// Answer

Which metrics should I track for AEO?

The core AEO metrics are citation rate, share of model, and sentiment per engine, backed by crawler access and the pages engines actually cite. Here is what each one tells you and why to track it.

The short answer

For AEO, track three outcome metrics across every engine: citation rate (how often ChatGPT, Claude, Gemini, Perplexity, and Google AI Overviews mention or cite you), share of model (your slice of those answers versus competitors), and sentiment (whether the mention is positive, neutral, or negative). Keep them per engine rather than blended, since results differ by engine. Behind those outcomes, track the inputs that drive them: AI-crawler access across the major agents, llms.txt, sitemap, JSON-LD, and which specific pages each engine cites.

The three outcome numbers

Citation rate is your raw presence: out of the prompts you run, how often does an engine mention or cite you. Share of model is the competitive view: inside the answers where your category comes up, how much of the citation space is yours versus rivals. Sentiment captures how you are described, because being cited negatively is a different problem from not being cited at all. Keep each one split by engine so a strong result in one place does not paper over a blind spot in another.

The input signals behind them

Outcome numbers tell you where you stand, not what to change. The signals that move them are more concrete: can AI crawlers reach your pages, do you publish an llms.txt and a clean sitemap, is your JSON-LD valid, and which exact pages is each engine pulling from. When citation rate drops, these are what you inspect first.

Why this is important

Tracking the right set turns AEO from guesswork into a loop: measure the outcome, read the input signals, make a change, and re-measure. AI citation tracking reports the outcome numbers per engine, and competitor AI visibility frames your share of model against rivals. To get a baseline now, run the free AI visibility scan.

Frequently asked questions

They answer different questions. Citation rate tells you how often you appear at all, which is the raw visibility number. Share of model tells you how you stack up against competitors inside the same answers, which is the competitive number. Track both: a high citation rate with low share of model means you appear but competitors appear more, and that gap is where the work is.

Should I blend metrics into one AEO score?

A single blended score is useful for a quick read, but it hides where you are weak. Engines retrieve and cite differently, so you can be strong in Perplexity and invisible in Google AI Overviews. Keep the per-engine breakdown alongside any rolled-up score so you know which engine and which pages to work on.

AI Citation Tracking AI Citation Tracking Competitor AI Visibility AI Citation Answer

Run your free scan in 60 seconds

Run a free scan to see where you stand across ChatGPT, Claude, Gemini, and Perplexity: which answers cite you, which cite competitors instead, and what to fix first.

Run a free scan