Collect recent customer reviews from a public review site, turn them into clean data, and store results in Google Sheets with sentiment labels. Great for marketing and CX teams that track brand reputation and want quick insights without manual copy and paste.

The run starts on a manual trigger. You set the company slug and max pages to scan. The workflow calls the review pages with pagination, grabs review links from the HTML, and splits them into single items. It checks Google Sheets for matching IDs to avoid duplicates, then opens each review page to pull full content. A DeepSeek model extracts fields like author, rating, title, text, and date. An OpenAI model scores sentiment. The results are written back to Google Sheets in a structured row.

For setup, you need a Google Sheets account plus API keys for OpenAI and DeepSeek. Add the company slug and choose how many pages to scan. Expect faster reporting, less manual work, and a steady view of customer mood. Use it for daily monitoring, campaign impact checks, and weekly summaries. It gives clear data your team can act on the same day.

• Manual trigger with simple controls for company slug and page limit
• HTTP pagination to fetch recent review pages with a safe request interval
• HTML parsing to collect review links from each page
• Item splitter and limiter to manage volume and testing
• Google Sheets lookup to skip reviews that are already saved
• Single review fetch to capture full review content from its page
• AI powered information extraction with DeepSeek for author, rating, title, text, and date
• Sentiment analysis with OpenAI to label and score each review
• Structured Google Sheets update to store IDs, content, and sentiment in one place

• Reduce manual review collection from 2 hours to 10 minutes
• Automate 90 percent of copy and paste work for review tracking
• Improve data accuracy by 80 percent with duplicate checks
• Handle up to several pages of reviews per run without extra effort
• Connect Google Sheets and AI models in one flow for easy insights

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with Google Sheets, OpenAI and DeepSeek. See the Tools Required section above for links to create accounts with these services.
3. In the n8n credentials manager, open any Google Sheets node, click the Credential to connect with dropdown, choose Create new credential, and follow the steps to connect your Google account with OAuth2. Name the credential clearly, for example Google Sheets Reviews.
4. For OpenAI: open the OpenAI Chat Model node, create a new credential, and paste your OpenAI API key from your OpenAI account API page. Save the credential with a clear name.
5. For DeepSeek: open the DeepSeek Chat Model node, create a new credential using API Key, paste your DeepSeek API key, and set the base URL to https://api.deepseek.com/v1 as noted in the template.
6. Open the Set Parameters node and set company_id to the company slug from the review site URL and choose max_page for how many pages to scan.
7. Open the Get reviews HTTP Request node and confirm the URL uses the company_id. Keep pagination enabled with the page parameter and the request interval to avoid rate limits.
8. Open the Extract HTML node and test the CSS selector to confirm it returns a list of review links. Run a partial test to see results in the Split Out node.
9. In the Google Sheets nodes, select your spreadsheet and sheet. Make sure the sheet has an Id column and any columns you plan to write like Author, Rating, Title, Text, Date, and Sentiment.
10. Check the Get rows filter so it looks up the review Id derived from the link and returns no rows when the review is new.
11. Open the Information Extractor node and review the required fields. Run a test on a single review to confirm it returns author, rating, title, text, and date correctly.
12. Open the Sentiment Analysis node and confirm the model is connected. Test with sample text to see a clear sentiment label or score.
13. Run the full workflow with the manual trigger. Verify only new reviews are added and that sentiment and fields map to the right columns. If you get no links, adjust the CSS selector. If you see rate limits, increase the request interval or lower max_page.