AI & Automation
Personas
SaaS & Startup
Time to ROI
Medium-term (3-6 months)
Most marketers are obsessing over Google rankings while completely ignoring the elephant in the room: AI assistants like Claude, ChatGPT, and Perplexity are becoming the new search engines. Last month, I discovered that one of my e-commerce clients was getting mentioned in AI responses dozens of times, despite being in a niche where "nobody uses AI for research."
This discovery sent me down a rabbit hole that completely changed how I think about content optimization. We're not just competing for Google's algorithm anymore – we're competing to be the source that AI models trust and cite.
The problem? Most businesses are still playing by Google's rules while their potential customers are asking Claude for recommendations. While everyone's focused on traditional SEO, there's a massive opportunity to dominate what I call GEO – Generative Engine Optimization.
Here's what you'll learn from my real-world experiments:
Why AI models favor certain content structures over traditional SEO optimizations
The chunk-level optimization strategy I discovered through client work
How to track AI mentions without expensive tools
Why citation-worthiness beats keyword density for AI visibility
The simple content restructuring that 3x'd our AI mentions
Industry Reality
What the SEO experts are saying about AI optimization
The SEO industry is scrambling to figure out this new landscape. Most experts are treating GEO like traditional SEO with a few tweaks, recommending familiar tactics:
The standard advice includes: Focus on featured snippets since AI models pull from those, optimize for question-based queries, create FAQ sections, and use schema markup for better understanding. Some are pushing expensive "AI SEO tools" that promise to track your mentions across AI platforms.
Here's where it gets interesting – and where most advice falls short. Traditional SEO assumes you're optimizing for one algorithm (Google's) with predictable ranking factors. But AI models work fundamentally differently. They're not just crawling and ranking; they're synthesizing information from multiple sources to generate original responses.
The conventional wisdom treats AI optimization like a new channel to conquer, using the same metrics and approaches that worked for Google. But after working directly with content that gets mentioned by AI models, I've learned this thinking is backwards.
Most SEO professionals are still thinking in terms of pages and rankings. They're asking "How do I get my page to rank in AI results?" But AI doesn't rank pages – it references specific chunks of information that help answer questions. This fundamental misunderstanding is why most traditional approaches aren't working.
The reality is more nuanced and, honestly, more interesting than just "SEO for AI."
Consider me as your business complice.
7 years of freelance experience working with SaaS and Ecommerce brands.
This journey started when I was working on a complete SEO overhaul for an e-commerce Shopify client. What began as traditional SEO work quickly evolved when we discovered something unexpected: their content was already appearing in AI-generated responses, despite being in a niche where most people assume AI isn't commonly used.
My client sold specialized B2B equipment – not exactly the type of industry where you'd expect people to be asking ChatGPT for recommendations. Yet when I started tracking, we found a couple dozen LLM mentions per month. This wasn't something we'd optimized for initially; it happened naturally as a byproduct of solid content fundamentals.
But here's what caught my attention: the content getting mentioned wasn't necessarily our highest-ranking pages on Google. Some pages that barely made page 2 of Google results were being cited by AI models regularly. Others that ranked in the top 3 for competitive keywords never got mentioned by AI at all.
This disconnect forced me to dig deeper. Through conversations with teams at AI-first startups and other practitioners, I realized everyone was still figuring this out. There's no definitive playbook yet, just experiments and observations.
The breakthrough came when I started analyzing which specific sections of our content were being referenced. I noticed that AI models weren't consuming entire pages like traditional search engines. Instead, they were breaking content into passages and synthesizing answers from multiple sources. Each section needed to be self-contained and valuable enough to stand alone.
This observation led to a complete restructuring of how we created content. Instead of optimizing for Google's algorithm, we started optimizing for how AI models actually process and reference information.
Here's my playbook
What I ended up doing and the results.
Rather than abandoning traditional SEO for experimental GEO tactics, I developed what I call a "layered approach" – building AI optimization on top of strong SEO fundamentals. Here's the exact process I developed:
Layer 1: Foundation First
I started with genuinely useful content for humans. This isn't just about keyword research – it's about creating content that actually solves problems. AI models seem to favor content that demonstrates clear expertise and provides actionable value. Every piece had to pass the "would a human expert reference this?" test.
