SEO just got replaced by GEO according to the public. But did it?

Background

What is SEO?
Search Engine Optimization (SEO) is the practice of improving a website’s visibility in search engine results. Website owners care because higher rankings mean more organic traffic—visitors who find you without paid ads. At its core, SEO is about aligning your content with what search engines prioritize: relevance, quality, and user experience.

What are web crawlers?
Web crawlers, also called “spiders,” are automated programs that scan websites to index their content. Think of them as librarians cataloging books. They follow links, analyze text, and report back to search engines like Google, which uses this data to rank pages.

Formal ways to improve SEO
Best practices include:

• Organized, accessible content: Clear headings, logical structure.
• Unique, up-to-date information: Originality matters.
• Speed and mobile optimization: Slow sites frustrate users (and crawlers).
• Helpful, reliable content: Answer questions thoroughly.

Traditional brute-force tactics
The SEO industry has long relied on shortcuts:

• Backlinks: Sites with high “domain authority” (DA)—a perceived measure of credibility—linking to your site can boost rankings. This spawned a market for buying backlinks, despite Google penalizing such practices.
• Keyword stuffing: Pages once hid blocks of popular keywords (e.g., “best pizza recipe pizza near me pizza delivery”)—sometimes invisible to users—to trick crawlers. Google now demotes this.
• Other tactics: Duplicate content, cloaking (showing different content to crawlers vs. users), and private blog networks (PBNs).

We don’t know
Google’s ranking algorithm is a black box. Much SEO knowledge comes from trial and error. Since 2015, Google’s RankBrain—a machine learning system—has further complicated things by interpreting user intent, not just keywords.

The Question

Headlines scream “SEO is dead!” as AI chatbots like Chat GPT rise. SEO experts push back, arguing it’s evolving. Who’s right?

People Fed Up with Traditional Search

• Clickbait overload: Top results often prioritize engagement over accuracy. (“You’ll Never Believe These 10 Secrets!”)
• Ad-riddled “authority” sites: High-DA publishers drown content in ads and sponsored posts.
• Paywalls and effort: Finding answers now requires dodging subscriptions and parsing jargon.

AI, Right Time Right Place

Why scroll through ads when AI gives direct answers? For users, it’s simpler. For businesses, the new challenge is: How do I get AI to mention my site?

AI, Machine Learning, Deep Learning: What’s the Difference?

• AI: Broad term dating to the 1950s. Early systems used “hard-coded logic” (explicit rules like “IF temperature > 100, THEN alert”).
• Machine Learning (ML): A subset of AI. Instead of hard-coding rules, ML algorithms learn patterns from data. Supervised ML uses labeled data (e.g., “this image is a cat”), while unsupervised ML finds hidden patterns (e.g., customer segments).
• Deep Learning: A subfield of ML using neural networks with many layers (“deep” structures). It excels at tasks like image recognition by mimicking brain-like learning.

Machine Learning and Statistics

ML is rooted in statistical learning theory. Consider temperature: it’s a statistic (average molecular energy), yet we rarely associate it with math. Similarly, ML applies statistics to predict outcomes—like ranking pages—without users realizing it.

However, traditional statistical models are “glass boxes”—you define the variables and interpret weights directly. ML models, particularly complex ones like deep learning, are more like “black boxes.” They learn patterns autonomously, often in ways even developers can’t fully explain. This loose control is a double-edged sword: while ML uncounterintuitive insights (e.g., RankBrain interpreting vague queries), it also makes outcomes harder to audit or tweak.

For SEO, this means strategies can’t rely on rigid rules. Instead, focus on what both statistics and ML prioritize: data quality, relevance, and user intent.

Transition Note:
We’ve covered key background, but unknowns remain. Just as Google’s algorithms are secretive, AI training datasets are often private. Reverse-engineering how models like GPT-4 work is nearly impossible.

How Does AI Get Its Information?

• Training data: Models learn from vast datasets—books, articles, code—scraped by crawlers (e.g., Common Crawl).
• Real-time searches: For fresh queries (“LaLiga’s top scorer”), some AI tools ping search engines via APIs, pulling top-ranked results.
• Implication: Ranking high on Google still matters. If an AI uses search APIs, your SEO efforts indirectly influence its answers.

The Emergence of GEO

If “SEO” feels outdated, enter Generative Engine Optimization (GEO)—optimizing content for AI mentions. Tactics might include structuring data for AI readability or emphasizing authority. But GEO is still theoretical; no proven playbook exists.

My Opinion

• GEO’s limits: Current tips (e.g., “use FAQs”) are speculative. AI models are too complex to reliably manipulate.
• Future tools: AI companies may offer dashboards (like Google Search Console) for businesses to manage their AI presence.
• Structured content wins: Use headless CMS (decoupling content from HTML) or JSON to future-proof for AI and traditional SEO.
• Coexistence: SEO isn’t dead. It’s adapting. Even AI tools depend on search engines—and thus, SEO—for real-time data.

The rush to capitalize on technological shifts reminds me of the internet boom era: countless companies raced to be “first,” only to crash when timing or execution faltered. Being early isn’t always the same as being right. Success often hinges on patience—waiting for tools like OpenAI’s or Google’s hypothetical GEO dashboard (akin to Search Console) to mature—while steadily evolving with current standards, like Google’s E-E-A-T framework (Experience, Expertise, Authoritativeness, Trustworthiness). Adaptation isn’t about shortcuts; it’s about aligning with sustainable practices that prioritize people. If users don’t find value in your content, no amount of optimization—for search engines or AI—will sustain it. Fortunately, the goal is shared: create something genuinely useful, and machines will follow.

References

Google Cloud. (n.d.). Applying Machine Learning to your Data with Google Cloud. Pluralsight.

Rose, D. (n.d.). Introduction to Artificial Intelligence. LinkedIn Learning.

Procopio, J. (2023). Is SEO Dead? Because That Would Make a Lot of Sense. Inc.com.

Patel, N. (2024). Is SEO Dead in 2025? NeilPatel.com.

IBM Technology. (2020). Machine Learning vs Deep Learning. YouTube.

Stewart, M. (2020). The Actual Difference Between Statistics and Machine Learning. Medium.

Bzdok, D., Altman, N., & Krzywinski, M. (2018). Statistics versus machine learning. Nature.

Cifuentes, M. (2024). GEO vs SEO: The New Rules for Search Visibility in an AI-First World. Flying Cat Marketing.

Handley, R., Skopec, C., & Paruch, Z. (n.d.). Google E-E-A-T: What It Is & How It Affects SEO. Semrush.