# Google is putting a redirect between its results and your site

Google confirmed it is wrapping search result links in google.com/goto. Here is what is documented, what is still unknown, and the one number worth watching.

Published: 2026-08-27T10:00:00.000Z

Canonical: https://thatseoagent.com/en/blog/google-goto-redirect

Since June 2026, people have been reporting Google search results whose links no longer point at
the destination. They point at `google.com/goto?url=…`, with the real address wrapped in an
encoding nobody outside Google has decoded. Click it and you still land where you expected, one
hop later.

Search Engine Roundtable reported it as a test in July. By August 26, Search Engine Land had
Google's confirmation and reports that the rollout was close to complete: Derek Perkins of the
rank tracker Nozzle said he was seeing wrapped links in nearly 100% of results across several
residential IP providers.

This post separates three things: what is established, what is still guesswork, and what actually
changes for anyone doing SEO. All three lists are shorter than the commentary suggests.

## What Google said

One sentence, from a spokesperson, and it is worth reading closely:

> We have a long history of deploying technical measures against evolving forms of abuse, and we
> regularly take steps to protect our services and users.

That is a statement about abuse and scraping. It is not a statement about ranking, about
attribution, or about anything a site owner is supposed to do. Google added that publishers
should watch their own data for anomalies and ask their tool vendors what they plan to do, and
that Search Console data is unaffected.

Note what is missing. There is no Search Central documentation for this, no help center page, no
changelog entry. The primary source for the whole story is a press statement, which is a thinner
foundation than most of the coverage implies.

## What is established

**The link shape changed.** Result links pass through `google.com/goto?url=[encoded]` before
reaching the destination. Everyone who has looked at the encoding describes it as a custom
base64-like format, and none of them has decoded it.

**The rollout is broad.** Multiple independent observers report it at or near 100% of results,
not a slice of a test.

**The target is programmatic collection.** A wrapped link cannot be read off the page; a scraper
has to follow every hop to learn where a result goes. On a page with dozens of results, that
multiplies the cost of harvesting a SERP by roughly the number of results on it. That is the
whole mechanism, and it lands on rank trackers, SERP APIs, and the AI systems that use scraped
Google results as a substrate.

## What is not established, and is being reported as though it were

Here is the honest state of the most-repeated claim: **nobody has published a header-level
analysis of the live behavior.** Not the status code, not the `Referrer-Policy` on the hop, not
what the destination server actually receives in `Referer`.

Everything you have read about organic sessions collapsing into `(direct) / (none)` is an
inference from that gap, not a measurement of it. It may turn out right. It is not currently
evidence.

What we can do is read the specs, and say plainly that this is a reading rather than a fact.

The Referrer Policy specification settles the referrer separately for each request in a redirect
chain, and a `Referrer-Policy` header on a redirect response replaces the policy for the rest of
that chain. Under the modern browser default, `strict-origin-when-cross-origin`, a cross-origin
request carries only the origin: `https://www.google.com/`, with no path and no query. That is
already what your server sees today when someone clicks a result.

So if the hop is an ordinary server-side redirect with no unusual policy header, the destination
most likely sees exactly what it saw before. GA4 classifies a session as Organic Search when the
source matches its list of search sites or the medium is `organic`, and an origin-only referrer
of `google.com` still satisfies that.

The version that would hurt is a hop that sets `Referrer-Policy: no-referrer`, or a
JavaScript-driven redirect that yields nothing. Then organic clicks arrive with no referrer at
all, and GA4 has no basis for anything but `(direct) / (none)`.

Both are possible. Only one of them is being written about.

## What it changes for SEO

**Nothing about ranking.** No ranking system changed, no guidance changed, nothing on your site
caused this and nothing on your site fixes it. Any advice you are given that involves editing
your pages in response to `goto` is advice about a mechanism that does not touch your pages.

**Third-party position data gets more expensive and probably noisier.** If following every link
is the only way to resolve a result, rank tracking costs more per SERP, and vendors will
economize somewhere: fewer keywords, less frequent refreshes, more estimation. Your tracker's
numbers were always a sample of a personalized surface. The sample is about to get harder to
take.

That points the same direction as Google's own published guidance on third-party tools: prefer
the data that comes from Search itself. Search Console reports clicks Google measured on its own
side, and a redirect Google runs cannot corrupt them.

