Customer Feedback Analyzer
The Customer Feedback Analyzer takes a batch of customer reviews, survey responses, support ticket comments, or NPS verbatims and produces a structured analysis with recurring themes, sentiment distribution, prioritized improvement recommendations, and direct customer quotes as evidence. It turns a pile of qualitative data into actionable product and service intelligence.
Product managers analyzing feature requests, support leads running quarterly reviews, customer success teams preparing voice-of-customer reports, and founders trying to understand why users churn use this template. It handles feedback from any source: app store reviews, G2 or Capterra reviews, NPS surveys, CSAT comments, support ticket feedback, social media mentions, or interview transcripts.
The prompt works because it applies a systematic analysis framework rather than reading feedback one at a time and trying to spot patterns manually. It categorizes feedback by theme, scores sentiment, distinguishes between frequent complaints and high-impact complaints (which are often different), and produces a prioritized action list that product and support teams can act on directly. The inclusion of exact customer quotes makes the output credible for stakeholders who need evidence, not just summaries.
This prompt is just the starting point
Score it with AI, optimize it with one click, track versions, and build your prompt library.
The Prompt
Analyze the following customer feedback and produce a structured insights report: **Product/Service**: [YOUR PRODUCT OR SERVICE NAME] **Feedback Source**: [WHERE THIS FEEDBACK COMES FROM, e.g., "App Store reviews (last 90 days)", "Post-support CSAT surveys", "NPS detractor verbatims Q1 2026", "G2 reviews"] **Total Feedback Items**: [APPROXIMATE NUMBER, e.g., "47 reviews"] **Feedback Data**: ``` [PASTE YOUR FEEDBACK DATA HERE. Include as many items as possible. Each item should ideally include: - The feedback text - A rating or score if available (e.g., 1-5 stars, NPS 0-10) - Date (if available) - Customer segment (if known, e.g., "Enterprise", "Free tier", "New user") Example format: --- Rating: 2/5 | Segment: Pro plan | Date: 2026-03-15 "Love the product overall but the export feature is broken half the time. I've reported it twice and got generic responses. Starting to look at alternatives." --- Rating: 5/5 | Segment: Free tier | Date: 2026-03-12 "Exactly what I needed. Simple, fast, does the job. Only wish it had a dark mode."] ``` **Analysis Focus** (optional): [ANY SPECIFIC AREA YOU WANT DEEPER ANALYSIS ON, e.g., "We just redesigned the onboarding flow, focus on onboarding-related feedback", "We are losing enterprise customers, focus on enterprise-specific pain points"] Produce the following: ### 1. Executive Summary 3-5 sentences: overall sentiment trend, the single biggest takeaway, and the most urgent action item. ### 2. Sentiment Distribution Break down the feedback by sentiment: - Positive (what percentage, what they praise most) - Mixed (what percentage, what they like vs. what they criticize) - Negative (what percentage, primary drivers of dissatisfaction) ### 3. Theme Analysis Identify the top 5-8 recurring themes. For each theme: - **Theme Name**: [Clear, specific label] - **Frequency**: How many feedback items mention this theme - **Sentiment**: Is this theme mostly positive, negative, or mixed? - **Representative Quotes**: 2-3 direct quotes that capture this theme - **Trend Signal**: Is this theme growing, stable, or declining (if dates are available)? ### 4. Priority Matrix Rank issues by combining frequency (how many customers mention it) and severity (how much it affects satisfaction or retention): | Issue | Frequency | Severity | Priority | Recommended Action | ### 5. Positive Signals What customers love and what you should protect, double down on, or use in marketing. Include quotable praise organized by theme. ### 6. At-Risk Signals Feedback patterns that indicate churn risk: mentions of competitors, language like "looking at alternatives", declining sentiment from previously happy customers, or repeated unresolved complaints. ### 7. Recommended Actions A prioritized list of 5-7 specific actions, each with: - What to do - Which team owns it (product, support, engineering, marketing) - Expected impact on customer satisfaction - Supporting evidence (quote count and representative examples)
Usage Tips
- Include at least 20 feedback items: Pattern analysis requires volume. Fewer than 20 items produces unreliable themes. Aim for 50 or more for meaningful insights.
- Keep ratings and metadata attached: A 2-star review saying "export is broken" carries different weight than a 4-star review saying the same thing. Ratings and customer segments add crucial context.
- Use the priority matrix in product meetings: The frequency-vs-severity matrix gives product teams a data-backed framework for roadmap decisions instead of relying on whoever speaks loudest.
- Run monthly for trend detection: A single analysis is a snapshot. Monthly analysis reveals whether issues are getting better or worse, which is often more actionable than the issues themselves.
- Feed the positive signals to marketing: The "Positive Signals" section contains authentic customer language that makes better ad copy and testimonials than anything a copywriter invents.
Get more from this prompt
Save it, score it with AI, optimize it, and track every version. Free to start.