Extract Website Data to Google Sheets with GPT-4o-mini

Turn any webpage into structured Google Sheets rows—automatically

This n8n workflow extracts specific data from a URL using OpenAI (GPT-4o-mini) and appends the results as clean, structured rows directly into Google Sheets—based on what you request in an n8n form.

What this workflow does

• Collects inputs via an n8n form: you submit a target page URL and a plain-English extraction request (which fields you want).
• Fetches the webpage content: it downloads the page using an HTTP Request with a browser-like User-Agent header to reduce fetch issues.
• Prepares text for extraction: it extracts readable text from the page HTML to reduce noise before sending anything to the model.
• Extracts structured data with GPT-4o-mini: the workflow sends the page text plus your instructions to OpenAI and requests a strict JSON array of row objects.
• Validates and converts JSON to rows: it parses and validates the AI response as JSON, then converts each object into an individual row item.
• Appends to Google Sheets: it adds the extracted rows to your selected spreadsheet and sheet tab with automatic column mapping based on the returned field names.

Use cases

• Competitive research: pull pricing, feature lists, or availability details from product pages into a spreadsheet.
• Lead enrichment: extract specific profile attributes from public webpages and organize them for review.
• Content tracking: capture targeted fields from pages and maintain an up-to-date dataset in Google Sheets.
• Ops and analytics: transform messy web data into structured rows without manual copy/paste.

Technical details

• n8n Form Trigger for URL + extraction instruction intake
• HTTP Request to fetch the page with a browser-like User-Agent
• HTML/text extraction to reduce noise before processing
• OpenAI (LangChain) with GPT-4o-mini to produce a strict JSON array
• Code steps for JSON parsing/validation and row conversion
• Google Sheets append with automatic column mapping

Setup is straightforward: add your OpenAI credential, select the chat model in the OpenAI (LangChain) node, connect Google Sheets credentials, choose the destination spreadsheet/sheet tab, and start submitting pages through the form.