Collect product details from web pages and save them into Google Sheets with the help of AI. Ideal for price tracking, promo audits, and competitor research across many product pages. Teams get a fast way to turn messy web content into clean rows without manual copy and paste.

Here is how it runs. A manual test starts the run, then a Google Sheets node pulls a list of URLs. Each URL is passed to ScrapingBee to capture a full page screenshot. An AI agent powered by Google Gemini reads the screenshot and extracts fields like title, price, brand, and promo flags. If the image is not enough, the agent calls a helper workflow that fetches the HTML with ScrapingBee, converts it to Markdown, and tries again. The Structured Output Parser shapes the results into a clean JSON array. Split Out turns the array into single items, and the final Google Sheets node appends rows in the Results tab.

Setup needs a Google Sheet with List of URLs and Results tabs, a ScrapingBee API key, and a Google Gemini API key. Expect big time savings on weekly price checks and catalog updates. Use it for ecommerce merchandising, promo tracking, and competitive monitoring. Map the columns to match the parser schema, then test with a small list and scale up once results look correct.

• Reads a list of URLs from Google Sheets and writes clean results back to a Results tab
• Captures full page screenshots through ScrapingBee for reliable visual parsing
• Uses a Google Gemini agent to extract product fields from screenshots
• Falls back to an HTML scraping tool that fetches page HTML and converts it to Markdown if needed
• Formats outputs with a Structured Output Parser aligned to an ecommerce schema
• Splits arrays into single items so each product becomes one row in Sheets
• Simple manual trigger for testing and safe runs before scaling

• Reduce manual copy and paste from hours to minutes for large product lists
• Automate up to 90 percent of repetitive scraping tasks with AI and fallback HTML parsing
• Improve data quality by using a fixed schema for titles, prices, brands, and promos
• Handle 10 times more URLs by batching them in Google Sheets
• Connect Google Sheets, ScrapingBee, and Google Gemini in one flow

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, ScrapingBee and Google Gemini. See the Tools Required section above for links to create accounts with these services.
3. Create a Google Sheet with two tabs named List of URLs and Results. Add a header row in Results that matches the fields product_title, product_price, product_brand, promo, and promo_percentage.
4. In the n8n credentials manager, create Google Sheets credentials. If using a Service Account, share the Google Sheet with the service account email. If using OAuth2, follow the on screen steps to connect your Google account.
5. Open the Google Sheets nodes. Set the Document ID and select the List of URLs tab for the input node and the Results tab for the output node. Click Execute Node to confirm access.
6. Open the ScrapingBee nodes. In the Credential to connect with dropdown, click Create new credential if available, or add your API key as a secure field. Ensure the query parameter api_key is set and screenshot_full_page is true on the screenshot node.
7. Open the Google Gemini Chat Model node. Create a new Google Gemini credential and paste your API key from the Google AI Studio or the official API page. Keep the model set to gemini 1.5 pro latest.
8. Review the Structured Output Parser node and confirm the JSON schema matches your Results columns. Adjust the schema or columns if you need different fields.
9. Open the Set fields node and confirm url maps to {{$json.url}} from the List of URLs sheet.
10. Check the HTML based scraping tool workflow. It should fetch HTML with ScrapingBee and convert it to Markdown before returning content to the agent.
11. Run a small test with the manual trigger. Watch the Vision based Scraping Agent, then Split Out, then Google Sheets Create Rows to confirm rows are added.
12. Troubleshoot common issues: if rows are blank, refine the system prompt or confirm the page loads in ScrapingBee. If you see auth errors, recheck API keys and Google sharing. If columns do not match, update the parser schema or sheet headers to align.