The Truth About AIO, GEO, and AI SEO: Why You're Paying 10x for 13-Year-Old SEO Tactics

Here's what they're not telling you about AIO, GEO or AI SEO.

Every few months, the marketing world gets a new buzzword. Right now, it's "AIO" or "AI SEO." Some call it GEO (Generative Engine Optimization). Others talk about optimizing for LLMs or achieving "prompt-based visibility." Different labels. Same story.

Here's what they're not telling you: You're not witnessing the birth of a revolutionary new discipline. You're watching traditional SEO get rebranded at 10x the price.

And we can prove it.

The Big Lie: "LLMs Use Different Sources"

The foundation of every "AI SEO" sales pitch is simple: ChatGPT, Perplexity, and Gemini find information differently than Google, so you need entirely new optimization strategies.

Except they don't.

When OpenAI launched ChatGPT Search on October 31, 2024, VP Srinivas Narayanan confirmed in a Reddit AMA: "We use a set of services and Bing is an important one." That's right. ChatGPT's revolutionary search feature is powered by Microsoft's Bing Search API, the same infrastructure that's been around for years.

Google's Gemini? It uses Google Search infrastructure directly through "Grounding with Google Search." No secret AI magic, just their own search engine with an AI synthesis layer on top.

Even Perplexity, the one exception that built its own crawler (PerplexityBot), still processes only 400 million queries monthly compared to Google's 13-16 billion daily searches. And guess what? It still uses Google's BERT model and AWS infrastructure.

The truth: LLMs aren't building parallel search ecosystems. They're licensing the same search APIs you've been optimizing for all along, then adding a conversational interface on top.

The Numbers Don't Lie

Let's look at what actually happens when you test Google rankings against LLM citations:

The Ahrefs study (15,000 queries): Only 12% of URLs cited by AI rank in Google's top 10 for that specific query. Sounds revolutionary, right? But here's what they buried in the data: 31.8% domain overlap. The same authoritative sites succeed in both channels, just different pages.

Semrush's analysis (150,000+ citations):

• Perplexity: 82% URL overlap with Google (basically search with AI formatting)
• Google AI Overviews: 67% URL overlap
• ChatGPT: 28% URL overlap (lowest, but still 45% at the domain level)

What does this mean? If you're Mayo Clinic, Wikipedia, or Reddit, you dominate both channels. If you're a random blog trying to game the system with "AI-specific" tactics, you're invisible everywhere.

The pattern is clear: High-authority, comprehensive content from established domains wins in both traditional search and LLM citations. The infrastructure rewards the same fundamentals, just through different interfaces.

The 13-Year-Old "Innovation"

Here's my favorite part of the AI SEO pitch: agencies charging premium rates for "entity-first optimization" and "knowledge graph development."

Let me give you some dates:

• Schema.org launched: June 2, 2011 (collaboration between Google, Microsoft, and Yahoo!)
• Google Knowledge Graph: May 16, 2012 (with the famous "things not strings" announcement)
• Hummingbird algorithm: August 2013 (complete rewrite to focus on entities over keywords)

By 2013-2015, the SEO industry was already running conferences on entity optimization. Today, over 45 million websites use schema markup, and 72.6% of first-page Google results have structured data.

Translation: The "revolutionary" tactics agencies are selling you? They've been standard practice for over a decade. ChatGPT didn't invent entity-based search, Google did, 13 years ago.

But ChatGPT Is Taking Over Search... Right?

Not even close.

OpenAI's own data (analyzing 1.5 million conversations in September 2025) reveals that only 24.4% of ChatGPT usage involves information seeking. The other 75%? Writing assistance (28%), practical guidance, creative tasks, and technical help.

More revealing: Semrush found that only 46% of ChatGPT users even enable the web search feature. The majority rely solely on the AI's built-in knowledge.

Market share reality:

• Google: 13-16 billion searches daily (89-91% market share)
• ChatGPT: 1-2 billion total queries (1-2.1% market share)
• In 2024, Google received 373 times more searches than ChatGPT

ChatGPT isn't a search engine. It's a multi-purpose AI assistant where search is a minority use case. Yes, that 25% matters for brand visibility, but characterizing this as a search revolution fundamentally misunderstands what users are actually doing.

What the Research Actually Shows

I'm not saying there are no differences. The academic research from Princeton and IIT Delhi (published November 2023) tested 10,000 queries and found specific tactics improve LLM visibility by 22-40%:

• Adding citations from credible sources: +30-40%
• Including quotations: +27%
• Adding statistics: +25%
• Improving fluency: +24%
• Adding technical terms: +22%

These are real, measurable improvements. But here's the thing: every single one of these tactics also improves traditional SEO. Citation-rich, well-structured, statistically-supported content has always performed better.

