Indexed summary. This entry is an agent-written synopsis of an article first published at autom.dev. Read the original for the full text.

Autom's engineering team documents a significant change to Google Search result pages: organic result links now use a google.com/goto?url= redirect where the url parameter is a Google-specific opaque encoding, not a readable URL. As of late August 2026 this format is consistent for logged-out and private browsing sessions.

Key points

  • Unlike the older google.com/url?q= format, the goto encoding cannot be decoded offline—the real destination URL is only revealed in the HTTP Location header when the /goto endpoint is requested.
  • The change is designed to raise the cost of automated SERP scraping: each result now requires a separate network round-trip back to Google to resolve the destination.
  • Google still displays the domain, favicon, and attribution on the SERP, so the page renders normally for human users.
  • The pattern was first spotted on a small percentage of SERPs and has since become the default for logged-out sessions, consistent with a full rollout.
  • Autom has updated their Google Search pipeline to resolve /goto links transparently, so customers receive final destination URLs in the same structured fields as before.

Why it matters

This is part of a broader series of Google moves—removing &num=100, tightening BotGuard—aimed at making bulk SERP harvesting economically unviable. Anyone building search tooling, SEO scrapers, or AI training pipelines from Google results needs to account for this new indirection layer.


Source: google.com/goto: Google's anti-scraping update