Every few years, a new technology reshapes search, and a wave of self-proclaimed experts rushes in to sell you a brand-new playbook. Right now, that technology is AI search (AEO, GEO, whatever you want to call it).
LinkedIn is overrun with posts promising that llms.txt is the new robots.txt, and that pumping out hundreds of AI-generated articles will somehow turn your brand into an AI citation machine. Most of it is hype. I’m here to tell you what’s hype and what isn’t.
Most AEO advice is worthless, and gets priorities completely mixed up. The same SEO fundamentals that drive Google rankings are largely what drive AI visibility, as boring as that may sound. That’s where you should start. If your content isn't ranking on page one of Google, it's almost certainly not showing up in AI search either.
Before chasing AI-specific hacks, you need a solid SEO foundation. The best practices below are built on that reality, starting first with fundamentals and expanding to more experimental and AI-specific practices …
What Is AEO & Why Does It Matter In 2026?
AEO is the practice of optimizing your content and brand presence so that AI mentions you more often, ranks you higher in AI-generated answers, and represents your brand both accurately and positively.
Where traditional SEO’s goal is to rank higher in Google SERPs, AEO is about showing up favorably across a full range of AI answers, whether that means being cited as a source or recommended as a solution.
AI Overviews now appear on approximately 48% of all Google searches as of early 2026, with some trackers putting that figure closer to 60% depending on methodology and keyword set.
But before you change up your strategy to focus entirely on AI, context matters:
- Google still processes around 210 times more searches than ChatGPT daily (SparkToro)
- AI tools account for only about 1% of website referral traffic (Conductor's 2026 AEO/GEO Benchmarks Report)
AI search is growing fast, but Google still drives the overwhelming majority of organic traffic for most websites.
AEO doesn't require a completely new playbook. Google and AI platforms share the same core goal: deliver the most helpful, trustworthy answer to a user's query.
The signals that earn Google's trust are largely the same signals that earn an AI model's trust. That's your foundation. Everything else builds on top of it.
AEO Best Practices For 2026
1. Make sure AI crawlers can access your site
Before anything else, confirm that AI crawlers can access your website. This is the most basic technical prereq, and two issues are worth flagging:
First, Cloudflare's Bot Fight Mode blocks AI crawlers by default
If you've added a new domain to Cloudflare recently, there's a chance you're blocking GPTBot, ClaudeBot, and PerplexityBot without knowing it. Audit your robots.txt and check your CDN and WAF settings for all major AI crawlers. If any are disallowed, fix it immediately.
Second, AI crawlers don't execute JavaScript
If your content is rendered client-side, AI crawlers are skipping it entirely, seeing only what's in the initial HTML. This isn't a new problem for SEO, but AI crawlers are less forgiving than Googlebot.
2. Rank in Google first
The single highest-leverage AEO move you can make is to get your content to rank highly in Google.
According to Grow and Convert's analysis of 400+ keywords across 16 clients, page-one Google rankings predicted appearing in ChatGPT and Perplexity 77% of the time. A November 2025 study by Originality.AI also found that ranking #1 gives a page a 58% chance of being cited in an AI Overview, and roughly 9 in 10 organic-style AI citations come from the top 30 results.
Why?
This is likely because Google's AI Overviews and AI Mode are grounded in Google's core search index.
Google itself has confirmed it:
"From Google Search's perspective, optimizing for generative AI search is optimizing for the search experience, and thus still SEO."
Additionally, AI platforms use query fan-outs, splitting a user's query into clusters of related sub-searches, and pull citations from pages that appear most often across those results. If you're not ranking across a cluster of related searches, you're largely invisible to the entire process.
3. Target the right keywords
Top-of-funnel "what is" topics may have high search volume, but they're dominated by AI Overviews that rarely cite external sources. According to Ahrefs' analysis of 300,000 keywords, 99.2% of keywords triggering Google AI Overviews are informational in intent. AI easily handles definitions. No citation is needed.
Instead, prioritize mid-funnel and high-intent keywords:
- Comparison searches
- Alternative searches
- Best-of lists
- How-to content tied to specific use cases
These are the searches where AI models are more likely to surface external sources because they require more nuanced, experience-backed answers.
This tiered framework will help you prioritize which topics to cover:
- Tier 1 (Prioritize). These formats target buyers actively searching for and/or evaluating specific tools. They’ve identified that they have a problem and are seeking a solution. These deliver the highest return.
- Tier 2 (Develop Selectively). How-to guides, best practices, and checklists attract mid-funnel readers who are building their knowledge. They build brand authority and help position your brand as a thought leader.
- Tier 3 (Limit). Definition and high-level content serves early-funnel audiences with low buying intent. Google AI Overviews dominate these searches. Produce this content only when it serves a specific internal linking or topical authority purpose.
4. Write answer-first
Lead every key section with a direct, self-contained answer in the first 40–60 words, before any context-setting or background explanation.
AI models heavily weight the opening of a document during retrieval. Kevin Indig's February 2026 analysis of 3 million ChatGPT responses found that 44.2% of citations come from the first 30% of a page's content.
This runs counter to how most marketers have been trained to write. The instinct is to warm readers up, build context, establish stakes, then deliver an answer.
Flip that structure. Give AI something clean to extract right away. Everything that follows can add depth, examples, and nuance, but the core answer needs to come first.
5. Use clear structure and formatting
A March 2026 peer-reviewed study from the University of Tokyo and University of Tsukuba found that structural changes alone, independent of content quality, produced a 17.3% improvement in citation rates across six generative engines.
Short paragraphs, clear headers (H2/H3 heading hierarchy) that signal what each section covers, and tabular data all reduce the friction between your content and an AI citation.
These choices also make content easier for human readers to scan and use. Good content structure has always served both audiences.
