AEO 101: What it is, why it matters, and how to start
AI referred website traffic is still small - but it's growing fast - and converts at a much higher rate than traditional search volume
Hi friends! Happy Sunday. Today, I’m going to talk about AEO (answer engine optimization). I’ve been reading, and listening to podcasts, and tinkering for months trying to decode this buzzword and form a POV.
Now I have some receipts (and a POV) and I am ready to share them with you!
One honest caveat up front: this is a brand-new space, and nobody is a true expert yet - not me, not the agencies, not the vendors. We’re all still figuring out how these models pick sources and answers.
But what we do know definitively is that usage is climbing fast and the traffic converts unusually well. That’s enough to make it worth your attention now, even while the playbook is still being written. Take everything here as informed early advice, not gospel.
And I have no doubt - this won’t be the last time I talk about this subject. I’ll continue to share as my understanding and opinions evolve. Let’s go!
AEO 101: What it is, why it matters, and how to start
For the last decade, SEO has basically meant one thing: rank on Google. And that’s changing.
ChatGPT now processes around 2.5 billion prompts per day and serves roughly 800 million weekly active users — about 12% of Google’s search volume when you filter for search-like queries. Claude doesn’t publish a comparable prompt count, but Anthropic is now estimated at north of 18 million monthly active users on web alone. And Gartner predicts traditional search query volume (Google) will decline 25% by end of 2026 as users shift to AI for answers.
But ALSO, as of Feb 2026 -
For every search-like query in ChatGPT, there are 8.33 searches in Google.
Google sends 190x more traffic to websites than ChatGPT.
In other words, Google is still important and dominant (for now) AND a meaningful and growing chunk of your buyers’ research is also happening inside ChatGPT, Perplexity, Claude, Gemini, and Google’s AI Overviews.
So this week we’ll break down AEO (Answer Engine Optimization). What it is, how it relates to the SEO you may already be doing, and how to get started without setting your existing strategy on fire.
The case for paying attention now
The volume of AI referral traffic to most sites is still small — Conductor’s 2026 benchmark (3.3 billion sessions across 13,770 domains) puts it at about 1.08% of total web traffic. Roughly 1 in 100 visitors. Easy to dismiss.
BUT…that 1% slice converts at dramatically higher rates than organic search:
Ahrefs found AI search drove 12.1% of signups from just 0.5% of visitors — a 23x conversion multiplier (June 2025).
Search Engine Land’s 13-month analysis showed LLM referrals converting around 18% — the highest of any source, including paid (February 2026).
Shopify reported AI-referred shoppers convert nearly 50% higher with 14% higher order values than organic (Q1 2026).
Here’s why I think the gap exists
When someone Googles “best onboarding software,” they type a couple keywords and scan blue links — they’ve barely articulated what they need.
When that same person opens ChatGPT/Claude, they have a full conversation: team size, tech stack, the workflow they’re fixing, budget, what they’ve already tried. They ask follow-up questions. By the time they land on your site, they’ve effectively run a 15-minute discovery call with themselves. They’re not researching anymore — they’re ready to evaluate. That pre-qualification is the whole story.
One caveat: this premium concentrates in research-heavy journeys (hint: B2B).
A study of 973 e-commerce sites ($20B combined revenue) found organic search actually beating ChatGPT by ~13% for purely transactional purchases. Impulse-buy products? More nuanced. B2B software with long research cycles? Very compelling.
The takeaway: AEO is a high-leverage, low-volume channel today, trending toward high-leverage and high-volume. Being an early adopter is one of the better asymmetric bets you can make right now.
What you can and can’t see
An important limitation to sit with. AI search splits into two questions, and only one of them is visible to you.
Demand is invisible
With traditional SEO, you have Search Console, keyword tools, and volume estimates telling you how many people search a term, how it’s trending, how you rank, AND how many visitors showed up on your site by typing that specific keyword.
None of that exists for AI. There is no Keyword Planner for ChatGPT — you cannot see how many people asked Claude about your category last month or what phrasing they used. The platforms don’t expose it (yet).
