AEO and GEO Content Strategy: I Thought It Was Just SEO With a Fancier Name. Then October 2024 Happened.

When I first heard "Answer Engine Optimization," my initial reaction was skepticism. Like, is this another marketing term someone invented to sell a course? I'd been writing and optimizing content for years, and SEO had always been the framework everything else orbited around. I wasn't about to throw that out for a buzzword.

Then October 2024 came around and Google rolled out AI Overview in Asia. And I watched a website I was managing start doing something strange in Google Search Console.

Impressions went up. Rankings collapsed. Clicks dropped. CTR tanked. All at the same time.

That's when I realized something had genuinely shifted.

This is my honest account of building an AEO and GEO content strategy from scratch while the ground kept shifting underneath it.

First, Let Me Catch You Up on All These Acronyms

If you're new to this space, the alphabet soup is real. I spent weeks getting confused between AEO, GEO, AIO, and SXO. Here's how I understand each one now:

Term	What It Means	What It's Actually About
SEO	Search Engine Optimization	Getting your pages to rank on Google and other search engines through keywords, backlinks, and technical structure
AEO	Answer Engine Optimization	Optimizing content to be the direct answer returned by AI assistants, voice search, and answer boxes. One question, one answer. You want to be it.
GEO	Generative Engine Optimization	Optimizing to be cited or referenced inside AI-generated responses across platforms like ChatGPT, Perplexity, and Google AI Overview, which pull from multiple sources to build one reply.
AIO	AI Overview or Artificial Intelligence Optimization	Google's AI-generated summary at the top of search results that answers a query without requiring a click. Side note: some in the industry also use AIO to mean Artificial Intelligence Optimization. The definition hasn't fully settled, so seeing it used two different ways depending on the article is normal.
SXO	Search Experience Optimization	Making sure your website is accessible, fast, and readable both by users AND by bots, including AI crawlers

They're not competing strategies. They're layers. SEO is still the foundation. AEO and GEO sit on top. SXO is what makes all of them possible.

That took me longer to figure out than I'd like to admit.

My Initial Take: SEO Is Still the Foundation (I Was Right About This Part)

When AEO started getting buzz, a lot of people in the industry were saying SEO was dying. I didn't buy that. And looking back, I think that instinct was correct, mostly.

Here's where I landed after months of testing and observing:

• Technical SEO is still non-negotiable. If your site can't be crawled, nothing else matters.

• Keyword research still drives what topics you target, even for AEO.

• Backlinks still build the authority that makes AI tools trust your content enough to cite it.

• On-page structure (headers, metadata, schema) is still how bots understand what your page is about.

What changed is the destination. Before, you were optimizing for a rank. Now, you're also optimizing to become a source. Those are two different goals, and they sometimes require different decisions.

What October 2024 Actually Looked Like in My Data

Here's where I stop being abstract. I'm sharing real GSC numbers because I think seeing the actual shift is more useful than me describing it vaguely.

Clicks dropped about 30%. CTR dropped by a third. Average position fell from 3.3 to 6.3, which sounds small but represents a significant ranking collapse for keywords that had been stable for months.

Impressions went up. From 27K to 28.5K.

More visibility, fewer clicks. This is unusual for my client's website, especially since it's consistently ranked #1. Why would our website get seen more but clicked less this time?

The answer came in two parts.

Part 1: The Mystery Traffic Was AI Bots

The impressions spike wasn't from real users discovering the site. A big chunk of it was AI crawlers, specifically bots from ChatGPT, Perplexity, and other AI search engines, showing up in the data. They were visiting pages, indexing content, and inflating impression counts without converting into any actual human traffic.

Once I tracked this properly in GA4, it became much clearer. Here's how to set that up if you haven't yet:

How to track AI referral traffic in GA4:

1. Open GA4 and go to Reports > Acquisition > Traffic Acquisition

2. In the top filter, click Add filter and select Session source/medium

3. Search for known AI referrers: chatgpt.com, perplexity.ai, claude.ai, bing.com/chat

4. You can also create a custom channel group under Admin > Data Display > Channel Groups

5. Add a new channel and set the condition to Session source contains chatgpt OR perplexity OR claude OR gemini

6. Name it "AI Referral" and save

Once I had this set up, I could see exactly how much traffic was coming from AI platforms vs. organic search. Some weeks, AI referral traffic was a meaningful portion of the total, but the session quality (pages per session, time on site, conversions) was noticeably different from traditional organic.

Part 2: People Are Doing Their Discovery Inside AI Now

The second part of the clicks drop is about user behavior, not bots.

