Every agency offers AEO now. Almost none of them know what they’re doing
That’s not a dig. It’s a structural problem with how fast the category has moved. Twelve months ago, AEO barely existed as a named discipline. Today it’s on every agency’s services page — usually wedged between “technical SEO” and “content strategy” with a paragraph that could have been written by someone who read one blog post on the topic.
If you’re evaluating agencies for AI visibility work, this matters. A lot. Because the gap between an agency that understands AEO and one that doesn’t isn’t marginal — it’s the difference between measurable citations in 90 days and a retainer that produces nothing you can track.
Here’s how to tell the difference.
They treat AEO as a content problem
The most common mistake agencies make is treating AEO as a content-only discipline. Rewrite some pages with Q&A formatting, add a few FAQ sections, call it done.
Content restructuring is real AEO work. But it’s one layer of three. Without the technical foundation (site speed, structured data validation, entity alignment, crawlability for AI platforms) and the strategic layer (which prompts to target, which platforms to prioritize, how to build topical authority in the right areas), content rewrites produce limited results.
Agencies that lead with “we’ll optimize your content for AI” without discussing schema implementation, entity signals, or citation monitoring are selling the middle of the process without the beginning or the end.
They can’t show you citations
Proof is the fastest filter.
Ask any agency you’re evaluating to show you examples of clients being cited in AI-generated answers. Not traffic reports. Not ranking improvements. Actual citations — the prompt that triggered them, the platform they appeared on, and the date.
Most agencies can’t do this. Not because their clients aren’t being cited, but because they haven’t built the measurement infrastructure to track it. If an agency can’t show you citations, they either don’t have them or don’t know how to find them. Neither is acceptable.
This matters because AI visibility without measurement is indistinguishable from no AI visibility at all. You need to know which prompts trigger citations, which platforms mention you, and whether that number is growing month over month.
They optimize for Google’s logic, not AI’s
SEO agencies are trained to think in terms of Google’s ranking factors — backlinks, domain authority, keyword density, page speed scores. These things matter for organic search. They matter much less for AI citation.
AI platforms don’t rank pages. They retrieve, extract, and synthesize content from sources they’ve determined to be trustworthy and relevant. The signals they use to make those determinations are different from Google’s — and in some cases directly opposite.
A page that ranks #1 on Google because of its backlink profile might be completely ignored by ChatGPT if the content isn’t structured for extraction. A page with modest domain authority but clear entity signals, answer-first formatting, and comprehensive schema markup might get cited consistently across multiple AI platforms.
Agencies that apply Google’s playbook to AI visibility work are optimizing for the wrong algorithm. The strategies that drive AI citations require different thinking, different technical implementation, and different measurement systems.
They don’t track share of voice
Rankings are a familiar metric for SEO agencies. Citation rate is a newer, more nuanced one. Share of voice — how often you appear relative to competitors across a defined set of target prompts — is the metric that most agencies haven’t built the infrastructure to track at all.
Share of voice matters because AI visibility is inherently competitive. You’re not just trying to appear in AI answers — you’re trying to appear more often than the three or four competitors AI currently favors. Without tracking competitor citations alongside your own, you have no way of knowing whether you’re gaining ground or losing it.
The agencies that truly understand AEO treat share of voice as a primary KPI, not an afterthought. They track which competitors appear for which prompts, reverse-engineer why, and build strategies to displace them systematically.
They have no clear process for AI platforms
Real AEO work requires platform-specific knowledge. ChatGPT retrieves differently from Perplexity. Google AI Overviews pulls from indexed content in ways Claude doesn’t. Gemini has its own entity evaluation signals.
Agencies that understand AEO have a documented process for each platform — how they audit visibility, what they optimize for, and how they measure results. Agencies that don’t tend to describe their process in platform-agnostic terms that reveal they haven’t tested across all five major platforms.
Ask any agency you’re evaluating: how does your approach differ for ChatGPT versus Perplexity? If the answer is vague or identical for both, that tells you everything you need to know.
What a real AEO practice looks like
For the record, here’s what separates agencies that genuinely understand AEO from those that don’t.
They start with an audit — mapping current citation rates across platforms before recommending any work. They optimize across three layers (technical, content, and strategy) rather than treating AEO as a content-only exercise. They track citations in real time and can show you exactly which prompts trigger mentions. They measure share of voice against competitors, not just absolute citation frequency. And they have platform-specific knowledge that informs how they approach ChatGPT differently from Perplexity or Claude.
Most importantly, they can show you results. Not promised results. Not theoretical frameworks. Actual clients being cited in actual AI answers with actual measurement behind it.
The bottom line
AEO is a young discipline moving faster than most agencies can keep up with. That’s not a reason to avoid it — it’s a reason to be selective about who you trust to do it.
The agencies that understand it have built their practice around AI visibility from the ground up. The ones that don’t added a services page and hoped no one would ask hard questions.
Ask the hard questions.
Related reading: how AI decides who to recommend · how to audit your AI visibility · talk to VisAbility.