Layer 2: Chunk-Level Optimization
This was the game-changer. Instead of optimizing entire pages, I restructured content so each section could stand alone as a valuable snippet. Each paragraph or section needed to be self-contained with enough context to be useful independently. This meant adding brief context at the beginning of sections and ensuring each chunk answered a specific aspect of a broader question.
Layer 3: Citation-Worthy Accuracy
AI models prioritize factual accuracy and clear attribution. I implemented a rigorous fact-checking process and ensured all claims were backed by credible sources. This wasn't just about avoiding errors – it was about building the kind of authority that AI models trust enough to cite.
Layer 4: Multi-Modal Integration
I discovered that content with supporting visuals, charts, and tables performed better for AI mentions. The key was making these elements contextually relevant, not just decorative. Tables summarizing key points, process diagrams, and comparison charts seemed to help AI models understand and reference the content more effectively.
Layer 5: Question-Answer Architecture
Instead of traditional keyword-focused headings, I restructured content around specific questions and comprehensive answers. This aligned perfectly with how people actually query AI models – they ask questions, not just search for keywords.
The implementation was systematic but not overwhelming. We applied these principles to existing high-performing content first, then rolled them out to new content creation. The focus remained on creating valuable content for humans while making it more accessible to AI models.
Content Structure
Each section must work independently while contributing to the whole story
Source Authority
Build the kind of credibility that AI models trust enough to cite regularly
Answer Architecture
Structure content around specific questions rather than keyword optimization
Multi-Modal Support
Integrate charts, tables, and visuals that help AI models understand context
The results validated my hypothesis about AI optimization being fundamentally different from traditional SEO. Within three months of implementing the layered approach, we tracked a significant increase in AI mentions across multiple platforms.
What surprised me most was the quality of these mentions. AI models weren't just citing our content – they were using it as authoritative sources to answer complex questions in our client's industry. We started appearing in responses about industry best practices, technical specifications, and even competitive comparisons where we weren't directly mentioned in the user's query.
The timeline was interesting too. Unlike traditional SEO where you might wait months to see ranking improvements, AI mentions started increasing within weeks of implementing the chunk-level optimization. It seems AI models incorporate new, high-quality content into their responses relatively quickly.
Perhaps most importantly, this approach didn't hurt our traditional SEO performance. If anything, the focus on creating genuinely valuable, well-structured content improved our Google rankings as well. The chunk-level optimization made our content more scannable for human readers, which improved engagement metrics.
The unexpected outcome was discovering that different AI models favor different aspects of our content. Claude seemed to prefer our detailed process explanations, while ChatGPT more frequently cited our comparison tables and data summaries.
What I've learned and the mistakes I've made.
Sharing so you don't make them.
Looking back on this experiment, several key insights emerged that completely changed how I approach content strategy:
AI optimization requires different metrics. Traditional SEO focuses on rankings and traffic, but AI optimization is about reference quality and citation frequency. You're not trying to rank; you're trying to become the go-to source for specific information.
Content architecture matters more than keyword density. How you structure information is crucial. AI models seem to favor content that's logically organized, with clear hierarchies and self-contained sections that can be referenced independently.
Authority building is accelerated. While traditional SEO authority building takes months or years, AI models can start referencing new, high-quality content relatively quickly if it meets their standards for accuracy and usefulness.
The landscape is still evolving rapidly. What works today might not work tomorrow as AI models evolve. The key is focusing on fundamentals – creating genuinely valuable, accurate, well-structured content – rather than trying to game specific algorithms.
Don't abandon SEO fundamentals. The best performing approach builds AI optimization on top of strong traditional SEO, not instead of it. The platforms may change, but quality content principles remain constant.
Manual tracking is still necessary. While tools for tracking AI mentions are emerging, manual verification remains important. AI models can interpret and cite content in unexpected ways that automated tools might miss.
Think in terms of decades, not months. This shift toward AI-powered search is permanent. Building for AI visibility now is like building for mobile-first design was a decade ago – it's not a trend, it's the new reality.
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 this approach:
Focus on creating detailed feature comparisons and use case documentation
Structure product documentation with self-contained sections for each feature
Build comprehensive integration guides that AI models can reference for technical questions
Create authoritative industry content that positions your expertise beyond just your product
For your Ecommerce store
For e-commerce stores implementing this strategy:
Develop detailed product comparison guides and buying decision frameworks
Create comprehensive category pages with expert-level information about product types
Build educational content around product usage, maintenance, and selection criteria
Focus on becoming the authoritative source for product information in your niche