How scraping works
You select a column of URLs in the Scrape tab, choose an output column, and start the run. Each URL is fetched, the visible text is extracted, and the result is written back to your sheet as it completes. On Auto (the recommended setting) the work runs on ReplyLabs servers with no setup. The one exception is the Google Sheets Scrape provider, which runs the fetch through your own Google Sheets session (see below).
Scrape providers
Pick the provider in the Scrape tab settings.
- Auto (recommended): Starts with ReplyLabs Scrape, our fast in-house scraper for company websites ($0.002 per URL), and uses an approved fallback when a page is blocked. Best default for most runs.
- Google Sheets Scrape:Runs the fetch through your own Google Sheets session, so requests originate from Google's IP infrastructure and from your daily Google quota. Slower, but useful as a fallback for sites that treat Google traffic more kindly. $0.002 per URL.
- Jina Reader: Routes each URL through Jina Reader, which renders the page in a real browser before extracting text. Better for JavaScript-heavy sites, single-page apps, and modern marketing pages. $0.005 per URL. Scale plan: bring your own Jina API key to make it free.
- Firecrawl: Routes each URL through Firecrawl for deeply rendered, structured output. Best for complex SaaS dashboards and pages where Jina returns incomplete results. $0.0075 per URL. Scale plan: bring your own Firecrawl API key to make it free.
What works and what does not
What all providers handle well: static company sites, marketing pages, blog posts, product pages, and most public web content.
What none of them can reach:
- Pages behind a login (any provider).
- LinkedIn /in/ personal profile pages: auth-walled by LinkedIn for everyone.
- Sites that actively block server-side scrapers (Cloudflare Bot Fight Mode, etc.).
LinkedIn /company/ pages (public organization profiles) work with the Jina or Firecrawl providers.
Best practices
Leave the provider on Auto for most runs. Switch to Jina when you see empty or partial results on JavaScript-rendered pages, and use Firecrawl for the most complex pages. Google Sheets Scrape is a useful fallback for sites that respond better to Google traffic. ReplyLabs respects robots.txt where applicable. Do not use scraping to extract data from sites whose terms of service prohibit it.