google.com/goto: the invisible door between Google and your site
In late August 2026, Google placed a door between its search results and your website. Nobody saw it, and that is normal: it is invisible. When someone clicks your link, they no longer come straight to you; they pass through an intermediate Google address, usually in the blink of an eye, and then land on your page. Google confirmed the rollout on 26 August. Which leaves the only question that matters to a small business: what does it change in your numbers? Honest answer: nobody knows yet. And that is exactly why this article exists.
What we know, and what we don't
What is established. The redirect is deployed at scale. Google mentions a measure against abuse; the industry mostly reads it as a device against the mass scraping of results by bots, including those feeding AI tools. It is presented as a protection, not as a change for your visitors, and the visitor personally notices nothing.
What is not. The effect on your analytics. If the pass through the intermediate address alters the visitor's origin information, what the tools call the referrer, part of your "Google" traffic could show up as "direct" in your reports. Conditional tense, deliberately: at the time of writing, no published measurement confirms it. The traffic would not disappear, it would change label. But a label that moves is enough to distort every conclusion drawn from the reports, starting with "SEO no longer works".
Two ways to face an unknown
Faced with a "we don't know yet", there are two reactions. The first: rewriting your site urgently on the strength of a rumour, switching analytics tools, buying an "anti-redirect" service. The second: setting a measurement rule and waiting for the facts. We chose the second, for a simple reason: if the effect exists, it will show in your own data before the press confirms it, and if it does not exist, you will have broken nothing for nothing. The rule fits in three series of numbers.
The measurement rule, to set today
- Your clicks in Search Console, over 28 days. That is the counter on Google's side: the redirect does not pass through it, it stays your reference.
- Your "google / organic" sessions in GA4 or your analytics tool, same period. That is the counter on your site's side, the one the redirect could disturb.
- Your share of direct traffic, same period. That is where mislabelled traffic would end up.
Note all three with the date, then compare every week, same days, same tools. And above all, read them together: if Search Console stays flat while "google / organic" falls in favour of direct, demand has not moved, the label has. You are holding a signal worth investigating. Not yet a cause, a signal, and that is already far more than a shiver.
Why "signal" is not "cause"
The nuance deserves two sentences, because it separates measurement from superstition. A gap between two counters can come from the redirect, but also from a change in your analytics tool, from a campaign sending untagged traffic, from a cookie consent banner configured differently. The signal says where to dig; it does not yet say what to conclude. It is the same reasoning as in our guide to measuring Google changes: never conclude from a single data point, and never repair what you have not first measured. If your curves have already moved and the worry is rising, the five-check protocol comes before any decision.
What not to do, above all
Do not rewrite anything because of this redirect. Do not switch analytics tools "to see". Do not sign up for a service that promises to "bypass" a device Google presents as protection of its own infrastructure. Each of those gestures adds a variable at the precise moment you should have none, and turns a measurable question into permanent fog. The only useful gesture costs ten minutes and three lines of a notebook; it is described in the box above, and in a month it will give you an answer nobody else can give you: your own.
Frequently asked questions
Does the google.com/goto redirect slow my site down for visitors?
The pass through the intermediate address is designed to be imperceptible and the visitor sees nothing of it. It does not modify your site: it changes the path travelled before arriving. The open question is not about speed, it is about how traffic gets labelled in analytics tools.
Will I lose traffic because of this redirect?
Nothing points to a loss of visitors: the redirect leads to you. The documented risk to date is a measurement risk, not a volume risk: part of the "Google" traffic could be counted as "direct" if the origin information gets lost on the way. Hence the rule of three series of numbers, noted now so you can compare later.
My GA4 report shows more direct traffic since late August: is that the proof?
It is a signal, not a proof. Compare with Search Console first: if it stays flat while "google / organic" falls, the shifted-label hypothesis becomes serious. Also check your own recent changes, tags, cookie consent, campaigns. A correlation does not choose its own direction.
Should I change something on my site because of goto?
No. There is no setting on your side that "fixes" or "bypasses" this redirect, and any service claiming otherwise deserves suspicion. The only useful work is measurement: three numbers noted today, compared every week, read together.
More questions? Many answers are already gathered on our FAQ page.
Want a calm reading of your analytics?
We set the baseline, cross Search Console with GA4, and tell you in writing whether your numbers describe a real change or a shifted label. The difference is often worth several months of decisions.
Information current as of 27 August 2026; the redirect's effect on analytics had not been publicly measured at that date and may have been since. Main sources: Google's confirmation of the redirect rollout, 26 August 2026, relayed by the trade press · analyses by Search Engine Land and Abondance, August 2026. Cover photo: Anton Porsche, Pexels.
Written byMaryna Skachek holds a PhD in biology, is a qualified e-commerce project manager and the founder of MariCleo Studio. She writes about digital strategy, content and the professional uses of AI.
Measure rather than guess
The full measurement guide
August's three changes and the whole method: baseline, log, cross-reading.
Read the guide →Traffic drop: the protocol
Five checks in order before touching anything, or paying anyone.
Read the article →Clear message & customer journey
Decisions grounded in your numbers, not in the shivers of the news cycle.
See the service →