Something broke loose in publishing this week. Executives who spent two decades treating Google as a load-bearing wall in their business are now saying, on the record, that they're ready to knock it down. The trigger is a steep AI search referral traffic decline that two separate outlets measured independently, using the same underlying Semrush data cut two different ways, and arrived at the same conclusion: this isn't a bad quarter, it's a structural shift.
AI search referral traffic decline pushes publishers toward the exit
Nieman Journalism Lab published a piece on July 22, 2026 with a headline that would have sounded absurd two years ago: search traffic has declined so much that some publishers are considering opting out of Google entirely. Not de-optimizing. Not diversifying. Opting out, as in blocking the crawler and walking away from the largest referral source the modern web has ever had. That's the framing from Nieman Lab's reporting, and it's not coming from a fringe blog. It's coming from the people who run the newsrooms.
USA Today CEO Mike Reed put it plainly: "It's time to take a stand and say enough is enough." Some publications, per Semrush data cited by Nieman Lab, were down more than 40% in search traffic between June 2025 and June 2026. Layer in Cloudflare's finding that over half of all web traffic is now bots, plus Cloudflare's plan to default-block multi-purpose AI crawlers starting September 15, 2026 unless a site owner opts back in, and you get an industry that feels like it's losing control of its own distribution twice over: once to AI Overviews eating clicks, once to crawlers eating content without permission. Reddit, meanwhile, is reportedly renegotiating its roughly $60 million a year licensing deal with Google, which tells you the tension isn't confined to news publishers.
“"Turning [Google] off and blocking them entirely is 100% on the table." — Neil Vogel, CEO, Penske Media / People Inc.”
Two outlets measure the AI search referral traffic decline, differently
Here's where you have to read carefully, because two different outlets published two different sets of numbers this same week, both sourced to Semrush, and they are not the same figure reported twice. ppc.land's July 23, 2026 piece, "Reddit and USA Today Face Google Exit as Search Traffic Drops," pulled one cut of Semrush data for June 2025 to June 2026. ghacks.net, publishing the same day, pulled a different cut of the same underlying trend. Treat these as two independent confirmations of a direction, not one number restated twice.
| PUBLISHER | PPC.LAND FIGURE (JUN 2025–JUN 2026) | GHACKS.NET FIGURE (JUN 2025–JUN 2026) |
|---|---|---|
| USA Today | -28% | ~-50% |
| Politico | -20% | -23% |
| CNN | -31% | -25% |
| Business Insider | -43% | over -85% |
Notice the gap isn't small. ghacks put Business Insider's decline at over 85%, more than double ppc.land's 43% figure for the same 12-month window at the same publisher. Different methodology, different time slicing, same publication in freefall either way. ppc.land also reported figures ghacks didn't cover: Reuters down 35%, Washington Post down 44%. And not everyone is losing. In the same window, ppc.land reported The Guardian up 16% and the BBC up 15%, which is the detail that should stop anyone from treating this as "Google broke search for publishers." Google didn't break search uniformly. Something about how AI Overviews interact with specific publishers, content structures, and licensing positions is producing winners and losers, and right now the losers are louder.
Traffic swings across publishers, same 12 months (ppc.land, Jul 23, 2026)
ghacks also cited Pew Research data that explains the mechanism better than any traffic percentage can. Pew found that a traditional link gets clicked in 15% of searches where no AI Overview appears. That drops to 8% once an AI Overview shows up on the results page. And for a link placed inside the AI-generated summary itself, rather than in the traditional results below it, the click rate is just 1%. That's the whole publisher complaint in three numbers: showing up isn't the same as getting clicked anymore, and getting cited inside the AI answer is closer to being acknowledged than being visited.
It's also why the UK's Competition and Markets Authority ordered Google to let publishers opt out of AI features specifically, without losing their regular search visibility as a penalty. Right now, in most markets, that choice doesn't exist: you either accept AI Overviews summarizing your work at a 1% click rate, or you disappear from search results altogether. The CMA order is a regulatory acknowledgment that the current bundle is coercive. It's also a preview of the kind of unbundling that could reshape this whole fight before the year is out.
Google's defense: billions of clicks, and rank still drives citations
Google isn't conceding the point. Per Barry Schwartz's Search Engine Roundtable "Daily Search Forum Recap" on July 24, 2026, Google's position is that its AI features send billions of clicks to websites every week. That's a real number, even if it's not broken out by publisher, and it's Google's standard rebuttal to every version of this complaint: aggregate click volume is up, even where individual publisher click-through is down. The same recap notes that Google Search is now running partly on the Gemini 3.5 Flash-Lite model, and that the EU imposed a €460 million fine on Google for self-preferencing in Search around the same period. None of that resolves the publisher argument. It does show Google is under regulatory pressure on this exact behavior from more than one direction at once, which is worth knowing before you decide how permanent any of this is.
“"It's almost always ranking number one on Google gets you more citations." — Lily Ray, VP SEO & AI Search, Amsive”
That quote, from Lily Ray's ppc.land interview published May 14, 2026, is the analyst-side counterweight to everything above, and it's an important one. Ray's point is that organic ranking is still, right now, the strongest predictor of whether a page gets cited inside an AI answer at all. If that's true, then walking away from Google doesn't just cost you the shrinking click-through Pew measured. It costs you the ranking signal that's still driving most of your AI citation volume in the first place. Blocking Google entirely is a defensible move for a publisher whose traffic is down 85% and whose content Google is summarizing at a 1% click rate with no real alternative distribution. It's a wildly overcorrected move for almost everyone else, including the overwhelming majority of B2B brands who will never see anything close to that decline.
What this means for your link building and digital PR strategy
Something Inc.'s clients are not mega-publishers running on ad-supported traffic at the scale Reuters or Business Insider operate at. Our audience is mid-size to enterprise B2B, where organic and AI-cited visibility feeds a pipeline, not a page-view count. For that audience, the right read on this fight isn't "block Google" and it isn't "this is publisher drama, ignore it." It's the same lesson we drew when we told clients to stop treating Reddit as a reliable AI citation channel after Reddit's own citation share collapsed following a licensing dispute with Google. That wasn't really a story about Reddit. It was a story about what happens when a strategy assumes one source, one platform, one relationship, will stay stable indefinitely. This week's publisher revolt is the same pattern with a bigger, louder triggering event.
No single referral channel earns the assumption of permanence anymore, not Google, not any one publisher relationship in your earned-media or digital PR mix, not any single AI engine's crawler policy. The publishers threatening to block Google are making a bet that they can survive without it. Most B2B brands can't make that same bet, and don't need to. What they do need is a link building and digital PR strategy built on the assumption that any one source in the mix could shrink, get renegotiated, or change its terms with no warning, the same way Reddit's access changed, the same way individual publishers' Google traffic just changed by 20 to 85 percent depending on whose Semrush cut you trust.
Do this next: pull your last twelve months of referral data and find out what percentage comes from Google search specifically, versus AI citations, versus direct earned placements. If any one source is carrying more than a third of your total, that's your concentration risk, and it's the same risk USA Today, Business Insider, and Reddit are all learning about in public right now. Build the next quarter's digital PR plan to bring that number down, not because Google is going away, but because you no longer get to assume it won't change on you the way it just changed on them.
See where you are cited today
A free snapshot audit of your rankings and AI citations before we ever talk.
Tyler leads work at the intersection of SEO and generative engines at Something Inc., helping B2B brands get ranked and cited across every major AI engine.