AI SEO Glossary

What is Citation Share?

TL;DR — Citation share is the percentage of category-relevant prompts on which a given AI engine cites your brand by name or links to one of your URLs. Tracked weekly across a defined prompt panel, per engine. The closest thing AEO has to a clean, comparable KPI.

Definition & scope

Citation shareis the metric AEO measurement is built around. Definition: across a defined, category-relevant prompt panel (typically 100-500 prompts run weekly), the percentage of answers in which a given engine cites the brand by name or links to one of the brand's URLs. Measured separately per engine (ChatGPT citation share, Perplexity citation share, Claude citation share, etc.) and aggregated to a cross-engine view.

Citation share is the closest equivalent to “average rank position” in classic SEO — a single number that moves measurably as the work moves, and that ties to a defined prompt set the engagement is being optimized against. Measured via prompt-panel monitoring: same prompts, same engines, same week-over-week diff. Without citation-share measurement, an AEO program is flying blind.

Where you'll encounter it

In any serious AEO/GEO retainer dashboard — Workspace surfaces citation share per engine, per prompt category, per week. In vendor product copy: most AI-SEO tooling (Profound, Otterly, Goodie, Peec) markets some version of citation-share tracking. In AEO case studies as the headline metric that moved.

For an executive sponsor, citation share is the AI-SEO metric to report monthly. For the operator, it's the leading indicator — citations move first, AI-referral traffic follows, organic referrals from classic search lift later as engines reinforce known-brand authority.

Related terms

AEO — citation share is the AEO KPI.
GEO — the umbrella under which citation tracking sits.
PerplexityBot — Perplexity is the easiest engine to measure citation share on.
RAG — the model architecture that produces citations.

Related services

AEO Services — citation work measured weekly by citation share.
Citation Share Monitoring Methodology — the panel and reporting design.

FAQ

How is citation share measured?

Run a defined panel of category-relevant prompts (typically 100-500) weekly across each engine. Record whether each answer cites your brand by name or links to one of your URLs. Divide by total prompts. Report per engine and aggregated. Full methodology on /methodology/citation-share-monitoring.

What's a good citation share?

Depends on category and engine. For a strong category leader in a focused vertical, 15-30% citation share on Perplexity is common; 5-15% on ChatGPT. For challenger brands, anything above 5% is a real win. The diff over time matters more than the absolute number.

Per engine or aggregated?

Both. Per engine is operationally useful — it tells you whether GPTBot/ClaudeBot access work moved citations on the corresponding engine. Aggregated is the headline for execs.

Why not just measure traffic?

AI-engine referral traffic is real (Perplexity, ChatGPT, Claude all send measurable click-through), but the citation surface is bigger than the traffic — many AI answers include the brand name without the click. Citation share captures the full mindshare lift; traffic captures only the click-through subset.

What prompt panel should I use?

Mix of commercial-intent (“best [category] for [use case]”), comparative (“[your brand] vs [competitor]”), and informational (“how to [task] in [category]”). 100-300 prompts is enough signal for weekly tracking on most categories.

How often should I re-measure?

Weekly is the standard cadence for an active engagement — frequent enough to catch shifts inside a sprint, not so frequent that noise dominates. Some teams run daily during AEO sprints; quarterly is too infrequent to course-correct.

Stop guessing whether AI engines cite your brand. Measure it.

We run weekly prompt-panel monitoring across 10+ engines via Workspace. 888-982-8269.

Get a Citation-Share Benchmark Call 888-982-8269