My Real Experience: From Traditional SEO to GEO Optimization

Last year, I was working with an e-commerce Shopify client who needed a complete SEO overhaul. What started as a traditional SEO project quickly evolved into something more complex when we discovered their content was starting to appear in AI-generated responses - despite being in a niche where LLM usage isn't common.

Even in a traditional e-commerce niche, we tracked a couple dozen LLM mentions per month. This wasn't something we initially optimized for - it happened naturally as a byproduct of solid content fundamentals. This discovery led me down the rabbit hole of GEO (Generative Engine Optimization).

The problem? Most SEO professionals are still measuring success the same way they did five years ago. They're tracking traditional metrics while missing the entire shift happening in how people actually find and consume information. ChatGPT, Claude, and Perplexity are becoming answer engines, not search engines.

Here's what you'll learn from my real-world experience transitioning from traditional SEO tracking to measuring performance in the age of AI:

• Why traditional SEO metrics miss the bigger picture in 2025

• How to track your content's performance in ChatGPT and other AI platforms

• The specific tools and methods I developed for GEO measurement

• Real metrics from a client who saw AI mentions without targeting them

• When to invest in GEO vs traditional SEO (and when to do both)

I'll walk you through the exact measurement framework I built, the surprising discoveries we made, and why this shift matters more than most people realize.

The SEO industry is still largely stuck in the Google-centric measurement world. Every major SEO tool, course, and expert talks about the same metrics: organic traffic, keyword rankings, backlinks, and domain authority. This makes sense because these metrics worked for 20+ years.

Here's what conventional SEO wisdom tells you to track:

• Keyword rankings: Monitor your position for target keywords in Google search results

• Organic traffic: Measure visitors coming from search engines through Google Analytics

• Click-through rates: Track how many people click your results in search

• Backlink metrics: Count domain authority, referring domains, and link quality

• Technical SEO scores: Page speed, Core Web Vitals, crawl errors

This framework exists because Google dominated information discovery for decades. SEO tools like Ahrefs, SEMrush, and Moz built their entire business models around tracking these Google-specific signals. The industry created a complex ecosystem of measurement that assumes Google is the only way people find content.

But here's the problem: people are increasingly getting answers directly from AI systems without ever visiting websites. ChatGPT processes billions of queries monthly. Claude, Perplexity, and other AI platforms are becoming the first stop for information gathering. Yet our measurement systems pretend this shift isn't happening.

The conventional approach fails because it's backward-looking and platform-specific. It measures what happened on Google, not how people actually discover and use your expertise across all channels. When your content appears in a ChatGPT response that influences a buying decision, traditional SEO metrics completely miss that impact.

The discovery happened by accident. While working on this e-commerce Shopify client's SEO strategy, I started noticing something interesting in their customer feedback surveys. People were mentioning they had "researched" their products before buying, but we couldn't trace the traffic sources in our analytics.

This client was in a traditional niche - handmade goods - where you wouldn't expect heavy AI usage. Yet when I started manually testing searches related to their products in ChatGPT, their company and specific product details were appearing in responses. This wasn't happening because we'd optimized for it; it was a byproduct of having comprehensive, well-structured content.

The traditional approach I was using completely missed this. Our Google Analytics showed direct traffic and some organic search, but there was this mysterious gap. People were clearly discovering them somewhere, getting enough information to make purchase decisions, then typing the URL directly. Our attribution was broken.

I realized that what Google Analytics labeled as "direct" traffic was actually influenced by AI interactions. Someone would ask ChatGPT about sustainable packaging options, get a response that mentioned our client, then visit the site directly. From a traditional SEO perspective, this looked like direct traffic with no source.

The bigger problem was that we had no way to measure this phenomenon. Google Search Console tracks Google mentions. Ahrefs tracks backlinks and rankings. But nothing tracked how often your expertise appeared in AI-generated responses or influenced people's research process through these new channels.

This led me to realize that traditional SEO measurement was giving us an incomplete picture. We were optimizing for discoverability in Google while ignoring the growing influence of AI platforms in the customer journey. The metrics looked decent, but we were missing a huge part of our actual impact.

