Trustpilot Sentiment Tracker
Automatically pull Trustpilot reviews, score sentiment, log everything to a spreadsheet, and alert your team on Slack.
The Challenge
Your Trustpilot page is a live feed of customer sentiment, but nobody has time to read every review, tag it by theme, and track whether things are getting better or worse. By the time a negative trend surfaces, it's already a fire drill. You need a system that watches reviews, logs the data, and pings you when something's off.
What This Prompt Does
Pull Reviews
Fetch recent Trustpilot reviews with ratings and text
Score Sentiment
Classify each review as positive, neutral, or negative
Log to Sheets
Append every review to a spreadsheet for trend tracking
Slack Alerts
Send a digest and warn when negative sentiment spikes
The Prompt
The Prompt
Task
Use @Trustpilot/Get ReviewsName it "Trustpilot/Get Reviews" and call it with @Trustpilot/Get Reviews to pull recent reviews for a given company domain, analyze the sentiment of each review, log the results to a Google Sheets spreadsheet using @Google Sheets/Append RowName it "Google Sheets/Append Row" and call it with @Google Sheets/Append Row, and send a summary digest to Slack via @Slack/Send MessageName it "Slack/Send Message" and call it with @Slack/Send Message.
Example: Track weekly sentiment for "acme.com" and alert the team when negative reviews spike.
Input
The user will provide:
- A company domain (e.g., "acme.com")
- A Google Sheets spreadsheet ID for logging
- A Slack channel name for alerts
Example: "Track sentiment for shopify.com, log to sheet ID 1abc123, and post to #brand-health"
Context
What to Analyze
For each review:
- Rating (1-5 stars)
- Sentiment classification: Positive (4-5), Neutral (3), Negative (1-2)
- Key themes mentioned (pricing, support, product quality, onboarding, etc.)
- Reviewer location for geographic patterns
Aggregate metrics:
- Average rating for the batch
- Sentiment distribution (% positive, neutral, negative)
- Top 3 complaint themes from negative reviews
- Top 3 praise themes from positive reviews
Workflow Steps
- Fetch 2-3 pages of recent Trustpilot reviews using the company domain
- Classify each review by sentiment and extract key themes
- Calculate aggregate metrics for the batch
- Append each review as a row to Google Sheets with columns: Date, Reviewer, Rating, Sentiment, Themes, Review Text
- Send a Slack message summarizing the sentiment breakdown and flagging any critical issues
What Counts as a Valid Result
- Use the actual rating and text from each review, never fabricate
- Sentiment must be derived from rating + text analysis together
- Flag any batch where negative reviews exceed 30% as a warning
- Include direct quotes from the most impactful negative reviews
Output
Slack Digest Format:
Trustpilot Sentiment Report for [domain]
- Reviews analyzed: [count]
- Average rating: [X.X]/5
- Sentiment: [X]% positive | [X]% neutral | [X]% negative
- Trending complaints: [theme 1], [theme 2], [theme 3]
- Action needed: [Yes/No] - [reason if yes]
Google Sheets: Each review logged as a row with Date, Reviewer, Rating, Sentiment, Themes, and Review Text columns.
Alert Trigger: If negative sentiment exceeds 30%, include a warning banner in the Slack message with the top 3 negative quotes.
Example Usage
Try asking:
- →"Track sentiment for shopify.com and post weekly to #brand-health"
- →"Pull this week's Trustpilot reviews for our domain and log to my tracking sheet"
- →"Are our Trustpilot reviews getting worse? Check the last 3 pages and alert #cs-team if so"