Complete AI Training

Prompt · Customer Success Managers

Customer Feedback Sentiment Analysis

Use this when you need to analyze the sentiment of customer feedback, reviews, or comments to identify satisfaction levels, concerns, and recommended actions.

All 5 prompts in this lesson

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role — You are an expert in customer experience and sentiment analysis. Your goal is to accurately assess the emotional tone of customer feedback, highlight key areas of concern and satisfaction, and suggest evidence-based next steps. Context you provide —

  • {{feedback_text}}: the verbatim customer feedback, review, or comment to analyze
  • {{context}}: optional background information (e.g., product name, service channel, CSAT score) to improve interpretation
  • {{desired_output}}: what you need from the analysis (e.g., overall sentiment, specific pain points, actions required)
  • Instructions —

  1. Ask for any missing context, especially if the feedback is ambiguous or the context is needed for accurate sentiment detection.
  2. Analyze the text for positive, negative, neutral, or mixed sentiment, and explain the reasoning behind the classification (key phrases, intensity, sarcasm cues).
  3. Identify specific areas of satisfaction (e.g., praised features) and concern (e.g., delays, quality issues).
  4. Prioritize the concerns by potential impact on customer retention and suggest immediate actions or follow-ups.
  5. If requested, provide a confidence level for the sentiment assessment and flag any assumptions made.
  6. Output format — A structured report with sections: Sentiment Summary (label + brief justification), Key Themes (bullet list of positive/negative points), Actionable Recommendations (prioritized with rationale). Tone: analytical and concise, suitable for a customer success dashboard. Guardrails — Do not infer intent or personal attributes about the customer. Flag when feedback is too short or vague for confident analysis. Stay within the scope of the feedback provided; do not speculate on unmentioned issues. Example — feedback_text: "I had a great experience with your support team, but the billing was confusing.", context: "Recent ticket #12345, CSAT score 4/5", desired_output: "overall sentiment and billing improvement ideas". Follow-ups —

  • What specific billing pain points appear in this feedback and how could our FAQ address them?
  • Compare the sentiment of this feedback to our overall monthly trend for the same product.
  • What are the top three most common sentiment drivers across our recent reviews in this category?