6. Satisfy Google's E-E-A-T framework
E-E-A-T stands for Experience, Expertise, Authoritativeness, and Trustworthiness, the framework Google uses to evaluate content quality. Content that scores poorly on E-E-A-T signals doesn't rank well in Google, and content that doesn't rank well in Google rarely appears in AI search.
Every piece of content you publish should be attributed to a named, credible author with verifiable expertise. Anonymous content signals lower trust to both search engines and AI models.
More importantly, the "Experience" component (first-hand, personal involvement with the topic) is the one signal AI cannot replicate.
A practitioner writing from real-world experience about their own field is going to produce content that is categorically different from a language model generating text about that same field.
7. Include specific data, named sources, and attributed claims
Vague claims get skipped. Specific, sourced claims get cited.
Research by Kevin Indig found that typical English text contains 5–8% proper nouns, while heavily AI-cited content averaged 20.6%.
Specific brands, named tools, real people, and attributed data points anchor AI answers and reduce ambiguity. A model looking for a reliable source on a topic will gravitate toward the page that says "according to Ahrefs' study of 17 million citations" rather than the one that says "research shows."
This doesn't mean cramming in data. Every statistic and cited claim should add substance and clarity to the point you're making. But when you have access to specific numbers, named sources, and real evidence, use them.
Original research and proprietary data are especially powerful here, because they give AI models something they can't find anywhere else.
8. Build off-site brand authority
Ahrefs studied 75,000 brands and found that brand web mentions correlated three times more strongly with AI visibility than backlinks (correlation of 0.664 versus 0.218). Brands in the top 25% for web mentions averaged roughly 169 AI Overview mentions, compared to about 14 for the next quartile. That's over 10x more visibility.
AI models build trust through corroboration. If only your own website says you're the best solution for a problem, an AI model treats that as a marketing claim and weighs it accordingly. When independent, credible sources mention your brand in relevant contexts, the model treats it as validated.
The channels that matter most, based on where AI models actually cite sources, are Reddit (the most cited domain across major AI platforms according to Peec AI's March 2026 analysis of 30 million sources), YouTube, LinkedIn, earned media and digital PR, and third-party review platforms like G2, Capterra, and Trustpilot.
However, citation sources can vary widely by industry. You should research and track the biggest citation sources within your own space.
9. Keep content fresh
AI-cited content is fresher than traditionally ranked content. Ahrefs' analysis of 17 million AI citations found that 76.4% of ChatGPT's most-cited pages were updated within the last 30 days.
Seer Interactive's crawl analysis found that 65% of AI crawler hits target content published within the past year, and only 6% targets content older than six years.
Outdated content is being filtered out of retrieval pools regardless of how well it ranks, which makes regular content refreshes one of the highest-ROI AEO maintenance tasks.
This doesn't mean rewriting content from scratch on a monthly basis. It means reviewing high-value pages regularly and updating them with new data, current statistics, and fresh examples.
A page that was thorough and well-structured when it was published two years ago, but hasn't been touched since, is competing against pages that were updated last month.
10. Track AI and Google performance together
Most brands are measuring Google rankings in one tool, AI visibility in another, and never connecting either to actual conversions. That siloed view makes it impossible to know what's working.
The goal is a single unified view: organic traffic, AI referral traffic, combined traffic, and inbound conversions together.
A clean dashboard that connects AI referral sources to conversions is how you know whether your AEO efforts are actually moving the needle.
You need to be able to answer:
“Is traffic going up and to the right, and is that traffic actually leading to qualified inbound conversions?”
I wrote a whole blog post on content reporting, which metrics you should be following and how to create a dashboard to monitor performance. Check it out here.
AEO Best Practices To Skip
Just as valuable as knowing what to focus on is knowing which AEO tactics aren’t worth your time. Two in particular get promoted aggressively and have the data working against them.
1. FAQPage schema for citation improvement
This one has been a cornerstone of AEO advice, but the evidence shows that it has little impact on AI search visibility.
Ahrefs' May 2026 study tracked 1,885 pages that added JSON-LD schema against 4,000 control pages and found no statistically significant citation improvement across any platform, and a small but notable 4.6% decline in AI Overviews.
To make things worse for schema advocates, Google deprecated FAQ rich results entirely in May 2026, removing the visible SERP reward that made FAQPage markup worth implementing in the first place.
Schema isn't worthless across the board. Organization, Article, and Product markup still serve legitimate purposes. But adding FAQPage schema specifically to boost AI citations is not supported by evidence.
2. Mass AI content generation for "topical authority"
AI platforms use fan-out queries that scan clusters of related searches, so producing dozens of articles covering every variation of a topic should help your brand appear across those clusters. The problem is the execution.
Google's Quality Rater Guidelines explicitly classify scaled content produced with AI as a "low-effort way to generate many pages that add little-to-no value," and that content is classified as spam under Google's Web Spam Policies.
Content that doesn't rank doesn't show up in fan-out search results. Content that doesn't show up in fan-out results isn't cited. Publishing 50 thin articles to "capture a cluster" doesn't build topical authority. It builds a content graveyard.
Start With The Fundamentals
The brands winning in AI search didn't find a new hack. They have strong SEO foundations, high-quality content that real experts wrote or contributed to, and off-site authority built through consistent presence on the channels that matter (Industry blogs, top-tier publications, G2, Reddit, LinkedIn, etc.). That's it.
The best practices above aren't a departure from what good SEO has always required. They're an extension of it.
The goal of AI search and traditional search is the same: provide the most helpful, trustworthy answer to a user's query.
The brands that do that consistently, across content quality, site structure, off-site presence, and keyword strategy, are the ones showing up in AI answers.
If you want hands-on help building that foundation, I work with SaaS brands specifically on this. Get in touch with me here.
.svg)
.svg)