Your presence is measurable — but only against prompts you supply
What tools (and manual testing) can show you is whether your brand appears for a specific set of prompts you define. That’s genuinely useful, but note the limit: you’re guessing at the questions, then measuring how you show up against your own guesses — not against what we know the world is actually asking. So plan to test, observe, and adjust rather than optimize against a known demand curve.
What AEO actually is
AEO is the practice of structuring your content so AI tools pull from it when answering a buyer’s question. When someone asks AI for the “best onboarding software for early-stage SaaS startups?”, AEO is what determines whether your brand shows up in that synthesized answer.
You’ll hear related acronyms — GEO, LLMO, AIO. The tactics overlap heavily; I’m using AEO as the umbrella term.
One mechanic worth understanding: query fan-out.
When you ask an AI a question, it often generates related sub-queries behind the scenes before answering — “how to fix a weedy lawn” might fan out into “best herbicides,” “remove weeds without chemicals,” “how to prevent weeds.”
The implication: covering the full cluster of questions around your topic, not just one headline keyword, gives you more surface area to get pulled into answers.
How AEO differs from SEO (and how it doesn’t)
The fundamentals are very similar — quality content, technical hygiene, and authority. The goal differs: SEO is about ranking #1 so someone clicks; AEO is about being the answer, click or not.
They overlap on the fundamentals — and 76% of AI-cited URLs come from top 10 organic results, so solid SEO is still part of the foundation.
Where they diverge: AEO purportedly rewards extractability (easy-to-parse structure), third-party authority (85% of AI brand mentions come from third-party pages, not your own domain), and freshness (recently updated pages earn 28% more citations).
You don’t pick one. AEO layers on top of a strong SEO foundation.
The vendors and Google don’t fully agree
Here’s where that “nobody’s an expert yet” caveat becomes very evident.
Most of the circulating AEO playbooks come from vendors and agencies, and they emphasize structure: extractable chunks, FAQ schema, answer blocks. There’s real research behind some of it — Princeton’s GEO study found expert quotes, statistics, and inline citations measurably increase how often AI engines cite you.
But Google — the OG company also running AI Overviews — published guidance pushing back on a lot of this. Their position: optimizing for their AI features is SEO, and several popular “hacks” are unnecessary.
So who’s right? Probably both, I’m guessing it varies by engine. Google’s guidance covers Google’s surfaces; ChatGPT, Perplexity, and Claude each retrieve and rank sources in their own way.
The honest synthesis: don’t obsess over structural hacks — they won’t hurt, but they’re not the main event — and focus on the one thing everyone agrees on: unique and high-quality content.
Google’s strongest stated signal is non-commodity content — unique, first-hand, experience-based writing over generic restatement. Their example nails it: “Why we waived the inspection and saved money: a look inside the sewer line” beats “7 tips for first-time homebuyers.”
For founders, this is the most actionable insight in the whole space: the content most likely to get cited is the stuff drawing on what you uniquely know — real customer stories, your contrarian take, your data. AI engines are drowning in generic content; the experience-led stuff is what they reach for.
Why startups can win at AEO faster than at SEO
Traditional SEO is a domain authority game — years of backlinks, content, and brand mentions. A new company can’t win that head-to-head fight against a 15-year-old incumbent in its first two years.
AEO breaks that moat.
AI tools care less about how long your domain has existed and more about whether you’re cited as relevant right now.
As Ethan Smith of Graphite put it on Lenny’s podcast: a startup mentioned in a Reddit thread today can show up in ChatGPT tomorrow.
A handful of strong third-party citations can put you in answer engines almost immediately — no 200-post backlog or three years of domain age required. For the first time in a while, the time-to-value curve actually favors smaller, faster-moving companies. That’s a multivitamin worth taking seriously.
Practical steps to get started
1. Audit how AI tools talk about you. Open ChatGPT, Perplexity, Claude, and Gemini. Run 10-15 prompts your buyers would actually use — “best [category] tool for [ICP],” “[your brand] vs [competitor].” Note who gets cited and where you don’t show up. Since you can’t see real demand, this manual testing is your data. But I am suspicious of the answers when we ask in the same workspace we have been using to run our marketing and business…
For a faster baseline, the free HubSpot AEO Grader scores you across the major engines in minutes.
