The Publisher Traffic Numbers That Went Round Last Week Were Wrong. The Corrected Ones Still Change Your Plan.
The publisher traffic figures the industry reposted last week were wrong, the corrected ones are smaller, and they are still bad enough to change how a retailer plans.
The Wall Street Journal reported on 21 July that Reddit had discussed ending Google's access to its content for artificial intelligence use as its reported $60 million a year licensing deal comes up for renewal. Reddit stock fell around 8 per cent the following day. The same report named USA Today Co., Politico, Reuters, the Economist and People Inc. as reassessing the relationship.
The original article carried Semrush traffic figures that turned out to be inaccurate. The Journal corrected them on 22 July. By then a dozen sites had reprinted the first set, and it is still circulating.
The math, on corrected United States organic search traffic, June 2025 against June 2026: USA Today's national paper down 18 per cent. Politico down 20 per cent. CNN and Business Insider down 31 per cent each. The version still doing the rounds claimed above 40 per cent for some titles. An 18 per cent fall is a channel under pressure. A 40 per cent fall is a channel in collapse. They lead to opposite decisions about whether to keep funding organic search this year, and the wrong one is in wider circulation than the right one.
Also worth stating plainly, because the headlines imply otherwise: no publisher has actually blocked Googlebot. Every named company is reassessing. That is negotiating posture ahead of contract renewals and regulatory deadlines, not a decision.
Our read: this is not search engine optimisation ending. The publishers hurt worst are the ones whose entire product was the click, and a retailer's product is a sofa. The transferable read is concentration risk. If the largest content owners on earth are openly hedging against dependence on a single traffic source, a business with one route to demand has the same structural problem at a smaller scale. The response is owned demand, meaning email, existing customer lists, repeat purchase and direct traffic, plus being citable through structured data and clear factual statements so you appear inside the answer when the click does not happen.
Which raises a question most teams cannot answer. If Google sent you zero clicks next quarter but cited you in every answer in your category, would your pipeline be better or worse?
Sources: Wall Street Journal, 21 July 2026, corrected 22 July 2026; Semrush data as reported; Search Engine Roundtable. Last verified 27 July 2026.
Written by P&C. We are a marketing accountability firm run by Yasmin and Arsen. We built the 5-Stage Pipeline Audit and we read platform changes so owner-operators do not have to. Connect on LinkedIn