Google Search implements google.com/goto redirect URLs to combat SERP scraping

Google Search replaces direct URLs with opaque redirects

Google Search is now rewriting organic result links to use google.com/goto?url=... instead of exposing the destination URL directly in the HTML. This change, which became consistent for logged-out and private browsing sessions in late August 2026, replaces the destination URL with a custom, Google-specific encoding that cannot be decoded offline.

To retrieve the actual destination URL, a client must now make a request to the /goto endpoint and read the Location header in the HTTP response. This shift ensures that the real URL is only revealed upon a request to Google's servers, rather than being available for bulk extraction from the search engine results page (SERP) HTML.

Strategic goal: Increasing the cost of SERP harvesting

Google is implementing these redirects as part of a broader effort to combat automated SERP harvesting, specifically targeting AI crawlers and SEO scrapers that build independent indexes by bulk-extracting result URLs.

By removing plaintext links, Google eliminates the ability for scrapers to parse thousands of URLs from a single HTML page without further interaction. Under the new system, every single result requires a separate request back to Google to resolve the destination. This makes scraping significantly slower, noisier, and easier for Google to detect via patterns of sequential link resolution.

This update follows other restrictive measures, including:

• The removal of the &num=100 parameter, which previously allowed users to see 100 results per page.
• The tightening of BotGuard and SearchGuard technical protections.
• Legal actions against SERP providers like SerpApi.

Technical implications for developers and users

Impact on automated tools

For developers building search-based tools, the goto format necessitates a change in how URLs are extracted. The recommended technical approach is to perform a HEAD request to the /goto URL and read the Location header without following the redirect to the final destination. This prevents the crawler from unnecessarily loading the target page while still obtaining the destination URL.

Impact on end-users

While the end result for a human user clicking a link remains the same, several side effects have been noted by the community:

• Broken Link Sharing: Users can no longer right-click and copy a direct link from the SERP to share with others; instead, they copy a google.com/goto redirect link.
• Increased Latency: Some users report a perceivable delay in loading pages due to the the redirect hop.
• Loss of Transparency: Users cannot hover over a link to see the destination URL before clicking, reducing the ability to verify the target site's legitimacy.

Community perspectives and alternatives

The shift toward obfuscated URLs has sparked significant criticism among power users and developers, who view it as a move toward a "closed" web where Google acts as an oracle rather than a directory.

"I want an extension that will let me resolve the true url without visiting the site, or even rewrites the entire page to show the url."

Many users expressed a preference for alternative search engines that maintain more traditional indexing behaviors. Frequently cited alternatives include:

• Kagi: A paid search engine praised for its lack of ads and user-centric approach.
• Brave Search: Noted for having its own independent index.
• DuckDuckGo: Used by those seeking privacy, though some users noted a decline in result quality.
• Mojeek: Cited as a viable API-based alternative for agents.

Some users suggest that the udm=web parameter may still provide the "old" behavior of returning traditional web results, though this remains an undocumented flag.