**AI answer engines that lean on scraped Google results lose a cheap input.** Whether that
changes what those systems say about your brand is unknown and probably unknowable from outside.
We would not claim a direction.

**The attribution question is real but unquantified.** The correct response is a measurement, not
a remediation.

## The one number to watch

If you want to know whether this touched you, one number answers it, and there is a second check
worth running while you are in there.

In GA4, pull Session source / medium for the last 90 days, broken down by week, for your top
landing pages. Record the ratio of `google / organic` sessions to `(direct) / (none)` sessions
each week. If the hop strips referrers, that ratio moves, and it moves on a specific week rather
than drifting. If it does not move, nothing reached you.

If you have server logs or edge analytics, search the `Referer` header for `google.com/goto`. The
expected result is zero hits, because a browser under the standard policy sends the origin and
not the path. A non-zero count would mean the full URL is being passed, which is a different and
more interesting finding than the one everyone is predicting.

Neither is a project, and the first one is something this product does for you. With Search
Console and GA4 connected to That SEO Agent, you can ask for it in plain language: sessions by
week and by session source/medium, last 90 days. It queries the Analytics Data API and returns
the weekly split, which is the ratio above. We ran that report against a connected site while
writing this, and the ratio had not moved; organic held its share through the weeks the rollout
was completing. One site is not evidence about the web. It is evidence that the check is cheap.

The second check we cannot do for you, and it would be dishonest to imply otherwise. We read
Search Console, GA4 and PageSpeed. We do not read your server logs, and the `Referer` header
lives only there.

Do both now, before the rollout finishes reaching wherever it is going, so you have a baseline
that predates it.

## The precedent

We have been here before. In October 2011 Google moved signed-in search to HTTPS and stopped
passing the query in the referrer, and the industry lost keyword-level organic data permanently.
`(not provided)` is still with us fifteen years later.

Two things about that are relevant. First, the loss was real and no workaround ever recovered it,
so "wait and see" is not always the wise posture. Second, the thing that replaced the lost data
was Search Console, which is where the query-level view lives now. If `goto` turns out to cost
you referrer fidelity, the same substitution is available, and the query-level view it gives
you is better than a referrer ever was for judging a page.

## Frequently asked questions

:::faq
### Do I need to change anything on my site?
No. This is a change to the path a click takes between Google and you. There is no markup, no
header, and no configuration on your side that participates in it.

### Will my organic traffic show up as direct now?
Nobody outside Google can say yet, because no header-level analysis of the live behavior has
been published. Under the standard referrer policy an ordinary redirect would preserve a
`google.com` origin and your classification would survive. Measure your own weekly ratio rather
than trusting either prediction.

### Does this affect Search Console?
No. Search Console reports what Google measured on its own side of the click, so a redirect Google
operates cannot distort it. Google said as much when confirming the change.

### Why is Google doing this?
The stated reason is protection against abuse of its search results. The practical effect is that
a result's destination can no longer be read off the page, which raises the cost of scraping
Google at scale for rank trackers, SERP APIs, and AI systems.

### Should I switch rank trackers?
Not on this news. Every tool in the category faces the same constraint at the same time. The
useful question for your vendor is what their coverage and refresh rate will look like in three
months, and whether they will tell you when a number is estimated rather than observed.
:::

## Sources

- [Google Search Rolling Out google.com/goto Tracking Parameters](https://www.seroundtable.com/google-search-goto-tracking-41957.html) — Search Engine Roundtable's reporting, the first sustained coverage
- [Google confirms deploying goto URL redirects to search results links](https://searchengineland.com/google-confirms-deploying-goto-url-redirects-to-search-results-links-485926) — carries Google's statement verbatim, and the rollout observations
- [Referrer Policy](https://www.w3.org/TR/referrer-policy/) — §8.2 on redirects, and the definition of `strict-origin-when-cross-origin`
- [\[GA4\] Default channel group](https://support.google.com/analytics/answer/9756891) — the exact conditions for Organic Search and Direct
- [Redirects and Google Search](https://developers.google.com/search/docs/crawling-indexing/301-redirects) — how Google treats each redirect type, and notably silent on referrers
- [Google's guidance on using third-party SEO tools, services, and advice](https://developers.google.com/search/docs/fundamentals/third-party-seo) — why Search Console is the recommended primary source