2. Create non-commodity content, then make it easy to parse. Note the order — quality first, structure second, which is where vendor and Google advice actually converge. Write the things only you can write: real customer stories, your POV, your own data. Then make your top pages extractable: lead with a clear answer in the first 100 words (44% of citations come from the first 30% of the article), use descriptive H2s that mirror how buyers ask questions, cover the full question cluster (remember fan-out), and cite specific stats with attribution. Just don’t contort content purely for machines — your human readers come first.
3. Invest in third-party presence — especially your own thought leadership. AI tools heavily weight what other sites say about you. Get listed and reviewed on G2 and Capterra (here’s exactly how), pitch yourself onto 3-5 podcasts your ICP listens to (scrappy PR starter kit), contribute guest posts, and be genuinely helpful on Reddit and Quora — genuinely being the operative word, since Google explicitly warns against inauthentic mentions.
But here’s the move I’d prioritize above all the others, because it validates something I’ve been preaching for years: founder brand works, and AEO just enhanced its value with data. Profound’s analysis of 1.4 million citations found LinkedIn became the #1 most-cited domain for professional queries across all six major AI platforms between late 2025 and early 2026 — and the citations increasingly come from published posts and articles, not static profile pages.
The AI engines are reaching straight for exactly the kind of opinionated, experience-led content that good thought leadership has always been. Personal branding (aka thought leadership) was already one of the highest-ROI things a founder could do for pipeline and trust; now it’s also one of the most direct paths into AI answers.
One caveat, consistent with everything else here: Profound’s statistic is based on a three-month data window in a fast-shifting space — directionally strong, not carved in stone.
Founder brand is an even stronger GTM multivitamin now: third-party presence drives backlinks, builds awareness, warms accounts, and now also feeds AEO.
4. Add FAQ schema — but keep it in perspective. It makes content easier to parse and keeps you eligible for rich results. I also personally find FAQs helpful as a human website visitor. Just don’t treat it as “the thing” that wins AEO; Google says outright it isn’t required for them.
5. Set up basic measurement. Most startups don’t have the bandwidth for a custom GA4 channel grouping or paid Semrush license — and don’t need to. If you’re on HubSpot, HubSpot AEO (recently embedded in the platform) tracks your visibility week over week across the major engines, compares competitors, and gives prioritized recommendations. Just remember what it’s actually measuring: your presence against the prompts you feed it, not confirmed real-world demand.
6. Don’t abandon SEO. It still drives the vast majority of traffic, it’s one of the best predictors of AI citation, and — as Google frames it — optimizing for AI search largely is SEO. AEO is an additive layer, not a replacement.
The takeaway
For most B2B startups, AEO could pay off disproportionately for the effort: the volume isn’t dramatic yet, but the conversion quality is exceptional, the volume is growing, and the smaller players can actually win quickly.
It’s also genuinely early. We can’t see AI demand the way we can on Google and the playbook is being written and rewritten in real time. None of that is a reason to wait — it’s a reason to start experimenting now, while it’s cheap to learn and the field is wide open.
Start with a baseline audit, write the content only you can write, and treat third-party presence as a real priority. Which candidly has been good advice before we ever had the acronym AEO.
See ya next week!
Resources I recommend
🎧 The ultimate guide to AEO with Ethan Smith (Graphite) — Lenny’s Podcast. One of the best episodes I’ve heard on this topic.
📄 Google’s official generative AI search guide — the primary-source counterpoint to much of the vendor advice. Worth reading directly.
🛠️ HubSpot AEO Grader (free) — fastest way to benchmark where you stand.
With love and gratitude -
If you want to learn more about working with me directly…
For B2B startups, we serve as your Fractional GTM executive or engineer. Learn more about -
When you’re ready, let’s connect to discuss your specific growth goals and challenges.
Subscribe for weekly education, ideas, and frameworks
In this newsletter I share the exact tips, playbooks, and GTM multi-vitamins I’ve used to help 30+ B2B startups scale their revenue 150-590% YoY.