The Ranking Illusion
For two decades, B2B teams operated on a single compass: Google rankings. Position three on a high-intent keyword meant qualified traffic. Position one meant dominance. Visibility was measurable, trackable, and — crucially — it predicted revenue.
That compass is broken.
Across markets from Singapore to the United States, we're watching a strange inversion happen. Teams maintain strong Google rankings while buyer discovery plummets. They rank for the right keywords, hit target positions, and see traffic drop simultaneously. The disconnect isn't a lag or a temporary anomaly. It's structural.
The shift is simple: buyers have moved. They're no longer starting searches in Google. They're asking ChatGPT. They're querying Claude. They're using Perplexity for research. And they're watching Google's own AI Overviews reshape what "first position" even means.
Where Buyers Actually Start Now
The fragmentation of discovery
Consider the B2B buyer journey in 2026. A procurement manager researching AI automation solutions no longer types "best AI automation platforms" into Google Search. They open ChatGPT, paste their use case, and ask for a comparison. If they want recent sources and attribution, they use Perplexity. For deep dives into technical architecture, they ask Claude.
Each of these engines operates on fundamentally different ranking logic than Google.
Google ranks pages. These engines rank responses. They synthesize, quote, and summarize content without ever driving clicks to your site. You can rank first on Google for "enterprise workflow automation" and never appear in a single ChatGPT response about the same topic — because the LLM's training data, cutoff date, and retrieval logic don't map to traditional SEO signals.
The vanishing middle funnel
Traditional SEO captured the middle of the funnel: buyers who weren't ready to talk to sales but were ready to research. Generative engines have collapsed that space. A buyer gets a synthesized answer — often good enough — in 90 seconds. They don't click through to your content. They don't visit your site. They don't become a lead.
Most teams are still optimizing for ranking position on a platform that's no longer their buyer's starting point. The best teams are optimizing for inclusion, attribution, and synthesis across four different discovery engines at once.
This is a harder problem than SEO ever was. Ranking on Google required on-page signals, links, and authority. Appearing in a ChatGPT response requires training data inclusion, semantic relevance, and the kind of content that LLMs choose to cite when synthesizing an answer — often primary research, unique data, or defensible claims that generative engines can't easily replicate from five other sources.
The Signals That Matter Now
If traditional rankings no longer predict buyer discovery, what does?
Citation likelihood. How often do LLMs reference your content when answering questions in your space? This is harder to measure than Google position but far more predictive of actual buyer awareness.
Synthesis compatibility. Can an AI engine cleanly extract and attribute your insights without confusion? Scattered, image-heavy, or Flash-based content fails here.
Training data presence. Older cutoff dates for some models mean your recent content doesn't exist for them yet. Distributing fresh, high-signal content across multiple surfaces — academic repositories, industry platforms, paid syndication — matters.
Unique claim ownership. The more your insights can't be easily restated in five other places, the more likely an LLM will cite you by name rather than synthesize your idea into a generic response.
What This Means for Your Team
The visibility paradox reveals a hard truth: you can't optimize your way to discovery if you're only optimizing for Google. Teams across Australia, Germany, the UK, and beyond are learning this the hard way. Visibility now requires presence across four engines, not one — and each demands a different strategy.
If you want a deeper look at how to build visibility in an engine-fragmented world, Modulus has published detailed methodology on Generative Engine Optimization (GEO) and how it differs from traditional SEO — worth exploring if your team is already seeing the cracks in rankings but can't yet explain the traffic loss.
Read next from Modulus1:
Originally published on the Modulus1 insights blog. Browse more analysis on AI, SEO, and automation.
Top comments (0)