Clicks Down 42%, LinkedIn Reach Down 47% — But That's Not the Real Problem

Anxiety has been building in the marketing community in recent months: organic search clicks and social reach are falling sharply. But as Search Engine Journal columnist Greg Jarboe writes, citing blogger Tony Apphoff, most of the numbers are accurate — the conclusion drawn from them isn't.

An analysis of Search Console data across 64 sites by Define Media Group found organic clicks have dropped roughly 42% since Google's AI Overviews rolled out broadly. Pew Research Center data backs this up: only 8% of users click through to an organic result when an AI Overview appears, versus 15% when it doesn't.

LinkedIn's situation is even sharper: after the platform rebuilt its feed around 360Brew, a 150-billion-parameter model that weighs a post's expertise and read time rather than likes and comments, organic reach dropped 47% over the year. Pangram Labs analyzed more than 1 million LinkedIn posts and found 41% were entirely AI-written — higher than on any other platform. That's part of why LinkedIn rolled out a "this looks like AI slop" button on July 30.

The problem isn't confined to content: Mercer's "2026 Global Talent Trends" study found the share of employees worldwide worried about losing their job to AI has climbed to 40%, up from 28% two years ago. Challenger, Gray & Christmas data shows AI was cited as a factor in more than 101,000 layoff announcements in the first half of 2026 — roughly double the count for all of 2025.

Even so, Apphoff calls these numbers a "structural shift," not a "decline" — the difference being that in a decline, there's no version of your work worth preserving, while in a structural shift the underlying function still matters, it's just changing shape. As he puts it: "The machines aren't taking anything from you that you weren't renting out." Curiously, one of the figures Apphoff cites — an 85% traffic collapse at Business Insider — was actually corrected by the Wall Street Journal on July 22, putting the real figure at 43% (28% for USA Today, 44% for the Washington Post) — a reminder of how risky it is to spread numbers unchecked in a fast-moving AI era.

As practical advice, Jarboe recommends three steps: first, track "function" metrics instead of channel metrics like clicks and reach — how often AI systems reference your content, what conversion your email subscribers deliver; second, strengthen owned assets — email lists, first-party data, and direct customer relationships; third, judge content by citation, retention, or getting an answer, not by ranking.

The bottom line: in 2026, every platform that's changed — Google included, LinkedIn included — hasn't stopped rewarding value, it's just changed what it rewards. Whoever adapts to these new criteria will be well ahead, 12 months from now, of those still arguing with the numbers.