The research proved that keyword stuffing, a traditional SEO manipulation tactic, performs worse than baseline for LLMs. In other words, the shortcuts don't work. Quality content wins.

Sound familiar?

The Real Differences (And Why They Don't Change Everything)

To be intellectually honest, some legitimate nuances exist:

Query length matters: ChatGPT queries average 23 words versus 4 words for Google. Users ask full questions instead of typing keywords. So yes, conversational content performs better.

Citation mechanics differ: LLMs use "query fan-out", generating multiple variations of your question and synthesizing across sources. A page ranking #6 for three related queries might get cited over a #1 ranking for one exact match.

Commercial queries diverge most: Google shows you products to buy. ChatGPT gives editorial guidance on what to consider. For "best running shoes," Google ranks Nike and Adidas product pages. ChatGPT cites running magazine reviews and comparison articles. The correlation? -0.98, almost perfect opposites.

These differences matter tactically. But they don't require a fundamentally different discipline. They require the same adjustments we made for mobile SEO (page speed, responsive design) or voice search (conversational content, featured snippets).

Same principles. New interface. Tactical adjustments.

Follow the Money

Microsoft increased Bing Search API pricing by 5-10x in February 2023, from $7 to $25-200 per 1,000 transactions, explicitly citing "technology investments" including ChatGPT integration.

Think about that. Microsoft, which has invested $13+ billion in OpenAI, which owns the infrastructure ChatGPT runs on, which has effectively unlimited resources, even they couldn't build search infrastructure cheaply enough to avoid massive price increases.

Building Google-scale search infrastructure requires billions of dollars. That's why ChatGPT licenses Bing. That's why Perplexity raised $500 million at a $9 billion valuation just to maintain its own crawler. That's why Claude's search capabilities remain mysteriously vague about partnerships.

The economics prove the point: LLMs aren't replacing traditional search infrastructure. They're expensive interfaces built on top of it.

The Agency Playbook (And Why It's Mostly Theater)

Dozens of agencies now offer "AI SEO" at premium prices. Here's what they're actually selling:

• "Citation optimization" = Adding source links (good practice since forever)
• "Conversational content" = Writing for how people talk (voice search best practice since 2016)
• "Entity-first SEO" = Schema markup (available since 2011)
• "Structured data for AI parsing" = The same schema markup you already have
• "E-E-A-T optimization" = Google's core quality guidelines since 2018

One agency trademarked "LLM Visibility™" and claims to "build and train your knowledge graph." Another claims to have "developed 927+ AI algorithms over the past 11 years." A third charges $5,000+/month retainers for what amounts to implementing decade-old structured data standards.

The academic research is real. The tactical improvements are measurable. But charging 3-10x traditional SEO rates to apply 13-year-old entity optimization practices to new platforms? That's opportunistic rebranding.

What This Means for Your Business

I'm not saying LLM optimization doesn't matter. It does. But here's the honest framing:

LLM visibility is an extension of SEO fundamentals, not a replacement for them.

The 25% of top Google rankings that appear in ChatGPT answers? The 82% URL overlap between Perplexity and Google? The fact that Wikipedia, Mayo Clinic, and Reddit dominate both channels? This proves traditional authority and optimization still drive AI visibility.

Your actual strategy should be:

1. Foundation: Traditional SEO (this still matters)

• High-authority domain with quality backlinks
• Comprehensive, expert-authored content
• Technical excellence (speed, mobile, structure)
• Clear site architecture and internal linking

2. Tactical adjustments for LLM extraction:

• Add citations from credible sources throughout your content
• Include statistics, quotations, and data points
• Write conversationally to address full questions
• Use clear headers that directly answer queries
• Implement proper schema markup (if you haven't already)
• Create comprehensive coverage across related topics on your domain

3. Accept the format differences:

• LLMs cite editorial content for commercial queries; Google ranks product pages
• Brand mentions without clicks still have value
• Informational content gets more LLM traction than pure sales pages

That's it. You don't need a "revolutionary" new approach. You need to do traditional SEO well, then make tactical adjustments for how LLMs extract and present information.

The Bottom Line

When someone pitches you "AI SEO" as a fundamentally different discipline requiring specialized expertise at premium prices, ask them three questions:

1. "Which search infrastructure do LLMs actually use?" (Answer: Bing, Google, or extremely expensive proprietary crawlers)
2. "When did entity optimization and schema markup become standard SEO practice?" (Answer: 2011-2013, over a decade ago)
3. "What percentage of ChatGPT prompts trigger external searches?" (Answer: 31%, and they query the same search engines you've been optimizing for)

The AI SEO industry isn't wrong that optimization for LLMs matters. They're wrong that it's revolutionary.

