AEO Analyzer: What It Measures and Which Type You Need
An AEO analyzer is a tool that checks whether AI engines such as ChatGPT, Perplexity and Google AI Overviews can find, understand, quote and recommend your brand or content. SEO tools track blue-link rankings. An AEO analyzer looks at answers instead. The name covers three different kinds of tool.
By Raúl Díaz, founder of Slibur
11 min read
In this article
- Brand-level analyzers ask AI assistants questions and record whether your brand appears. Page-level analyzers grade one URL. Technical checkers test whether AI crawlers can reach your site.
- Scores from different tools are not comparable, so pick the type before you pick the product.
- A sensible first workflow is a free crawler check, then one page audit, then tracking of real prompts.
- In our run on 2026-10-29, ChatGPT cited 22 sites for four questions, and only 1 of them (hubspot.com) is also in Google's top 10 for this search.
- Re-check on a schedule, because AI answers change from run to run.
We make Slibur, which includes a free AEO checker and an AEO tracker, so weigh our view accordingly. We say what it does not do where it matters.
What an AEO analyzer is (and why the name covers three different tools)
An AEO analyzer evaluates how well AI-powered engines understand, quote and recommend your content, rather than where a page ranks in a list of links. In practice, the label is stuck on three different products.
Pixelmojo, in its AEO score guide (July 2026), says products marketed as AEO checkers measure at least three different things: how AI models perceive your brand, whether a specific page is structured for citation, or whether AI crawlers can reach your site at all. We use that split here because it is the most useful decision a buyer can make first.
| Class | What it grades | Input | What it cannot tell you |
|---|---|---|---|
| Brand-level | Whether AI answers name or cite you | Prompts, a brand name | Why a page is or is not quoted |
| Page-level | Whether one page is built to be quoted | One URL | Whether engines actually mention you |
| Technical | Whether AI crawlers can fetch your site | A domain | Whether the content is any good |
The three scores do not add up. A site can pass the crawler check, score well on a page audit and still never be named in an answer. The reverse happens too.
Brand-level analyzers: do AI answers name you?
These tools send real questions to AI assistants and log the answers. They report how often your brand shows up, in what position, with what tone, and which sources the AI used.
They answer the question owners care about most: are we being recommended? They do not fix anything by themselves. HubSpot's AEO tool, Profound, Otterly, Peec, AthenaHQ, Ahrefs Brand Radar and Semrush most likely sit in this group, with different engines and prompt limits.
Page-level analyzers: can an engine quote this page?
A page-level analyzer takes one URL and checks things such as heading structure, a direct answer near the top, schema and readable text. It tells you what to edit on that page.
It cannot tell you whether any engine has ever quoted the page. Treat it as a checklist, not a verdict. Our free AEO checker is this type: paste a page URL and it runs 17 checks and returns a score.
Technical checkers: can AI crawlers reach the site?
Technical checkers test robots.txt rules, sitemaps, llms.txt and whether named AI bots are allowed in. ChatGPT cited openai.com when we asked how to check AI crawler access, so start there for which bots to allow or block.
This is the cheapest check and the one to do first. If a bot is blocked, nothing else you do on the page matters for that engine.
What these tools actually measure
Most analyzers measure two families of things: what AI answers do with your brand, and how ready your content is to be used. The first is observed behavior. The second is a set of readiness proxies.
These come from brand-level tools and are computed from the answers themselves.
- Visibility: the share of answers that name or cite your brand across a set of prompts.
- Mentions: a direct mention names you. An indirect one cites your site without naming you.
- Position: where your brand sits when an answer lists several.
- Sentiment: whether the tone about you is positive, neutral or negative.
- Cited sources: which URLs and domains the AI used, and whether they are yours, a competitor's or a third party's.
- Competitor benchmarking: which rivals appear next to you or instead of you.
Meltwater's overview of AEO tools makes a similar point: effective tools track how brands appear in AI answers to reveal sentiment, prompt performance and citation trends.
The cited sources list is often the most useful output. It shows which third-party pages the engines trust on your topic, and that is where you may want to be mentioned.
Page and technical tools look at structure: clear headings, short paragraphs, a direct definition in the opening lines, and structured data. Google's introduction to structured data (opens in a new tab) explains what markup is and how it is read.
Treat schema and llms.txt as readiness proxies. They are cheap to add and some engines may use them, but no source we have shows they cause citations. The llms.txt proposal (opens in a new tab) describes the format and nothing more.
Speakable markup is a good example of a nuance. Per Pixelmojo (checked July 2026), Google documents speakable as a beta feature scoped to Assistant news playback for US English. A tool that rewards it heavily is rewarding something narrow.
How to run your first analysis in four steps
Start with the cheapest check and move toward the most informative one. Each step answers a different question, so do them in this order.
Step 1: check AI crawler access, robots.txt, sitemap and llms.txt
Open your robots.txt and look for lines like this:
User-agent: GPTBot
Disallow: /That blocks the GPTBot crawler from the whole site. You may want that, or you may have inherited it from a template. Know which.
