The Scenario
You're a researcher at a B2B services firm and a new vendor just made it to your shortlist. Their documentation site is thorough — too thorough, actually. You need to read through up to 30 pages of it to evaluate their integration depth, and whoever asked for this evaluation wants it by end of day Thursday.
Cell A1 of your Google Sheet has the vendor's docs site URL. You need each sub-page's URL and body content written into the sheet, one row per page, so you can ctrl+F through the output and tag which pages cover the capabilities you're evaluating.
The bad version:
- Click into the docs site. Find the first sub-page. Copy the URL into a cell. Copy the body content into the next cell.
- Navigate back. Find the second sub-page. Repeat.
- Lose track of which pages you've already visited somewhere around page 12.
There's a quarterly review on Friday and this is one of three vendors you're evaluating this week.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent inside your Google Sheet. It takes the seed URL, calls Tavily to crawl sub-pages, and writes each page's URL and content as a new row — you decide how many pages to include.
Crawl the website starting at the URL in cell A1 and extract content from up to 30 pages — write each page's URL into column A starting at row 2 and its body text into column B. Stop when you reach 30 pages or run out of discoverable sub-pages.
What You Get
- Rows 2 onward populated with one page per row: column A holds the URL, column B holds the clean body text.
- Crawl stops at 30 pages or when Tavily exhausts reachable sub-pages — whichever comes first.
- Column A, row 1 stays untouched (your seed URL remains in place).
- Any page Tavily couldn't fetch gets a "FETCH ERROR" entry in column B with the URL still recorded in column A.
What If the Data Is Not Quite Ready
The seed URL in A1 redirects — you need to use the canonical URL instead
Resolve the redirect from the URL in cell A1 to its final destination, then crawl that destination for up to 30 sub-pages. Write each page URL to column A and body text to column B starting at row 2.
You only want pages that contain specific keywords (e.g., "API", "webhook", or "integration")
Crawl up to 30 pages starting at the URL in cell A1 using Tavily. Write each page URL to column A and body text to column B, but only include rows where the body text contains the word "API", "webhook", or "integration". Skip pages that don't mention any of those terms.
You need a one-sentence summary of each page instead of the full body text
Crawl up to 30 pages starting at the URL in cell A1 using Tavily. For each page, write the URL to column A and a one-sentence summary of the page content to column B. Limit to 30 pages total.
Crawl, filter by keyword, summarize, and flag missing pages — all in one shot
Resolve any redirect from the URL in A1. Crawl up to 30 sub-pages using Tavily. Write each page URL to column A and a one-sentence summary to column B. Only include pages that mention "API", "webhook", or "integration". Write "FETCH ERROR" in column B for any page Tavily couldn't retrieve. Stop at 30 rows.
One prompt handles redirect resolution, crawling, keyword filtering, and summarization in sequence.
Try It
Get the 7-day free trial of SheetXAI and open any Google Sheet with a seed URL in cell A1, then ask it to crawl the site and write each page into a new row using Tavily. See also mapping full competitor site structure and the hub overview.