The Algorithmic Earthquake and What Comes After It

The ground shifted. And most marketing departments missed it entirely.

Imagine the most profitable main street in a major city. Ten incredibly popular stores that get all the foot traffic. Overnight, without warning, nearly two of those stores just vanish from the city map entirely. Not moved to a side street. Erased. That is exactly what happened in December 2025 to nearly 15% of the top 10 search results on the internet.

This was not a punishment. Google called it a "recalibration of relative usefulness." The algorithm did not decide your website was bad. It simply decided that a completely different set of websites were much, much better at answering what people were actually asking. And it acted on that decision at a scale never seen before.

We are now living in a zero-click world. And it is accelerating fast.

For 20 years, the internet ran on an unwritten contract: publishers spend money to create great content, Google indexes it, and in exchange, Google sends the publisher human traffic to monetize. That contract is now broken — permanently.

The fundamental flow of human attention on the internet is being rerouted. People are getting answers without ever visiting the website that created those answers. Here is what that looks like in hard numbers:

Your marketing team spent real money ranking that "how to choose payroll software" article at the top of Google. The AI Overview reads it, summarizes the best parts, delivers it directly to the user — and your link sits at the bottom, unclicked. Organic click-through rates on informational queries have dropped 61%, from a historical average of 1.76% down to a fraction of that.

And this is just traditional search. On ChatGPT — which now handles 12% of Google's total global query volume — the damage to traffic is even more extreme. ChatGPT sends 190 times less traffic to external websites than Google does, because it is built from the ground up to resolve queries internally, not hand users a map of ten places to look.

PR is no longer a soft metric. It is the machine's curriculum.

Here is the data point that should stop every CMO cold: 90% of ChatGPT's citations come from pages that don't rank in Google's top 20 traditional results. Read that again. Your company could be invisible in traditional search — buried on page three or four — and ChatGPT might be citing you as the industry authority every time someone asks a relevant question.

Why? Because large language models — the AI brains behind ChatGPT, Perplexity, and Google's Gemini — don't train on what ranks. They train on what is credible. And credibility, to a machine, looks exactly like rigorous, third-party, earned media coverage.

Consider PR Newswire. Its human traffic has actually dipped — fewer people browse dry press releases for fun. But behind the scenes, PR Newswire receives over 113,000 direct referrals from ChatGPT every single month. The AI is mining structured, verified, timestamped corporate announcements to build its knowledge graph of objective facts about the corporate world.

Gartner predicts that corporate budgets dedicated to PR and earned media will double by 2027. The smart money already understands: you are not just buying press coverage. You are buying your way into the training data that shapes what the AI believes to be true about you.

How AI judges authority today — and what actually gets you cited.

Before a CMO doubles their PR budget, they need a surgically precise understanding of what actually works in the post-March 2026 landscape. There are two non-negotiable mandates that govern whether your content gets rewarded or ignored.

What about AI-generated content? This is the most common fear in boardrooms right now — and the answer is more nuanced than most people expect. Google does not punish the use of AI. It punishes unreedited, lazy scale.

And the metrics that measure success have to change too. The old KPIs are dead. If your SEO director walks in and says "We got 500 backlinks this month," the right response from a CEO is: "I don't care." A thousand irrelevant links from obscure blogs do nothing to teach an AI to trust your brand.

The new KPIs that actually matter:

Back to the future. The two-track playbook for what to do on Monday morning.

The irony of this entire technological revolution is that to survive the bleeding-edge future of AI, marketing departments have to look backward. Not to last year. Back to 2014. The era right before hyperautomation took over — before cheap AI content spinners allowed a single marketer to generate a thousand mediocre articles a minute. Back when there were no shortcuts to authority.

In 2014, you won online by doing genuinely grueling work: being a legitimate expert, building a real trustworthy brand, writing deeply researched content that actually helped people, picking up the phone and pitching skeptical journalists based on the merit of your ideas. That is exactly what wins again today.

But the transition between eras is not instantaneous. The old world hasn't vanished entirely while the new world takes over. Marketing departments must explicitly split their work into two parallel tracks — and fund both deliberately.

The difference between these two tracks is not volume — it is context. A backlink inside a spring-cleaning lifestyle article that briefly mentions your brand in paragraph eight? Massive Track One win. Zero Track Two value. The AI searching for the best commercial pizza oven is not reading spring-cleaning tips. It needs your brand cited in an article explicitly titled "The Definitive Guide to High-Volume Commercial Pizza Ovens in 2026."

The five principles of the new playbook: