How AI is rewriting the rules of SEO

Not long ago, SEO was a mechanical exercise. Pack your copy with keywords, chase backlinks wherever you could find them, fix a few technical quirks, and watch the rankings climb. That playbook no longer works.

The rise of AI agents and large language models (think ChatGPT or Gemini) is reshaping how people search for information. Increasingly, they’re bypassing the old link-and-click model in favour of direct, conversational answers. That shift is eroding traditional SEO’s dominance and forcing a rethink of how websites earn visibility.

From ranking to being referenced

The decline in organic traffic isn’t necessarily a reflection of bad content. It’s a sign that the rules of engagement have changed. Search engines are no longer just matching keywords, they’re weighing context, intent, and authority.

Success today isn’t about squeezing onto page one of a results page. It’s more about becoming the trusted source an AI draws on to formulate its answer, a new kind of “zero-click” victory.

Why the old playbook fails

Traditional SEO was built for algorithms that behaved like robots. They rewarded repetition, links at scale, and technical tweaks. AI doesn’t work that way. It “reads” more like a human, evaluating meaning, relationships between ideas, and credibility. That’s why tactics like keyword stuffing or indiscriminate backlinking no longer deliver, and why even top-ranking pages are losing traffic to AI-driven summaries.

A smarter way forward

Adapting to this AI-first environment means shifting from optimisation for bots to optimisation for understanding. The focus is on producing content that’s clear, structured, and above all, trustworthy.

The essentials:

Clarity and structure: Logical headings and scannable formats help both readers and machines digest your content.Topical authority: Instead of shallow keyword coverage, aim to become the definitive resource on a subject. Credibility (E-E-A-T): Demonstrate real expertise, authority, and trustworthiness: the signals AI uses to determine reliability. Conversational writing: Anticipate the natural questions people ask and answer them directly.

Tools for the transition

Tech providers are already building solutions designed for AI-first discoverability. Optimizely, for example, recently announced a set of groundbreaking CMS features that go beyond traditional SEO and into AEO (AI Engine Optimisation) and GEO (Generative Engine Optimisation).

The updates include:

Page optimisation: Enhancing pages with GEO-specific metadata, markdown summaries, and auto-generated Q&A pairs based on content. Site-wide optimisation: Auto-generated llms.txt files to guide LLMs, plus the ability to apply GEO metadata in bulk across existing content. GEO analytics: new performance insights, including a “GEO health index” that scores content against best practices and tracks crawl patterns to detect LLM traffic.

These kinds of tools will help brands ensure that content is both human-readable and machine-ready, keeping them relevant in a landscape where visibility depends on AI as much as it does on search.

Blending old and new

The smartest strategies combine the fundamentals: meta tags, sensible keywording, good site hygiene with AI-friendly practices like structured data and authoritative content. Done well, your site isn’t just readable, it’s reference-worthy.

The opportunity ahead

The shift to an AI-driven web isn’t the end of SEO. It’s the start of a new chapter in its memoir. By investing in clarity, authority, and trust, brands can position themselves not just to be found, but to be cited.

The next step doesn’t have to be sweeping. Start by adding structured data to key pages, or by rewriting one cornerstone article with a stronger summary and a clearer structure. Small moves now can put your site ahead as the rules of search continue to evolve.