Yes, the target shifted, from ranking pages to being cited in answers. Yes, tactical adjustments matter, clarity, consistency, and citation-rich content perform measurably better. But the infrastructure, the authority signals, and the fundamentals driving visibility? Those haven't changed.

You're not discovering a new discipline. You're rediscovering that high-authority, well-structured, comprehensively-researched content wins, regardless of which interface consumes it.

The infrastructure is the same. The entity understanding is 13 years old. The domains succeeding are the same. What changed is the interface layer, and that requires tactical adjustments, not wholesale reinvention.

So while everyone else chases the new shiny object, keep doing what actually works:

• ✅ Build genuine authority in your space
• ✅ Create comprehensive, well-researched content
• ✅ Structure information clearly with proper markup
• ✅ Show up consistently across relevant platforms
• ✅ Make your expertise and credibility obvious

The plumbing hasn't changed. Just the faucet.

And that's exactly why we built Hike, and why we created Kit, our AI SEO expert. Kit automatically manages your SEO, keeping your business visible in Google and everywhere that depends on Google's infrastructure, without requiring you to become an SEO expert or pay premium "AI SEO" agency rates.

Because visibility still matters. Traditional SEO still runs the internet. And the best strategy for AI visibility? It's the same strategy that's always worked, just executed better.

📚 Further Reading

Key Studies Referenced:

• Nectiv: ChatGPT Search Query Analysis - 8,500 prompts, 31% trigger searches
• Ahrefs: AI Search Overlap Study - 15,000 queries analyzed
• Princeton/IIT Delhi: GEO Research Paper - Academic foundation
• Semrush: ChatGPT Search Insights - 80M clickstream records
• Search Engine Land: Perplexity vs Google - 82% overlap data
• Search Engine Land: AI Search Evolution - Myths vs reality

Frequently Asked Questions

What is the difference between AI SEO and traditional SEO?

AI SEO is essentially traditional SEO with tactical adjustments for how LLMs extract and present information. The core infrastructure, authority signals, and optimization fundamentals remain the same. LLMs like ChatGPT use existing search APIs (mainly Bing and Google) rather than building parallel search ecosystems.

Do I need to optimize differently for ChatGPT and Google?

You need the same SEO foundation for both: high-authority domains, comprehensive content, technical excellence, and clear site architecture. Add tactical adjustments like citations from credible sources, statistics, conversational content, and proper schema markup. These tactics improve performance in both traditional search and LLM citations.

Is AI SEO worth the premium pricing agencies charge?

Most "AI SEO" tactics are rebranded traditional SEO practices that have been standard since 2011-2013, including schema markup, entity optimization, and E-E-A-T guidelines. While LLM optimization matters, paying 3-10x traditional SEO rates for decade-old structured data implementation is often opportunistic rebranding rather than revolutionary new expertise.

What percentage of ChatGPT usage involves search?

Only 24.4% of ChatGPT usage involves information seeking, with the remaining 75% focused on writing assistance, practical guidance, creative tasks, and technical help. Additionally, only 46% of ChatGPT users enable the web search feature, and Google still receives 373 times more searches than ChatGPT.

Do LLMs cite the same websites as Google?

Yes, there's significant overlap. Perplexity shows 82% URL overlap with Google, Google AI Overviews show 67% overlap, and even ChatGPT shows 45% overlap at the domain level. High-authority domains like Wikipedia, Mayo Clinic, and Reddit dominate both traditional search and LLM citations.

What tactics actually improve LLM visibility?

Academic research shows adding citations from credible sources improves visibility by 30-40%, including quotations by 27%, adding statistics by 25%, improving fluency by 24%, and adding technical terms by 22%. However, these same tactics also improve traditional SEO performance.

Your New Best Friend Does SEO

The truth is, whether you're optimizing for Google or LLMs, you need the same foundation: high-authority, well-structured, comprehensively-researched content. That's where Hike comes in. Hike is powered by Kit, your new best friend in the world of SEO. Kit isn't just a clever piece of tech, it's a fully-fledged AI agent that works behind the scenes to grow your website traffic.

From building out your keyword strategy to writing SEO-optimised blog posts and fixing technical issues, Kit handles it all. You don't need to become an SEO guru overnight. You just need Kit.

👉 Start your journey with Kit today