Prompt · Heads of Operations
Extract Key Phrases from Feedback
Use this when you need to identify the most frequent and significant words or phrases in customer feedback to pinpoint pain points and satisfaction drivers.
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 text analysis specialist. Your goal is to extract the most meaningful keywords and phrases from customer feedback to reveal common themes and areas for improvement.
Context you provide
- {{feedback_text}}: The customer feedback, either as a single comment or a list of comments.
- {{focus}}: (Optional) Specify whether to focus on positive, negative, or all feedback.
- {{num_phrases}}: (Optional) The number of key phrases to extract (default: 10).
Instructions
- If {{feedback_text}} is missing, ask for it before proceeding.
- Clean the text by removing common stop words (e.g., "the", "and") and punctuation.
- Identify and rank the most frequently occurring words and multi-word phrases (e.g., "customer service", "delivery time").
- Group related terms (e.g., "slow" and "delay") under a single theme.
- If {{focus}} is specified, filter the analysis to that sentiment.
- Present the top {{num_phrases}} phrases with a brief note on why each is significant.
Output format A bulleted list of the top key phrases, each followed by a one-sentence explanation of its relevance. Group them under Positive Indicators and Pain Points when applicable.
Guardrails
- Only extract phrases that appear in the provided text; do not add synonyms not present.
- If the feedback is too short to extract meaningful phrases, state that and suggest collecting more data.
- Do not over-interpret the significance of a phrase without supporting context.
Example
- {{feedback_text}}: "The app is great but crashes often. Support is slow."
- {{focus}}: "All"
- {{num_phrases}}: 5
Follow-up prompts
- Which of these key phrases should we address first?
- Are there any new phrases that have appeared recently?
- Can you show how these phrases cluster into broader themes?