Everyone Wants to Block Google. The Data Says Don't Panic.

Pull up any trade-press comment section this week and you'll find the same fantasy playing out: publishers threatening to just shut Google out. Cut off the crawler. Stop feeding the machine that stopped feeding you back. USA Today's CEO said it out loud — "enough is enough" — and Reuters, Politico, People Inc., The Economist, and Time are reportedly all weighing the same move. The instinct makes sense. The fear underneath it has a name now: google zero, the idea that AI Overviews are quietly starving the open web of the referral traffic Google used to send it. I get the instinct. I don't think the data backs the response.

The 'block Google' chorus, and the 'google zero' fear behind it

Barry Schwartz at Search Engine Roundtable flagged the story on July 23, working from a Wall Street Journal report that hasn't stopped circulating since. Some of the biggest names in US publishing are seriously discussing blocking Googlebot outright — not de-indexing a section, not tweaking one crawler variant's access, blocking Google, the whole company, from crawling their sites at all. USA Today CEO Mike Reed put it plainly: "It's time to take a stand and say enough is enough." Reuters and Politico are reportedly weighing the same move internally. People Inc. CEO Neil Vogel went further, calling it "100% on the table." The Economist and Time were named as considering it too. Schwartz's site resurfaced the story again in its August 4 roundup of Google webmaster news, which tells you the conversation hasn't cooled off — it's still the one everyone in publishing is having.

Here's the part that should make you pause before nodding along: this isn't a grievance somebody invented for a headline. Semrush data cited in the reporting shows measurable referral-traffic drops at the publishers now weighing this move. Real enough to put a sitting CEO on record about it. We've written before about the citation and click-through damage AI Overviews cause at the query level, and that damage doesn't need embellishing — it's documented, it's uneven across query types, and it lands hardest on exactly the kind of straightforward informational content whole verticals of publishing were built on. Nobody credible is arguing AI Overviews are harmless. The argument that matters is what you do about it.

“A real, painful, concentrated problem does not require an all-or-nothing response. It requires a proportional one.”

What the referral-share data says about google zero

Here's what "blocking Google" actually trades away. TechnologyChecker.io tracks this using Cloudflare Radar's referer data across a large tracked network of sites, and the page, last updated August 1, gives you the number every publisher weighing a block should look at before they flip the switch: Google sent 87.63% of all referral traffic across that network in May 2026. In July, after a full quarter of AI Overviews running at roughly half of all Google searches, that share was 88.08%. Not down. Up, by a fraction, over the exact window the panic narrative says should have shown collapse.

Sit with that for a second. Roughly half of all Google searches now surface an AI Overview, according to the same August webmaster roundup that's been carrying the block-Google story. That's the mechanism publishers are furious about, and they're right to be furious about pieces of it. But the aggregate referral number moving from 87.63% to 88.08% over that exact stretch means Google isn't quietly draining the open web of traffic the way "google zero" implies. It means the losses are real but concentrated — landing hard on specific query types and specific publishers, while Google's overall role as the biggest door into the open web stays basically where it was, maybe a hair bigger. That's not the shape of a platform in collapse. That's the shape of a platform redistributing where it sends people, unevenly, in a way that hurts some sites a lot and most sites barely at all.

Blocking Google is the wrong instrument for a targeted problem

Picture the actual mechanics of what a full Googlebot block does to a newsroom or any content site of scale. It doesn't strip out the AI Overview appearances that are costing clicks on "what is X" queries. It strips out every ranking, every News box placement, every Discover feed appearance, every branded-query click, every piece of the roughly half of searches that never trigger an AI Overview in the first place. You're not amputating the tumor. You're amputating the limb because part of it hurts.

That's the trade Mike Reed, Neil Vogel, and the editors at Reuters, Politico, The Economist, and Time are actually weighing, whether the public framing says so directly or not. Google is, per the same Cloudflare Radar data, sending 88% of all referral traffic across the open web — not 88% of any one site's traffic, 88% of the tracked network's traffic, meaning it's the largest single door into almost every publisher's site, by a wide margin over the next-biggest source. Walking away from that over a problem that's demonstrably concentrated on a subset of query types isn't bold. It's using a decision built for total war on a fight that's actually about territory.

Diversify earned media before you consider google zero as a strategy

The actual fix isn't denial, and it isn't amputation either. It starts with accepting the damage is real, then refusing to let any single platform's referral behavior determine whether your quarter looks good or bad. That's a concentration problem, not a Google problem, and concentration problems get fixed by diversifying, not by cutting off your biggest source of concentration and hoping the smaller ones absorb the shock.

This is where link building and digital PR earn their keep as strategy, not just tactics. A citation profile built across trade press, industry newsletters, podcast mentions, Reddit threads, LLM-cited third-party sources, and syndicated wire pickups doesn't collapse when one platform changes its SERP layout, because no single referral source was ever carrying more than a fraction of the weight. That's the opposite of what most newsrooms and content teams built over the last decade, when Google search was reliable enough that concentrating almost entirely on it looked like discipline instead of risk. It was discipline, right up until the platform it depended on stopped behaving the way it always had.

The same logic applies to how you show up in generative engine optimization work now, not as a replacement channel but as one more leg on a wider stool. Getting cited inside an AI Overview or a chatbot answer isn't the same referral mechanism as a blue link, and it shouldn't be measured as one, but it's still a form of reach that exists independent of whether Googlebot is allowed near your site. Our breakdown of the link building strategy shift prompted by this exact referral anxiety goes deeper into the specific tactics — the point here is the principle underneath them: the fewer eggs in the Google basket, the less any single week's trade-press panic actually threatens your traffic.

None of this means sit still and hope the next Search Engine Roundtable roundup brings better news. It means the fix for a platform sending you less of a specific kind of traffic is building more kinds of traffic, not fewer. USA Today, Reuters, Politico, People Inc., The Economist, and Time are all big enough to survive their own experiment either way. Most sites reading this aren't, and for most sites, walking away from 88% of the open web's referral traffic over a problem the aggregate data says is real but contained is the kind of decision you only get to make once.

See where you are cited today

A free snapshot audit of your rankings and AI citations before we ever talk.