Then run a tool. Our free AI crawler checker tests whether GPTBot, ClaudeBot, PerplexityBot and others can read the site. The llms.txt checker tests whether that file exists and follows the format, and the llms.txt generator builds a draft from your sitemap if you have none. AEO Engine also offers a free suite with robots.txt validation, sitemap auditing and llms.txt generation, per its tools page.
One caution: a crawler check reads your rules. A firewall or CDN can still block requests that robots.txt allows.
Step 2: audit one important page
Pick the page you most want quoted, such as a pricing explainer or your main product page. Run it through a page-level tool.
Read the failed checks, not just the score. Typical fixes are a direct answer near the top, a question-style heading, and removing text hidden behind scripts. Our free 18-point GEO audit covers the homepage, robots.txt, sitemap and llms.txt, but it does not crawl the whole site.
A before and after makes the point. Before: an opening paragraph about "our mission to transform how teams work." After: "Project X is a task manager for teams of 5 to 50 that costs $12 per user per month." The second can be quoted as is.
Step 3: track real prompts across engines
Write 10 to 20 questions your customers actually ask. Do not put your brand name in them, because buyers who do not know you will not either.
Run them across engines and record whether you appear, in what position, and which sources are cited. You can do this by hand in a spreadsheet, or use a tracker. Slibur's AEO tracker does this on five AIs (ChatGPT, Google AI Overviews, Claude, Perplexity and Grok) and sends a weekly summary.
By hand, log the date, engine, prompt, brands named and sources. That is enough to see a pattern after a few weeks.
Step 4: fix, re-run and keep the change
Change one thing at a time where you can: a page rewrite, a robots.txt rule, a mention on a source the engines already cite. Then re-run the same prompts.
Keep a changelog with dates. Without it, you cannot tell whether a move in visibility came from your edit or from the engine changing its answer.
How to read an AEO score without fooling yourself
An AEO score is one tool's estimate, and it only means something compared with the same tool's earlier scores. Pixelmojo defines it as a number, usually 0 to 100, that estimates how ready your content is to be found, understood and quoted by AI answer engines.
There is no shared scale. Each vendor weights its own categories. Saram Technologies, for example, says its scores run 0 to 100 across five weighted categories.
For context only: Pixelmojo reports that across 59 domains that ran its audit between April 26 and June 3, 2026, the average score was 26 out of 100, and about 9 in 10 scored below 40. That is one vendor's sample of people who chose to run an audit. It does not describe the web.
In that one vendor's sample, low scores were common, so do not panic over yours. Track your own trend.
Ranking first does not guarantee citation, and other assistants use their own search and sources.
Our own run below points the same way. ChatGPT cited 22 sites across four answers, and one of them appeared in the Google top 10 for this search.
A few prompts give a noisy picture. Saram's free audit runs a real scan with 3 AI prompts and returns a partial report, per its guide. Three questions can tell you whether the pipe works. They cannot tell you your visibility.
AI answers also vary between runs, so one answer is one draw. Ask a vendor how many prompts it runs, how often, in which country and language, and whether prompts name your brand. A tool that includes your brand in the prompt may inflate your visibility.
Free vs paid: what you get at each level
Free analyzers check readiness and give a single snapshot. Paid ones track answers over time across many prompts. Pick by the question you need answered.
What free scanners can answer
Free tools answer yes-or-no questions well: can a crawler get in, does llms.txt exist, does this page have a direct answer, is schema present. AEO Engine's free hub and Slibur's free tools sit here.
They are weaker on brand visibility, because a handful of prompts is a thin sample.
When tracking over time is worth paying for
Pay when you need trends, competitor comparison or alerts for a topic where AI answers matter to sales. HubSpot's AEO tool is described on its page at about $50 per month, with tracking across ChatGPT, Gemini and Perplexity (page checked October 2026). Profound, Otterly, Peec, AthenaHQ, Ahrefs and Semrush offer paid tracking too, with their own prices and limits on their sites.
Slibur is a paid tracker with a 7-day free trial (no card, up to 10 prompts). Paid plans per month before taxes: Starter $59 (20 prompts), Growth $129 (75 prompts), Scale $299 (250 prompts), with 15% off annual billing. ChatGPT and Google AI Overviews run daily, and Claude, Perplexity and Grok weekly. Limits you should know: no Gemini app, Google AI Mode or Copilot, no CSV or PDF export, no API, and alerts only in the weekly email. See pricing for details.
Before buying, ask these questions:
- Which engines are covered, and which are not?
- How many prompts, and can you add your own?
- How often does it run, and is that daily or weekly per engine?
- Which country and language does it measure?
- Do prompts name your brand?
- Can you see the full answer and the sources, or only a score?
