The distribution of online traffic is shifting rapidly: Google Search is sending significantly fewer visitors to publisher websites as Google increasingly answers queries directly on its own platform using AI.
Chartbeat data shared with Axios shows Google Search traffic to publishers fell 34% over the past year. The impact is uneven: over the past two years, small publishers lost 60% of search referrals, medium publishers lost 47%, and large publishers lost 22%.
This is consequential because search visitors typically arrive with specific intent and are therefore more valuable than casual visitors from social networks.
Platform saturation and AI-generated content
At the same time, platforms such as LinkedIn and Reddit are becoming crowded with AI-created content, making it harder for genuine material to stand out and for readers to distinguish real from fake. In response, Substack announced a partnership last week with AI-detection firm Pangram, explicitly calling out LinkedIn and warning that platforms that reward fakeness "will create a race to the bottom."
How publishers are adapting
Many online-focused professionals are already preparing for a shift toward large language model (LLM) environments similar to ChatGPT. The industry is developing the concept of GEO — the term for shaping how content appears inside LLMs such as Claude and ChatGPT — which functions as a modern version of SEO. GEO practices are still early-stage and opaque, so time is of the essence.
Axios reports that its Smart Brevity format — putting the most important point first, followed by context and information ordered by importance, often in bullets — performs well with AI. Axios is among the top publishers surfaced by leading LLMs and is working with companies to see whether others can achieve similar visibility.
Practical steps recommended for publishers
Axios recommends publishers try to: appear inside LLMs now, improve their standing there, reduce reliance on search and the open web, and build direct relationships with audiences via newsletters, events and other direct channels.
The next phase: agents and AI as a distribution layer
The coming "agent" wave will further transform distribution: savvy publishers are shifting from treating AI engines as mere referral sources to treating them as a distribution layer. Increasingly, the person comparing prices or researching products online may be an individual’s AI agent acting as an intermediary.
Overall, publishers will need to rethink content formats and distribution strategies to serve both human readers and AI systems in order to maintain visibility and audience connections as the ecosystem evolves.
