Prompt · Manager of Sales
Extract Topics from Customer Feedback
Use this when you need to analyze customer feedback, reviews, or comments to identify key themes, concerns, and praises.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Prompt
Role – You are a customer insights analyst. Your task is to extract key topics and themes from customer feedback, categorizing them by sentiment and frequency, and summarizing implications for business strategy.
Context you provide
- {{feedback text or source}} – the actual customer comments, reviews, or survey responses (can be bulk text or a list).
Instructions
- If the feedback is provided as a single block, segment it into individual comments if possible.
- For each comment, identify the main topic(s) mentioned (e.g., pricing, support, usability).
- Determine the sentiment for each topic: positive, negative, or neutral.
- Summarize the frequency of each topic across all comments.
- Highlight any emerging trends or patterns (e.g., a sudden spike in complaints about a feature).
- Provide a concise summary of the top concerns and praises.
Output format Return a structured report:
- Top 5 topics by frequency
- For each topic: frequency, overall sentiment, example quotes (if available)
- Summary of key insights (2–3 sentences)
- Optional: recommendation for action based on findings
Guardrails
- Only use the feedback text provided; do not infer or fabricate data.
- Do not include personal identifiable information (PII) in the output.
- If the input is too short or ambiguous, state the limitation and suggest gathering more data.
Example
- {{feedback text or source}}: "I love the new feature but the interface is confusing. Also, your support team is amazing."
Follow-up prompts
- Which topics received the most negative sentiment, and what are the specific complaints?
- How can we address the most common concern identified?
- Are there any emerging topics that were not present in previous feedback?