Get cited, not clicked

For twenty years the goal was simple: rank the link, earn the click. But when an AI engine answers a question inline, there’s no link to click — there’s a sentence with a citation. The citation is the new click. Optimizing for it is called Generative Engine Optimization (GEO), or Answer Engine Optimization (AEO). Same idea.

This isn’t a fringe concern. AI engines — ChatGPT Search, Perplexity, Gemini, Copilot, AI Overviews — now handle an estimated 12–18% of English-language informational queries, up from under 2% a year earlier, and AI-referred sessions have grown roughly 527% year-over-year (Enrich Labs). The competition for citations is still young — which is exactly why it’s worth moving now.

Why a citation is worth chasing

Even with no click, a citation does two things a ranking never could:

• It builds your brand inside the answer. “According to AdsenseThemes…” is an impression on the highest-trust surface there is. Branded search is the one query type still growing.
• It compounds. Models develop source preferences; sites cited frequently get cited more. A citation advantage built now is hard to dislodge later.

What AI engines actually reward

The patterns are consistent across studies:

1. Answer first. Put a direct, quotable answer in the first 2–3 sentences. ChatGPT and AI Overviews lift clean definitions and bullet lists almost verbatim.
2. Structure for extraction. FAQs, comparison tables, and step lists are machine-liftable. (This is the same structure screen readers want — accessibility and GEO are one job.)
3. Freshness. Content updated within 30 days earns roughly 3.2× more AI citations than stale pages. Maintain “living” references with a real Updated date.
4. Get indexed in Bing. It’s a non-negotiable prerequisite for appearing in ChatGPT’s cited sources.
5. Be where the models look. Community platforms account for a striking share of citations across engines — Reddit alone is a heavy source for Perplexity — so meaningful participation in relevant communities pays off.

A practical checklist

• Lead every key page with a one-paragraph answer.
• Format answers as clean Q&A for humans and machines. (Google retired FAQ rich results in May 2026 — the schema still validates and is still parsed, but it no longer earns a SERP feature, so use it only where the FAQ is visible and genuinely useful, never as a trick.)
• Turn data into tables an AI can quote.
• Stamp and genuinely update your evergreen pages.
• Confirm you’re indexed in Bing, not just Google.
• Cite your own primary data — the thing no competitor and no model can reproduce.

Google’s own line: it’s still SEO

Google’s May 2026 AI optimization guidance is blunt: its generative features are “rooted in our core Search ranking and quality systems,” and you do not need llms.txt, content chunking, AI-specific rewrites, or special schema to appear in them (Search Engine Journal). There’s no machine-only shortcut — the work that earns rankings is the work that earns citations.

That’s the freeing part: you optimize for one thing, not two. Clean structure, direct answers, real authority, and freshness serve the human reader, the screen reader, and the model at the same time. The only way to get this wrong is to chase machine tricks at the cost of readability.

Measure it

As of June 2026, Google Search Console has new Search Generative AI performance reports showing how often your pages appear in AI Overviews and AI Mode — impressions by page, country, and device (clicks aren’t there yet) (Google). There’s also a toggle to opt out of generative AI features; leave it on — being cited is the goal.

The bottom line

GEO isn’t a replacement for good content; it’s good content shaped so a machine can quote it without distorting it. Win the citation and you build the brand and authority that — eventually — bring the direct, durable traffic the algorithm can’t take away.