When AI Overview gives someone a solid answer at the top of the page, they don't need to click through to a website to get what they came for. The entire awareness and consideration stage of their research is happening inside a single AI response. By the time they click on a link, they've already compared options, read summaries, and made up their mind on what they're looking for.

Think about what the traditional marketing funnel looked like before AI Overview:

• Awareness: Someone searches "best event spaces in Singapore" and lands on a blog post listing 10 options. They read, they browse, they form opinions. That happens on your website.

• Consideration: They come back, compare a few venues, maybe read reviews. Still on your website or others in your space.

• Decision/Conversion: They contact a venue, request a quote, or book.

Now that funnel looks like this:

• Awareness + Consideration: Someone asks AI Overview "what are the best event spaces in Singapore for a corporate dinner for 40 people." They get a structured, cited answer with options, price ranges, and context. All of that happens inside Google. They never visit a website during this stage.

• Decision/Conversion: They click one link. They're already close to converting when they arrive.

This is actually a good thing if you're the source being cited. The visitor who lands on your page from an AI citation is warmer than the visitor who found you through a regular organic result. They've already been pre-sold to some extent by the AI summary. So yes, lower traffic volume, but the intent of the remaining clicks is higher.

The scary part? If you're not being cited at all during that awareness and consideration stage, you've been removed from the consideration set before anyone even visits your site. The user never had a chance to discover you.

That's what makes AEO a brand visibility issue, not just an SEO issue.

How I Track AI Overview Visibility (The Ahrefs Method)

Before I get into the steps, a quick note on where the formula comes from. I watched Neil Patel's webinar "The Great Google Reset" (NP Digital) and one of the slides covered how to measure AI visibility in the new era of search. He presented a broader AI Visibility Score formula meant to track how often a site appears across all major AI platforms:

AI Visibility Score = (Appearances in AI / Total Relevant Queries) x 100

He also laid out complementary metrics like Citation Frequency, Entity Mention Velocity, and Cross-Platform Presence, basically a full measurement framework for the AI era vs. the traditional traffic era.

The honest caveat: my client only has Ahrefs, and Ahrefs currently gives me reliable AI Overview data for Google only, not cross-platform appearances across ChatGPT, Perplexity, and the others. So I adapted the formula specifically for AI Overview visibility, which is a narrower but still useful slice of the full picture.

Here's how I calculate it using Ahrefs:

1. Go to Ahrefs and open the site dashboard

2. Click Organic Keywords under Organic Search

3. Click Add Filter > SERP Features > AI Overview > Apply

4. This shows all keywords the site ranks for that also trigger an AI Overview. Record this number. This is your AI Overview Exposure count.

5. Go back to the filter and this time select Include Target In instead of Include

6. This now shows keywords where your site is actually cited inside the AI Overview, not just ranking below it. Record this second number. This is your AI Citations count.

7. Calculate: (AI Citations / AI Overview Exposure) x 100 = AI Overview Visibility Rate

So if your site ranks for 411 keywords that trigger an AI Overview and is cited inside 52 of them, your AI Overview Visibility Rate is roughly 12.6%. That's the number I track week over week to see if the content and credibility work is actually moving the needle.

For multi-platform AI citations, also in Ahrefs

Under the site Overview dashboard, there's an AI Citations section that shows citation counts across five platforms:

• Google AI Overview

• ChatGPT

• Perplexity

• Gemini

• Copilot

I still think there are better tools out there for this. On a separate client, I've been exploring Peec AI, which I like so far because it shows you competitor brand mentions on pages where you're not being cited. That's a gap analysis angle that Ahrefs doesn't give you as directly. I'll share a more detailed review once I've used it longer.

How Each AI Engine Decides What to Cite (They're All Different)

This is the part that makes AEO genuinely complicated. There's no single algorithm to reverse-engineer because each platform has its own logic.

Here's what I've observed so far:

• Google AI Overview tends to favor pages that already rank well organically. If you have strong SEO, you have a head start. It also favors structured, answer-first content with clear headers.

• ChatGPT (via Bing) seems to pull from a mix of Bing-indexed pages and sources with strong domain authority. It also references pages that directly and comprehensively answer a specific question rather than covering a topic broadly.

• Perplexity is the most citation-heavy of the group. It pulls from multiple sources per answer and tends to favor pages with fresh, specific data. It's also more likely to cite niche but authoritative sources.

• Gemini is deeply integrated with Google's index, so similar signals to AI Overview apply, but it seems to weight content recency and E-E-A-T signals (experience, expertise, authoritativeness, trustworthiness) more visibly.

• Copilot (Microsoft) relies heavily on Bing search data and tends to favor well-structured pages from established domains.

