AI & Automation
Personas
SaaS & Startup
Time to ROI
Medium-term (3-6 months)
Last year, while working on an e-commerce client's SEO strategy, I discovered something that completely changed how I think about content optimization. We were seeing our content mentioned in AI-generated responses despite being in a niche where LLM usage wasn't common. This led me down a rabbit hole that most SEO professionals are just starting to explore.
The question "Does ChatGPT consider keywords?" isn't just academic anymore. With AI-powered search experiences becoming mainstream, understanding how large language models process and reference content has become crucial for anyone serious about organic visibility.
While traditional SEO focused on ranking in Google's ten blue links, we're now entering an era where your content needs to be discoverable by AI systems that synthesize information rather than just index it. The rules are different, but the opportunities are massive.
In this playbook, you'll learn:
How AI systems actually process and reference content (it's not what you think)
Why traditional keyword strategies fall short in the AI era
My real-world experiment transitioning from traditional SEO to GEO optimization
The chunk-level thinking approach that actually gets AI mentions
Practical implementation steps for both SaaS and e-commerce businesses
Reality Check
What the AI optimization gurus won't tell you
If you've been following the AI and SEO space lately, you've probably heard conflicting advice about how to optimize for AI-powered search experiences. The "experts" are divided into two camps: those who claim traditional SEO is dead, and those who insist nothing has changed.
Here's what the industry typically recommends for AI optimization:
Keyword stuffing for AI: Some suggest cramming keywords thinking AI systems work like primitive search engines
Conversational content: Writing everything in question-answer format assuming AI prefers dialogue
Abandoning traditional SEO: Completely pivoting strategies without understanding how AI systems actually work
Over-structuring content: Adding excessive schema markup and structured data thinking more is better
AI-first content creation: Using AI to create content for AI, creating a feedback loop of mediocrity
This conventional wisdom exists because we're in the early days of AI-powered search, and most people are guessing rather than experimenting. The reality is more nuanced than either extreme suggests.
The truth? AI systems don't "consider" keywords the way traditional search engines do. They process content contextually, understanding meaning and relationships rather than matching exact terms. But this doesn't mean keywords are irrelevant—it means they serve a different purpose in the AI era.
Most content creators are either completely ignoring this shift or overreacting to it, missing the opportunity to build authority in AI-powered discovery systems while they're still developing their preferences.
Consider me as your business complice.
7 years of freelance experience working with SaaS and Ecommerce brands.
When I started working with this e-commerce client, they had a solid traditional SEO foundation but were curious about emerging AI trends. We discovered their content was appearing in ChatGPT responses organically—not something we'd optimized for, but it was happening naturally.
This discovery led me to dig deeper into what I now call GEO (Generative Engine Optimization). The client operated in a traditional e-commerce niche where you wouldn't expect much AI interaction, yet we were tracking a couple dozen LLM mentions per month.
My initial assumption was that we could simply apply traditional SEO principles to AI optimization. I spent weeks researching how different AI models process content, talking to teams at AI-first startups, and analyzing patterns in how content gets referenced.
What I learned was eye-opening: AI systems don't crawl and index like search engines—they consume content in chunks and synthesize information from multiple sources. This meant our entire approach to content structure needed to change.
The traditional SEO approach I'd been using focused on page-level optimization—title tags, meta descriptions, keyword density, internal linking. But when I analyzed which parts of our content were being referenced by AI systems, I noticed something crucial: they were pulling specific passages that could stand alone as complete thoughts, regardless of the page's overall optimization.
It wasn't about the page ranking #1 for a keyword. It was about individual sections being comprehensive and authoritative enough to serve as source material for AI-generated responses. This realization completely shifted my understanding of how content needs to be structured in the AI era.
Here's my playbook
What I ended up doing and the results.
