Before You Worry About AEO/GEO - Check These 5 SEO Fundamentals | SerpCtrl
2026-07-02·7min
Before You Worry About AEO/GEO - Check These 5 SEO Fundamentals
SEOAEOGEO
1. Can the crawler actually read your content?
This is the one that kills most sites quietly. Google and Bing render JavaScript - slowly, imperfectly, but they do it. The AI crawlers mostly don't. GPTBot, ClaudeBot, PerplexityBot, CCBot - they pull your raw HTML and leave. If your headings, your body copy, your structured data get injected client-side after the page loads, those bots see an empty shell.
So the test is simple: disable JavaScript in your browser and load your most important page. What's left? If it's a blank box with a spinner, you don't have an AEO problem. You have a rendering problem that's been costing you in classic search too - you just never measured it.
Server-side render the content that matters. This isn't a GEO tactic. It's table stakes, and it's been table stakes for years.
2. Is your site even indexable?
Before "will an AI cite me," answer "can anything index me at all." Open your robots.txt. Open your sitemap. Check Search Console coverage. The boring failures are the common ones: a Disallow: / left over from a staging environment, a sitemap full of URLs that redirect three times before 404ing, canonical tags pointing at the wrong version of a page, whole sections sitting behind parameters no crawler bothers with.
None of that is exotic. It's the stuff that gets skipped because it isn't fun, and then quietly caps everything else you do. An engine - Google's or an LLM's - can't choose you from a set it never crawled. You cannot be the answer to a question if you're not in the room.
So before you budget a cent for "AI visibility," confirm the machines can actually see the pages you care about. If they can't, that's not a GEO problem. It's an indexing problem, and it's been costing you all along.
3. Does your content actually answer a question?
Here's where AEO advocates are half-right. AI engines do favor content that directly answers a specific question, in plain language, near the top of the page. But - and this matters - so does Google's featured snippet, and has done since 2014. Same instinct, older name.
If your blog post buries the answer under 600 words of "in today's digital landscape" preamble, no engine is pulling a clean citation from it. Lead with the answer. Use a clear question as your heading and answer it in the next two sentences. Then expand. This is good writing and good SEO at the same time - the AI angle is just the newest reason to finally do it.
The mistake is treating this as an AI-specific rewrite. Write for the human who's scanning. The machines are trained on what works for humans.
And here's the one-up, because most people skip it: don't guess what your content is about - go check what the machine thinks it's about. Spend a few euros and run your pages through Google's Natural Language API. It'll hand you back the entities and categories it actually extracts, which is frequently not the topic you thought you were writing on. That gap - between what you meant and what the machine reads - is exactly the gap that loses you the citation.
You can do it at scale, too.
4. Do you have the authority to be trusted?
This is the part the hype sellers skip, because it's slow and unglamorous. Building genuine authority means earning links, publishing consistently in your niche, and being mentioned by other credible sources. The exact same off-page work that's defined SEO for two decades. If anything, AI search raises the stakes - these engines lean hard on established, frequently-cited sources, which means the rich get richer and newcomers have to work harder, not less.
If your domain has no authority, "optimizing for Perplexity" is rearranging deck chairs.
5. Is the technical foundation healthy?
A slow, broken, badly-structured site won't get cited by an AI engine for the same reason it won't rank: the crawler gives up, or the ranking systems discount it, or it never gets crawled deeply enough to matter. Fix that foundation before anything else - schema on top of a broken site is lipstick.
But let's talk about schema directly, because it's where the hype and the cynicism both get it wrong. One camp sells structured data as a magic AEO unlock. The other waves it off - "Google's walking away from FAQ rich results, schema never really mattered." Both are missing the point, and the point is almost stupidly simple: schema is just a JSON file.
That's it. JSON-LD is a small, structured, machine-readable block that states, in plain key-value terms, what your page is and how its parts relate. No prose to parse, no layout to interpret, no ambiguity to resolve. And structured, unambiguous JSON is exactly what machines want to be handed - which is why the entire AI field keeps converging on it. Look anywhere outside SEO: AI video tools are standardizing on JSON prompts precisely because JSON is clear, modular and developer-friendly enough that the model doesn't have to guess. Same instinct, different room. When you ship clean JSON-LD, you're not gaming a rich result - you're handing every engine, search or LLM, an unambiguous description of your content in the one format they all read without friction.
So whether Google retires this or that rich snippet is beside the point. The snippet was always just a reward for structured data, never the reason for it. The reason is that you're describing your page to a machine in machine language. That doesn't stop mattering because one SERP feature gets deprecated - if anything, as more reading shifts to LLMs that ingest rather than render, handing them clean structured data matters more, not less.
Do it because it's cheap, it's unambiguous, and it speaks the language the machines were built to read. Not because someone on LinkedIn told you it's the new growth hack.
So what about AEO and GEO, actually?
I'm not telling you to ignore AI search. It's real, and being cited in an AI answer is becoming genuinely valuable. But notice that even this deserves a number, not a vibe. One analysis making the rounds on LinkedIn tracked AI-assistant referral traffic over twelve months: it peaked near 4M visits and then fell to around 2.8M - roughly a 30% decline - over the same stretch everyone was calling AI adoption "an all-time high." Growing technology, volatile and still-small traffic. As that post put it bluntly: "if your SEO foundations are weak, your AI visibility will probably be weak too." That's the whole argument, from someone who actually showed the chart.
So when the next viral post tells you "SEO is so easy now, just let the AI do it," remember what the honest version of that list actually looks like: qualify keywords, fix technical and UX issues, build pages for users and agents, earn links and brand mentions, maintain entity consistency, test what works in the field instead of in theory, and keep up as search, AI and agents shift weekly. That's not a dead discipline. That's a bigger one - and no acronym makes the work disappear.
What I'm telling you is that there is no separate AEO/GEO project that lives apart from your SEO. The checklist above is the AEO checklist. Crawlable content, indexability, question-first writing, authority, technical health - nail those, and you are optimized for AI search, because AI search is built on the same signals.
The agencies selling you a standalone "GEO package" are mostly selling you SEO fundamentals at a markup, with a more exciting name on the invoice. Some of them know it. Some of them don't, which is worse.
Do the boring work. Then, once your foundation is solid, the AI-specific refinements - tightening your answer formatting, strengthening entity clarity, making sure your structured data is complete - actually have something to stand on. Refinements on a foundation. Not a replacement for one.
Where to go from here
If you want the occasional plain-spoken take on what's real versus what's hype in search - no acronym soup - the SerpCtrl newsletter is where I write it down.
And if you'd rather just know whether your foundation is solid before you spend on AI optimization, book a consultation. Ten minutes usually tells me whether you have a GEO opportunity or an indexing problem wearing a GEO costume.
/// NEWSLETTER
Get practical ideas
for digital growth
Join our newsletter for concise, useful insights on web development, SEO, and digital performance.