• Claude (Anthropic) is interesting because it primarily relies on its training data for responses, but when web search is enabled (which it is on claude.ai), it actively fetches and cites real-time sources. This means Claude's citation behavior depends on the context it's being used in. When browsing, it tends to favor pages that are clearly structured, well-sourced, and easy to parse. Cloudflare blocks and JavaScript-heavy pages are a real problem here too, same issue as with other bots.

What this means practically: a page that does well on Google AI Overview won't automatically perform the same way on Perplexity. The fundamentals overlap, but the weightings are different enough that you have to think about all of them when planning content.

My AEO and GEO Content Strategy Right Now (And What I'm Still Figuring Out)

I want to be transparent here. It's still early. Anyone claiming to have a fully proven AEO playbook right now is probably oversimplifying it. But here's what I've been testing with observable results:

Content structure

• Answer-first writing: The first two to three sentences of every section directly answer the question raised by the header above it. This matches how AI tools scan for answers.

• Question-based headers: Headers written as questions mirror how people actually search and how AI interprets intent.

• Compound query targeting: Instead of targeting a single keyword, I look for topics that combine two or more requirements. Most pages on the web target one thing at a time. A page that comprehensively addresses a compound query (like "best event spaces in Singapore for corporate team dinners under 50 people") becomes the single resource an AI can cite rather than having to pull from three different pages.

Credibility signals

• Author bios with real credentials and links to social profiles

• Citing primary sources within the content (studies, official data, named experts)

• Publishing original data or insights that no other page has, even if it's just internal analysis you're comfortable sharing publicly

• Clear "about this content" context at the top of pages so AI knows who wrote it and why they're qualified

On-page SEO (still very much relevant)

• Proper internal and external linking, because how you connect pages still affects how bots understand your site structure

• Clean header hierarchy (H1, H2, H3 in order) so both search engines and AI can parse your content outline

• Metadata optimization: title tags and meta descriptions still influence how AI tools label and summarize your pages

• Alt text on images, because that's content too and it's one of the few things bots can always read

• Schema markup, especially FAQ, HowTo, and Article schemas, which give AI structured signals about what your content contains

Off-page SEO (still very much relevant)

• Backlink building, because authority signals from external sources are still one of the strongest trust signals for AI citation decisions. A page with credible backlinks is more likely to be cited than one without.

The SXO Problem Nobody Talks About Enough

All of the content and optimization work above assumes that AI bots can actually read your website. A lot of sites, including one I'm currently working on, have a serious crawlability problem that makes everything else less effective.

The site relies heavily on JavaScript rendering. What this means is that when an AI bot (or Googlebot, for that matter) does a raw HTML fetch of the page, the content isn't there yet. The page looks empty because the content only loads after JavaScript runs. AI crawlers, and many search engine bots, don't wait around for JavaScript to execute. They read what's in the raw HTML and move on.

I diagnosed this by doing something simple. I used AI tools to fetch and read the site the same way a bot would, then compared what they saw versus what a human user sees in a browser. The gap was significant. Whole sections of content, internal links, schema markup, and metadata were invisible to the initial crawl.

Some other issues I found using that same method:

• Cloudflare security settings blocking certain bots from accessing pages at all

• Template code diluting the authority signals on key pages

• Missing schema and hidden information that only loaded after user interaction

• Structural bugs that made pages harder to parse

The result? Competitors with simpler, more crawlable sites were being cited more frequently, even when the content quality wasn't dramatically different. Visibility wasn't a content problem. It was an access problem.

SXO, making your site physically readable by both humans and bots, is the layer everything else depends on. I'd argue it's the most urgent thing to fix before investing heavily in content strategy.

There's Still a Lot I'm Figuring Out

The landscape is changing fast enough that some of what I've written here might need an update in three months. Rankings are still volatile. Citation patterns shift. New tools are launching constantly.

What I do know is that the fundamentals of good content (clear, specific, authoritative, well-structured) still hold. The distribution and discovery layer has changed. The quality bar for the content itself has actually gone up, not down.

There's more to come on all of this.

Want to Work Together?

If you're dealing with the same shifts I've been navigating, whether that's unexplained traffic behavior, declining CTR despite growing impressions, or trying to get your site cited in AI Overviews, I'd genuinely like to help.

I'm Sofia Isabel Martillano, a freelance SEO, GEO & AEO content strategist based in the Philippines and available for remote work. You can reach me at sofiaisabelmartillano@gmail.com or connect with me on LinkedIn. My portfolio is at sofiaisabelmartillano.com if you want to see the work first.

I'm open to content writing and on-page SEO projects. Let's figure this out together.