Search is evolving faster than ever. Whilst Google still dominates, AI-powered search engines like ChatGPT, Perplexity, and Google’s own Search Generative Experience are changing how people find information. If your SEO strategy only focuses on traditional search, you’re already behind.
The good news? The fundamentals still matter. But you need to layer in new optimisation strategies for LLM-based search. Here’s how.
The Foundation: Traditional SEO Still Matters
Before you worry about AI search, make sure your traditional SEO foundation is solid. This is non-negotiable baseline work.
Technical SEO essentials:
- Site speed under 3 seconds (Google prioritises fast-loading sites)
- Mobile-first design (60%+ of searches happen on mobile)
- Clean URL structures and logical site architecture
- SSL certificates and HTTPS everywhere
- Core Web Vitals optimisation
Think of technical SEO as your infrastructure. Without it, everything else struggles to gain traction.
On-page fundamentals:
Target one primary keyword per page, but write naturally. Your title tag should be 50-60 characters with the keyword near the front. Meta descriptions (150-160 characters) don’t directly affect rankings, but they influence click-through rates significantly. Use header hierarchy properly—H1 for your main topic, H2s for major sections, H3s for subsections. Include alt text on all images for both accessibility and search engine understanding. Internal linking between related pages helps search engines understand your site structure and spreads authority throughout your domain.
Content quality signals:
Depth beats volume every time. One comprehensive, well-researched article outranks ten shallow pieces. Google’s algorithms can now distinguish between regurgitated information and original insights backed by real expertise. Regular updates keep content current and signal ongoing relevance. The E-E-A-T framework (Experience, Expertise, Authoritativeness, Trustworthiness) has become central to how Google assesses content quality. Write like you actually know what you’re talking about—because you should.
The New Layer: Optimising for AI Search Engines
AI search engines work differently. They’re not matching keywords—they’re understanding context, synthesising information, and providing direct answers. This requires a different approach.
Write for featured snippets and direct answers
AI models pull from content that answers questions clearly and concisely. Structure your content to be “snippet-worthy”:
Start sections with direct answers. If someone asks “What is technical SEO?”, give them a clear definition in the first sentence: “Technical SEO refers to optimising your website’s infrastructure to help search engines crawl and index your content effectively.” Then elaborate with context and examples.
Think inverted pyramid—most important information first, supporting details following. Don’t make readers (or AI models) hunt for the answer.
Implement structured data and schema markup
Schema markup helps both Google and AI models understand your content context:
- Article schema for blog posts (author, date, topic)
- FAQ schema for Q&A content
- HowTo schema for instructional content
- Product schema for e-commerce
- Organisation schema for business information
Tools like Google’s Structured Data Markup Helper make this straightforward. The time investment is minimal compared to the visibility benefits.
Create comprehensive, context-rich content
LLMs favour content that provides full context rather than assuming prior knowledge. This doesn’t mean dumbing down—it means being thorough. Define terms even if they seem basic. Explain the “why” behind recommendations, not just the “how.” Include relevant background that helps readers understand the bigger picture. Use clear, descriptive language. AI models reward content that can stand alone as a complete resource.
Structure content with clear hierarchy
AI models parse structure to understand importance and relationships:
- Use descriptive headers that preview content (not clever/vague ones)
- Break complex topics into logical subsections
- Include a table of contents for long articles
- Add summary sections for key takeaways
Clear structure helps both human readers navigate and AI models understand your content.
Optimise for conversational queries
People ask AI search engines differently than they use Google:
- Traditional: “best CRM small business”
- AI search: “What’s the best CRM for a small business with a remote team and limited budget?”
Write content that answers these longer, more specific queries naturally. Address common follow-up questions. Use conversational language that mirrors how people actually speak.
Practical Steps to Implement This Week
Day 1: Audit your top 10 pages
Check technical SEO basics—speed, mobile responsiveness, indexing status. Identify pages that could include direct answers near the top. Note missing schema markup opportunities. This gives you a baseline and helps prioritise improvements.
Day 2: Restructure one high-traffic article
Pick your best-performing piece and optimise it:
- Add a clear, direct answer to the primary question in the intro
- Break content into sections with descriptive H2/H3 headers
- Add a FAQ section at the bottom with related questions
- Include contextual information that helps readers understand the “why”
This becomes your proof of concept.
Day 3: Implement schema markup
Start with Article schema on blog posts. Add FAQ schema to pages with Q&A content. Test implementation with Google’s Rich Results Test. Don’t try to do everything at once—focus on your highest-traffic pages first.
Day 4: Optimise for conversational queries
Use tools like AnswerThePublic to find question-based queries in your niche. Rewrite one section to answer these naturally. Add contextual information that AI models value. Think about the follow-up questions someone might ask after getting an initial answer.
Day 5: Create an internal linking strategy
Connect related content with descriptive anchor text. Build topic clusters around your main services or expertise areas. This helps both users navigate and search engines understand your authority.
How to Measure Success
Track these metrics monthly to see if your efforts are working:
Traditional SEO metrics:
- Organic traffic growth in Google Analytics
- Keyword rankings for target terms (use Ahrefs or SEMrush)
- Click-through rates from search results
- Time on page and engagement signals
- Conversion rates from organic traffic
AI search visibility:
- Monitor mentions in AI search results (manually check ChatGPT, Perplexity for your topics)
- Track “zero-click” searches where your content appears in snippets
- Look for increases in branded searches (people finding you through AI, then searching directly)
Don’t expect overnight results. SEO is a compound investment—improvements take weeks or months to show in rankings and traffic.
The Bottom Line
SEO in 2025 isn’t about choosing between traditional search and AI search—it’s about serving both. The strategies overlap more than they diverge. Focus on creating genuinely helpful, well-structured content that directly answers questions, and you’ll perform well across all search platforms.
Start with your technical foundation. Layer in AI optimisation strategies. Track what’s working. The businesses that adapt now will dominate search for years to come.
Traditional SEO built on keywords and backlinks. Modern SEO builds on clarity, context, and comprehensive answers. Master both, and you win.