Every agency offers AEO now. Almost none of them know what they’re doing
That’s not a dig. It’s a structural problem with how fast the category has moved. Twelve months ago, AEO barely existed as a named discipline. Today it’s on every agency’s services page — usually wedged between “technical SEO” and “content strategy” with a paragraph that could have been written by someone who read one blog post on the topic.
If you’re evaluating agencies for AI visibility work, this matters. A lot. Because the gap between an agency that understands AEO and one that doesn’t isn’t marginal — it’s the difference between measurable citations in 90 days and a retainer that produces nothing you can track.
Here’s how to tell the difference.
They treat AEO as a content problem
The most common mistake agencies make is treating AEO as a content-only discipline. Rewrite some pages with Q&A formatting, add a few FAQ sections, call it done.
Content restructuring is real AEO work. But it’s one layer of three. Without the technical foundation (site speed, structured data validation, entity alignment, crawlability for AI platforms) and the strategic layer (which prompts to target, which platforms to prioritize, how to build topical authority in the right areas), content rewrites produce limited results.
Agencies that lead with “we’ll optimize your content for AI” without discussing schema implementation, entity signals, or citation monitoring are selling the middle of the process without the beginning or the end.
They can’t show you citations
Proof is the fastest filter.
Ask any agency you’re evaluating to show you examples of clients being cited in AI-generated answers. Not traffic reports. Not ranking improvements. Actual citations — the prompt that triggered them, the platform they appeared on, and the date.
Most agencies can’t do this. Not because their clients aren’t being cited, but because they haven’t built the measurement infrastructure to track it. If an agency can’t show you citations, they either don’t have them or don’t know how to find them. Neither is acceptable.
This matters because AI visibility without measurement is indistinguishable from no AI visibility at all. You need to know which prompts trigger citations, which platforms mention you, and whether that number is growing month over month.
They optimize for Google’s logic, not AI’s
SEO agencies are trained to think in terms of Google’s ranking factors — backlinks, domain authority, keyword density, page speed scores. These things matter for organic search. They matter much less for AI citation.
AI platforms don’t rank pages. They retrieve, extract, and synthesize content from sources they’ve determined to be trustworthy and relevant. The signals they use to make those determinations are different from Google’s — and in some cases directly opposite.
A page that ranks #1 on Google because of its backlink profile might be completely ignored by ChatGPT if the content isn’t structured for extraction. A page with modest domain authority but clear entity signals, answer-first formatting, and comprehensive schema markup might get cited consistently across multiple AI platforms.
Agencies that apply Google’s playbook to AI visibility work are optimizing for the wrong algorithm. The strategies that drive AI citations require different thinking, different technical implementation, and different measurement systems.
They don’t track share of voice
Rankings are a familiar metric for SEO agencies. Citation rate is a newer, more nuanced one. Share of voice — how often you appear relative to competitors across a defined set of target prompts — is the metric that most agencies haven’t built the infrastructure to track at all.
Share of voice matters because AI visibility is inherently competitive. You’re not just trying to appear in AI answers — you’re trying to appear more often than the three or four competitors AI currently favors. Without tracking competitor citations alongside your own, you have no way of knowing whether you’re gaining ground or losing it.
The agencies that truly understand AEO treat share of voice as a primary KPI, not an afterthought. They track which competitors appear for which prompts, reverse-engineer why, and build strategies to displace them systematically.
They have no clear process for AI platforms
Real AEO work requires platform-specific knowledge. ChatGPT retrieves differently from Perplexity. Google AI Overviews pulls from indexed content in ways Claude doesn’t. Gemini has its own entity evaluation signals.
Agencies that understand AEO have a documented process for each platform — how they audit visibility, what they optimize for, and how they measure results. Agencies that don’t tend to describe their process in platform-agnostic terms that reveal they haven’t tested across all five major platforms.
Ask any agency you’re evaluating: how does your approach differ for ChatGPT versus Perplexity? If the answer is vague or identical for both, that tells you everything you need to know.
What a real AEO practice looks like
For the record, here’s what separates agencies that genuinely understand AEO from those that don’t.
They start with an audit — mapping current citation rates across platforms before recommending any work. They optimize across three layers (technical, content, and strategy) rather than treating AEO as a content-only exercise. They track citations in real time and can show you exactly which prompts trigger mentions. They measure share of voice against competitors, not just absolute citation frequency. And they have platform-specific knowledge that informs how they approach ChatGPT differently from Perplexity or Claude.
Most importantly, they can show you results. Not promised results. Not theoretical frameworks. Actual clients being cited in actual AI answers with actual measurement behind it.
The bottom line
AEO is a young discipline moving faster than most agencies can keep up with. That’s not a reason to avoid it — it’s a reason to be selective about who you trust to do it.
The agencies that understand it have built their practice around AI visibility from the ground up. The ones that don’t added a services page and hoped no one would ask hard questions.
Ask the hard questions.
Related reading: how AI decides who to recommend · how to audit your AI visibility · talk to VisAbility.
