The core AEO metrics are citation share (how often engines cite you), answer presence (whether you appear at all for target prompts), share of AI voice (your citations versus competitors), and referral conversion (revenue from AI traffic). Measure each across a fixed prompt set and your analytics, tracked over weeks.
Part of the AEO cluster. Optimizing without measuring is guessing. Once you have worked through ranking factors, these are the numbers that tell you if the work landed.
AEO measurement is not one number. Traditional rank trackers do not see inside a generated answer, so you need a small set of purpose-built KPIs. Four cover most of what you need to know.
The four metrics that matter
| Metric | Question it answers | How to measure |
|---|---|---|
| Citation share (citation rate) | How often do engines cite me? | Citations of your domain ÷ total queries tested × 100 |
| Answer presence | Do I appear at all for target prompts? | Share of tracked prompts where your brand is mentioned or cited |
| Share of AI voice | How do I stack up against competitors? | Your citations ÷ total citations for the prompt set × 100 |
| Referral conversion | Does AI traffic turn into revenue? | Conversions and revenue from AI-referral sessions in analytics |
1. Citation share
Citation share, or citation rate, is the proportion of your tracked queries where an answer engine cites your domain at least once.
Formula: Citation rate (%) = (queries where your brand appears ÷ total queries tested) × 100.
How to measure it: build a fixed set of 50 to 200 buyer-intent prompts, run them across ChatGPT, Perplexity, Google AI Mode, and Copilot on a regular cadence, and log whether you were cited. Keep the prompt set stable so trends mean something.
2. Answer presence
Presence is the coarser, upstream version of citation share: are you in the answer at all, whether cited with a link, mentioned by name, or quoted without attribution? A brand can be mentioned without being linked, and that still shapes buyer perception.
How to measure it: on the same prompt set, tag each result as cited, mentioned, or absent. Presence rate is the share tagged cited or mentioned. Watch mentions specifically, since unlinked mentions still influence buyers and often precede citations.
3. Share of AI voice
Share of AI voice measures how often you appear in AI answers compared to your competitors for the same prompts. It turns raw citation counts into competitive standing.
Formula: Share of AI voice = (your citations ÷ total citations across you and named competitors) × 100.
How to measure it: on your prompt set, log every brand cited, not just yours. Your share is your citations over the total. This is the metric to report upward, because it frames progress against the market, not against zero.
4. Referral conversion
Visibility is not the goal; pipeline is. AI referral traffic is visitors who arrive from a citation in a generated answer, and referral conversion is what those visitors do next.
How to measure it: segment AI-referral sessions in analytics (referrers like chatgpt.com, perplexity.ai, and similar), then track conversion rate, revenue, and assisted pipeline from that segment. Context: AI platforms drove 1.13 billion referral visits in June 2025, up 357% year over year, with ChatGPT accounting for about 87.4% of AI referral traffic. This channel is real and growing, so give it its own reporting line.
Putting it together
Read the metrics as a funnel:
Answer presence tells you whether you are in the conversation.
Citation share tells you whether you are trusted enough to be named.
Share of AI voice tells you how you rank against rivals.
Referral conversion tells you whether any of it pays.
Two practical rules. First, hold your prompt set constant, or your trend lines are noise. Second, measure over weeks; crawl cadence and answer variance make day-to-day numbers jumpy. If presence is high but citation share is low, revisit extractability and authority signals. If you are early and unsure where to begin, start with AEO for beginners.
FAQ
What is the most important AEO metric?
Share of AI voice for competitive reporting, and citation share for tracking your own progress. Presence is the leading indicator, and referral conversion is the business outcome.
How do I track citations inside AI answers?
Run a fixed prompt set across the major engines on a schedule and log whether you were cited. Tools automate this, but the method is the same: stable prompts, consistent cadence, logged results.
How is AI referral traffic different from organic?
It arrives from a generated answer rather than a search results page. Segment it by referrer in analytics and report it separately, because intent and conversion behavior differ from classic organic.
How often should I measure?
Weekly for prompt-based metrics, and continuously for referral traffic in analytics. Avoid reading too much into a single day; answer engines vary run to run.
What is a good citation rate?
It depends on your niche and competition, which is why share of AI voice matters more than an absolute number. Track direction over time against your own baseline and named competitors.
Can I measure AEO for free?
You can start manually by running prompts and logging results, and by segmenting AI referrers in analytics. Tools save time as your prompt set and page count grow.
Stop guessing at your AI visibility. SEO AEO Specialist tracks citation share, presence, and share of AI voice across your pages and shows what is moving. Free scan €0 (50 pages), one-time report €9, continuous monitoring €19/mo per domain.
Sources: HubSpot, Cairrot, Discovered Labs
AEO for Beginners: A Zero-to-One Guide to Getting Cited by AI | SEO AEO Specialist
See where you stand
SEO AEO Specialist runs a free AI-visibility audit and hands you the exact fixes. One-off report €9, unlimited €19/mo.
Run your free audit →