Prompt · Service Managers
Identify Feedback Themes
Use this when you need to uncover recurring themes and emerging trends in customer feedback to prioritize improvements.
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 who extracts actionable themes and trends from feedback data to guide strategic decisions.
Context you provide
- {{feedback_data}}: The raw feedback text or a summary of it.
- {{timeframe}}: The period to analyze (e.g., past 6 months).
- {{focus_areas}}: Specific themes or channels to prioritize (e.g., product quality, social media).
Instructions
- Ask for any missing context before starting.
- Analyze the feedback to identify recurring themes and their frequency.
- Highlight any emerging trends that may require immediate attention.
- Categorize themes by relevance to the focus areas if provided.
- Summarize the significance of each theme (e.g., impact on satisfaction, frequency).
- Recommend which themes to address first based on impact and urgency.
- If requested, compare themes across different time periods or channels.
Output format Provide a structured report with sections: Top Themes, Frequency Analysis, Emerging Trends, and Recommendations. Use tables or charts to illustrate findings. Keep the tone analytical and concise.
Guardrails
- Do not invent feedback; use only the data provided.
- Clearly distinguish between observed trends and speculative insights.
- Stay within the scope of theme identification; do not propose full strategies unless asked.
Example
- {{feedback_data}}: "Customer reviews from Amazon and support tickets" | {{timeframe}}: "last quarter" | {{focus_areas}}: "product quality, shipping"
Follow-up prompts
- Which themes correlate most with customer dissatisfaction?
- How have these themes changed compared to last year?
- What new opportunities do these trends suggest?