SEO in the age of AI search — what actually changed

For twenty years, SEO was about ranking. Now it's about being cited. That single shift changes almost everything downstream.

The three surfaces that matter

1. Google search — still the biggest, still the most valuable, but increasingly answered inside the AI Overview.
2. AI answer engines (ChatGPT, Perplexity, Claude, Gemini) — smaller traffic, higher intent, growing fast.
3. Vertical AI tools — model-powered assistants inside SaaS products that quietly cite sources.

Your content needs to show up in all three. The techniques overlap, but the tactics differ.

What still works

• Deep, primary-research content
• Clear H-tag hierarchy and clean semantic HTML
• Internal linking that maps to entity clusters
• Fast, mobile-friendly pages with real Core Web Vitals

What's changed

• Thin "SEO content" is worse than useless — it actively hurts you
• FAQ schema and clear question-answer structure now drive AI citations
• Brand mentions across the open web (Reddit, YouTube transcripts, podcast notes) feed retrieval

Where to invest first

If you have one quarter and one budget, spend it here:

1. Audit the ten pages that drive 80% of your revenue
2. Rewrite them for extractability — short, scannable, quote-worthy paragraphs
3. Add primary research or original data that answer engines will cite
4. Build brand mentions in the sources those engines actually retrieve from

The rest is optimization on the margin.