Test methods deserve the same scrutiny. AI Rank Lab says it spent eight weeks testing 10 platforms on the same 25 prompts and compared results with manual checks, with 30% of its score for tracking accuracy. That is a clear method, but it is a vendor-run test, so check it yourself.
Slibur dataChatGPT, United States, 4 answers, October 29, 2026
We asked ChatGPT about AEO analyzers: what came back
We asked ChatGPT (chatgpt.com, logged out, with web search, from the United States) four questions on 2026-10-29, and all four were answered. Across the four answers, it named 23 distinct brands and cited 22 distinct sites. This is a snapshot, not a study, and it supports no general rule.
Which sites ChatGPT cites versus Google's top 10
Only 1 of the 22 sites ChatGPT cited also ranks in Google's top 10 for "aeo analyzer": hubspot.com. Seven of the eight top-10 domains were not cited: youtube.com, pixelmojo.io, meltwater.com, airanklab.com, saramstechnologies.com, aeoengine.ai and reddit.com.
For the question "What is an AEO analyzer?" it cited hubspot.com, techradar.com, elementera.com, semrush.com and ahrefs.com. For the crawler question it leaned on cloudflare.com, synthesisinsights.com, sitequiry.com, checkseo.dev and openai.com.
The most cited sites were ahrefs.com, hubspot.com and semrush.com, with two citations each. Everything else appeared once.
Which tools the assistants name
For "What is an AEO analyzer?" ChatGPT named HubSpot AEO, Semrush and Ahrefs Brand Radar, in that order. For the brand-mention question it named Otterly.AI, Peec AI, Profound, Semrush and Ahrefs Brand Radar.
On page audits it listed AI Rank Inspector, AEO.wiki, Ten Speed, HubSpot, iTechGuides, AuditLamp, SEO Scrapper and AEO.co. For crawlers it named bots (GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, PerplexityBot, Google-Extended) plus Sitequiry, CheckSEO and Cloudflare.
Semrush and Ahrefs Brand Radar each came up in two answers. The page-audit answer separated checklists for learning how to audit a page from tools that scan a URL.
Which analyzer fits your situation
Match the tool class to the decision you need to make, not to the biggest feature list.
One site and no budget
Use free tools only. Run the crawler check, then the llms.txt check, then audit your top page with the free AEO checker. Ask five or ten customer questions by hand in ChatGPT and Perplexity and write down the results.
This costs nothing and covers the three classes at a basic level.
A marketing team tracking competitors
You need a brand-level tracker with enough prompts and a stable cadence. Compare engines, prompt limits, country and language, and source transparency using the checklist above.
If five engines on a weekly rhythm fit, Slibur's AEO tracker is one option, and the 7-day trial (no card) lets you test it. It does not measure traffic or conversions from AI, offers no API and no exports, and tracks one country and one language per workspace. Rule it out if you need those.
Content team fixing pages
Start with page-level audits and the sources list from a tracker. Fix the pages that fail basic checks, then look at which third-party pages the engines cite for your topic.
Re-run your prompts every week or two so you can link edits to results. If you want to try tracking, start the free trial.
Frequently asked questions
What is an AEO analyzer?
An AEO analyzer is a tool that checks whether AI engines like ChatGPT, Perplexity and Google AI Overviews can find, understand, quote and recommend your brand or pages. Depending on the product, it tracks brand mentions in answers, grades one page, or tests whether AI crawlers can reach your site.
What is the difference between an AEO analyzer, checker and grader?
The names are used loosely, and no standard separates them. What matters is what the tool grades: your brand in AI answers, one page's structure, or crawler access. Read the product page for its input and method, because scores from different tools are not comparable.
Is there a free AEO analyzer that is worth using?
Yes, for readiness checks. Free tools can tell you whether AI crawlers are blocked, whether llms.txt exists and whether a page has a direct answer. They are thin on brand visibility, since some free audits run only a few prompts, such as three in Saram's case.
What is a good AEO score?
There is no shared scale, so a good score depends on the tool. Pixelmojo reported an average of 26 out of 100 across 59 domains in its own audit, with about 9 in 10 below 40. Use that as context only, and compare your score with your own earlier runs.
Does ranking #1 on Google mean AI engines will cite me?
No. In our ChatGPT run on 2026-10-29, only 1 of the 22 sites it cited also ranked in Google's top 10 for "aeo analyzer". That is a snapshot of four answers, not a rule, but it shows rankings and citations can differ.
How often should I run an AEO analysis?
Run technical and page checks after each change, and track prompts on a steady schedule, weekly at minimum for most teams. AI answers vary between runs, so one result is a single draw. Keep the same prompts and a dated changelog so you can see real trends.
Raúl Díaz is the founder of Slibur, the tool that asks ChatGPT, Google AI Overviews, Claude, Perplexity and Grok the questions your customers ask and shows which brands they recommend. He writes about answer engine optimization from the data Slibur collects.
This article was drafted with AI assistance from our own data and a read of the top results at the time of writing, then checked before publishing. How we write the blog