AEO Toolkit
See your page the way search and AI answer engines do. Check which bots can crawl it, which sub-questions behind your keyword it actually answers, which passages are quotable, and where competitors out-write you.
How it works
- Crawler access. Reads robots.txt and checks 13 search, user-fetch and training bots, plus noindex/nosnippet directives.
- Query fan-out. Expands your keyword into the sub-questions answer engines typically ask: definition, pricing, features, alternatives, and so on.
- Passage coverage. Splits the page by heading into passages and scores how well each answers each sub-question.
- Extractability. Flags passages that open with “It…”, bury the answer, or contain no concrete facts, since those are hard for an engine to quote.
- Competitor gaps. Runs the same analysis on up to three competitor pages and shows where their passage wins.
Frequently asked questions
What does the AEO Toolkit actually measure?
It fetches your page the way a plain crawler would, splits it into passages, expands your target query into the sub-questions an answer engine typically asks, and checks which passages answer them. It also reports which search, user-fetch and training bots your robots.txt allows, and scores each passage for whether it can be quoted on its own.
Does it tell me what ChatGPT, Perplexity or Google will cite?
No. It is a directional simulation of retrieval, based on keyword coverage, not a prediction of any specific engine. It is most useful for before/after checks: change a passage, re-run, and see whether coverage and citation-readiness improved.
Why do some pages say my fetcher is blocked?
The web version respects robots.txt, so if a site disallows our crawler (MarketerTools-AEOToolkit) for that page, we do not fetch it. The free desktop app runs on your own machine and can analyze pages you control without that limit.
Is my data stored?
We do not store the URLs you analyze or the results. Short-lived per-connection request counters are kept only to rate limit abuse and keep the tool free for everyone.
What does the desktop app add?
JavaScript rendering with a plain-vs-rendered diff, semantic (embedding) scoring, an LLM-based query fan-out and passage judge, and a citation probe, all running locally with Ollama and no API keys.
Related tools