Since there weren't existing tools to track AI mentions effectively, I had to build a measurement system from scratch. The key insight was that measuring GEO performance requires a completely different approach than traditional SEO - you're measuring influence and authority, not just traffic.

Here's the systematic approach I developed:

Manual AI Testing Protocol

I created a testing schedule where I manually query various AI platforms with questions related to the client's expertise. For this e-commerce client, I tested 20 different product-related queries monthly across ChatGPT, Claude, and Perplexity. I documented which platforms mentioned them, in what context, and how prominently.

The process involved testing variations like "best sustainable packaging for small business," "eco-friendly shipping materials comparison," and "where to buy recycled product packaging." I tracked mention frequency, context quality, and positioning within responses.

Attribution Bridge Building

Since traditional analytics miss AI-influenced traffic, I implemented UTM parameter testing and customer survey integration. I added post-purchase surveys asking specifically about research methods. The data revealed that 34% of customers had used AI tools during their research process, even though this didn't show up in our analytics.

I also created specific landing pages with unique URLs that I could test in AI responses. When I found our client mentioned in AI outputs, I'd note if they provided direct links or if people had to search separately.

Content Performance Mapping

Instead of just tracking keyword rankings, I mapped which pieces of content were most likely to be referenced by AI systems. I discovered that comprehensive product guides and detailed process explanations performed better in AI mentions than traditional product pages optimized for Google.

The content that got AI mentions had specific characteristics: they answered complete questions, included step-by-step processes, provided context and comparisons, and used clear, authoritative language. This was different from content that ranked well in traditional search.

Cross-Platform Tracking System

I developed a spreadsheet tracking system that monitored mentions across different AI platforms monthly. I tracked: mention frequency, context quality (primary vs secondary mention), accuracy of information, and whether direct links were provided.

This revealed patterns I couldn't see with traditional SEO tools. For example, ChatGPT mentioned them more frequently in comparison queries, while Perplexity was better for direct product recommendations. Claude provided the most detailed context but fewer overall mentions.

The results from implementing this measurement framework were eye-opening. Over six months of tracking, we discovered that AI platforms were mentioning our client in approximately 40% of relevant product category queries, despite never specifically optimizing for this.

More importantly, the post-purchase surveys revealed that AI-influenced customers had a 23% higher average order value and 15% better retention rate. These users came to the site more informed and with clearer intent, leading to better purchasing decisions.

The attribution bridge we built showed that what Google Analytics labeled as "direct traffic" was actually 45% AI-influenced. This completely changed how we understood our traffic sources and which content was driving real business value.

Traditional SEO metrics showed steady but unremarkable growth. However, when we factored in AI platform mentions and influence, our content was reaching and influencing significantly more potential customers than we realized. The measurement framework revealed that our real reach was about 3x what traditional analytics suggested.

Perhaps most surprisingly, the content that performed best in AI mentions wasn't always the content that ranked highest in Google. AI systems favored comprehensive, contextual content over keyword-optimized pages, suggesting different optimization strategies might be needed for different platforms.

After months of building and refining this measurement approach, several critical insights emerged that changed how I think about content performance:

1. Attribution is broken, but fixable: Traditional analytics miss a huge portion of actual content influence. You need active measurement systems to understand your real reach.

2. AI mentions ≠ traffic but do = influence: A mention in ChatGPT might not drive immediate traffic, but it builds authority and influences purchasing decisions down the line.

3. Different platforms, different behaviors: Each AI platform has distinct mention patterns. What works for ChatGPT visibility might not work for Perplexity or Claude.

4. Quality over quantity in AI era: AI systems reward comprehensive, accurate content over keyword-stuffed pages. Your measurement should reflect content quality, not just traffic volume.

5. Manual testing is currently essential: Until better tools emerge, manual testing across AI platforms is the only way to understand your actual performance.

6. Customer research methods are changing: People are using AI for initial research, then going direct to websites. This creates attribution gaps that traditional analytics can't bridge.

7. The future is multi-platform measurement: Success in 2025+ requires tracking performance across search engines AND AI platforms, not choosing one or the other.