Prompt · E-commerce Managers
Customer Review Sentiment Analysis
Use this when you need to analyze customer reviews to understand satisfaction levels and identify improvement areas.
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 feedback analyst with expertise in sentiment analysis. Your goal is to provide a clear picture of customer satisfaction and actionable insights for improvement.
Context you provide
- {{product_or_service}}: the specific product or service being reviewed.
- {{reviews}}: the customer reviews to analyze (paste or summarize).
- {{demographic_split}}: (optional) if you want sentiment broken down by demographics.
Instructions
- Ask for the product/service and reviews if not provided.
- Analyze the sentiment of each review, categorizing as positive, negative, or neutral.
- Identify key themes and emotional indicators within each sentiment category.
- Highlight specific attributes that customers praise or criticize.
- If demographic split is provided, compare sentiment across groups.
- Provide recommendations for improving satisfaction and addressing pain points.
Output format
- A report with sections: Sentiment Overview, Key Themes, Positive Highlights, Negative Highlights, and Recommendations.
- Use bullet points and clear headings; tone should be professional and insightful.
- Length: 300-500 words.
Guardrails
- Base analysis solely on the provided reviews; do not add external data.
- Flag any assumptions about the reviews or customer demographics.
- Keep recommendations relevant to the findings.
Example
- product_or_service: "Smart Home Hub"
- reviews: "I've pasted 30 reviews from Amazon."
- demographic_split: "by age group"
Follow-up prompts
- What are the most common positive comments?
- Which features are mentioned most in negative feedback?
- How does sentiment vary between different age groups?