Look, I get it. Every time a new search technology appears, we try to map it to what we already know.
- When mobile search exploded, we called it “mobile SEO.”
- When voice assistants arrived, we coined “voice search optimization” and told everyone this would be the new hype.
I’ve been doing SEO for years.
I know how Google works – or at least I thought I did.
Then I started digging into how ChatGPT picks citations, how Perplexity ranks sources, and how Google’s AI Overviews select content.
I’m not here to declare that SEO is dead or to state that everything has changed. I’m here to share the questions that keep me up at night – questions that suggest we might be dealing with fundamentally different systems that require fundamentally different thinking.
The questions I can’t stop asking
After months of analyzing AI search systems, documenting ChatGPT’s behavior, and reverse-engineering Perplexity’s ranking factors, these are the questions that challenge most of the things I thought I knew about search optimization.
When math stops making sense
I understand PageRank. I understand link equity. But when I discovered Reciprocal Rank Fusion in ChatGPT’s code, I realized I don’t understand this:
- Why does RRF mathematically reward mediocre consistency over single-query excellence? Is ranking #4 across 10 queries really better than ranking #1 for one?
- How do vector embeddings determine semantic distance differently from keyword matching? Are we optimizing for meaning or words?
- Why does temperature=0.7 create non-reproducible rankings? Should we test everything 10 times over now?
- How do cross-encoder rerankers evaluate query-document pairs differently than PageRank? Is real-time relevance replacing pre-computed authority?
These are also SEO concepts. However, they appear to be entirely different mathematical frameworks within LLMs. Or are they?
When scale becomes impossible
Google indexes trillions of pages. ChatGPT retrieves 38-65. This isn’t a small difference – it’s a 99.999% reduction, resulting in questions that haunt me:
- Why do LLMs retrieve 38-65 results while Google indexes billions? Is this temporary or fundamental?
- How do token limits establish rigid boundaries that don’t exist in traditional searches? When did search results become limited in size?
- How does the k=60 constant in RRF create a mathematical ceiling for visibility? Is position 61 the new page 2?
Maybe they’re just current limitations. Or maybe, they represent a different information retrieval paradigm.
The 101 questions that haunt me:
- Is OpenAI also using CTR for citation rankings?
- Does AI read our page layout the way Google does, or only the text?
- Should we write short paragraphs to help AI chunk content better?
- Can scroll depth or mouse movement affect AI ranking signals?
- How do low bounce rates impact our chances of…
Source link
Disclaimer
We strive to uphold the highest ethical standards in all of our reporting and coverage. We blogs.grocliq.com want to be transparent with our readers about any potential conflicts of interest that may arise in our work. It’s possible that some of the investors we feature may have connections to other businesses, including competitors or companies we write about. However, we want to assure our readers that this will not have any impact on the integrity or impartiality of our reporting. We are committed to delivering accurate, unbiased news and information to our audience, and we will continue to uphold our ethics and principles in all of our work. Thank you for your trust and support.
Website Upgradation is going on for any glitch kindly connect at [email protected]