Based on my experiments with this client and subsequent projects, I developed a layered approach to AI optimization that builds on traditional SEO rather than replacing it. Here's the exact framework I used:
Foundation Layer: Traditional SEO First
Despite what AI optimization gurus claim, you still need solid SEO fundamentals. AI systems need to find and access your content, which means:
Clean site architecture and fast loading times
Proper crawlability and indexing
Quality backlinks and domain authority
Comprehensive coverage of your topic area
Content Restructuring: Chunk-Level Thinking
This was the breakthrough insight. Instead of optimizing pages, I started optimizing individual sections for standalone value:
Each content section needed to be self-contained with context, explanation, and conclusion. For example, instead of writing "As mentioned above, the benefits include..." I rewrote sections to say "The key benefits of X include..." with full context.
I implemented what I call "snippet-ready structuring"—each paragraph or section could be extracted and still make complete sense. This meant more comprehensive explanations but also better user experience.
Authority Building Through Depth
Rather than surface-level coverage, I focused on creating the most comprehensive resource on each topic. AI systems favor authoritative, factual content when synthesizing responses.
I added data, statistics, case studies, and multiple perspectives to each piece. The goal was making our content the obvious choice for AI systems looking for reliable information to reference.
Multi-Modal Integration
AI systems increasingly process images, charts, and structured data alongside text. I integrated visual elements that supported the textual content and added proper alt text and captions that provided context.
Tables, charts, and infographics weren't just visual aids—they became additional touchpoints for AI systems to understand and reference our content.
Key Discovery
AI systems process content in passages, not pages. Each section must stand alone.
Implementation Method
Built comprehensive sections that work as independent knowledge units rather than connected narratives.
Measurement Approach
Tracked AI mentions across different platforms and analyzed which content structures performed best.
Success Metrics
Increased LLM mentions from dozens to hundreds per month through systematic content restructuring.
The results of this approach were significant and measurable. Within three months of implementing the GEO optimization framework, we saw:
Traditional SEO maintained its strength: Organic traffic continued growing at the same rate, proving that AI optimization doesn't hurt traditional search performance when done correctly.
AI mentions increased substantially: We went from occasional mentions to consistent referencing across multiple AI platforms. The content was being cited as authoritative source material.
User engagement improved: The chunk-level approach made content more scannable and valuable for human readers. Bounce rates decreased and time-on-page increased.
Authority building accelerated: The comprehensive, self-contained approach to content creation positioned the client as a definitive resource in their niche.
Most importantly, we achieved these results without sacrificing traditional SEO performance. The approach proved that AI optimization and traditional SEO can work together rather than compete.
What I've learned and the mistakes I've made.
Sharing so you don't make them.
This experiment taught me several crucial lessons about the intersection of AI and content strategy:
Don't abandon what works: Traditional SEO fundamentals remain essential. AI optimization should layer on top of solid SEO practices, not replace them.
Think in chunks, not pages: AI systems consume content differently than search engines. Structure content so each section provides complete value independently.
Authority beats optimization: AI systems favor comprehensive, factual content over content optimized for specific queries. Focus on being the best resource, not the most optimized one.
Context is crucial: Every section needs enough context to stand alone. Avoid references like "as mentioned above" that break when content is extracted.
Multi-modal matters: AI systems increasingly process visual elements alongside text. Integrate charts, tables, and images that add informational value.
Quality compounds: Unlike traditional SEO where you can optimize thin content, AI systems reward depth and comprehensiveness.
Early adoption advantage: While most businesses ignore AI optimization, implementing these practices now builds authority before the space becomes competitive.
How you can adapt this to your Business
My playbook, condensed for your use case.
For your SaaS / Startup
For SaaS companies looking to implement GEO optimization:
Create comprehensive feature documentation that works as standalone resources
Structure use case pages with complete context and examples
Build integration guides that AI systems can reference for technical questions
Focus on problem-solution content that addresses user queries comprehensively
For your Ecommerce store
For e-commerce stores implementing this approach:
Create detailed product guides that work as buying resources
Structure category pages with comprehensive buying advice
Build comparison content that helps users make informed decisions
Include technical specifications and use cases for each product