [Theme Name]
**Impact**: [Business impact if unaddressed]
**Evidence**: [Data summary]
**Recommendation**: [Strategic action]

### 2. [Theme Name]
[Same format]

## Competitive Intelligence
[What customers are saying about competitors]

## Customer Quotes
[3-5 impactful quotes with context]

## Recommended Actions
1. [Priority action with owner]
2. [Priority action with owner]

## Appendix
[Detailed data tables]

Red Flags

• Echo chamber: Only hearing from vocal minority
• Recency bias: Overweighting recent feedback
• Volume bias: Prioritizing loudest over important
• Missing segments: Not hearing from key customers
• Action gap: Collecting but not acting
• No closure: Customers don't know they were heard
• Stale categories: Taxonomy doesn't match current product
• Sentiment-only: Missing nuance in analysis