Signal amid the noise
Google offers only high-level SEO guidance, leaving the web’s SEO practitioners to divinate and read signs to work out how the search algorithm behaves. That tactic has yielded results before, but not every SEO recommendation succeeds.
The volatile state of the Internet today—marked by uneven traffic and the rapidly growing adoption of AI—may push struggling publishers toward more SEO snake oil like content chunking. When traffic is thin, any uptick is seized upon and credited to recent changes. If numbers fall, it’s simply written off as a bad spell.
This new content superstition can seem to work initially, but at best it’s an outcome of Google’s current quirks—the company isn’t designing LLMs to prefer broken-up content. Sullivan acknowledges there may be “edge cases” where content chunking looks effective.
“Fine. That’s what’s happening now, but tomorrow the systems may change,” he said. “You’ve created all these things aimed specifically at a ranking system, not at a human, because you were trying to do better in the rankings instead of focusing on people. Then the systems improve—likely the way they always do—to reward content written for humans. All the tweaks you made to please this LLM-driven system that might have worked now may not carry forward in the long run.”
Chunking probably won’t vanish while publishers can point to short-term gains. Nevertheless, Google appears to view chopping up content for LLMs as an unsustainable future for SEO.
