Are Ghost Citations Hiding Your Brand in AI Search?

Around 40% of AI citations leave the source brand unnamed, exposing a blind spot in how companies measure visibility in AI search. Across roughly 16 million brand appearances, new data from Writesonic found that around 40% of AI citations didn't name the source brand in the generated answer. On Perplexity, the rate reached 52%.

What Happened

Writesonic analyzed roughly 16 million brand appearances across major AI engines and found that around 40% of citations linked to a source page without naming the brand in the answer itself. Search Engine Land published the findings from Nikki Lam and Samanyou Garg, who coined the term ghost citation for this gap. Perplexity led at 52%, followed by Google AI Mode at 49%, AI Overviews at 41%, ChatGPT at 37%, Gemini at 25%, Grok at 22%, and Copilot at 19%.

The Numbers in Context

A year ago, most AI visibility dashboards treated a citation as a win equivalent to a top-three organic ranking. The new data reframes that math. If Perplexity strips your brand name from 52% of its citations and Google AI Mode from 49%, roughly half of your reported wins on those engines deliver zero brand recall to the reader. Compare that to traditional SERP features, where a featured snippet typically carries the source domain in view. The ghost citation rate is not a rounding error; it is the majority behavior on two of the fastest-growing answer surfaces.

Why This Matters for HR Tech and FinTech Marketers

You operate in categories where buyers run months-long evaluations and rarely click through on a first pass. If your research, benchmark report, or product explainer gets cited without attribution, you fund the answer and a competitor collects the recall. That distorts every downstream metric. Attribution models built on AI referral traffic will undercount influence, while brand lift studies will miss the mentions that never happened. For CMOs reporting AI visibility to the board, citation count alone is now a vanity metric. You need paired tracking: where you were cited, where you were named, and where the two split. Otherwise you are optimizing for engine ingestion, not buyer awareness.

The Starr Conspiracy's Take

This is the measurement correction the category needed. We have argued that AI visibility reporting was drifting toward the same trap that inflated impression-based display metrics a decade ago. Ghost citations prove it. The fix is not to abandon citation tracking but to pair it with mention tracking and to weight engines by their naming behavior. If you want a framework for this, our take on how to measure brand visibility in AI search walks through the split-signal model. Your team should also revisit content structure. Answers that embed the brand name in the claim itself survive summarization better than answers that rely on the byline or logo to carry attribution.

What to Watch Next

Expect Perplexity and Google to face pressure from publishers and B2B brands to raise naming rates. Several platforms are already testing mention-versus-citation splits, and broader rollout as a default view could land in 2026. The likely near-term winner: brands that rewrite flagship content so the brand name lives inside the quotable claim.

Related Questions

How do you write content that AI engines quote with your brand name attached?

Embed the brand name inside the factual claim, not in the surrounding sentence. Instead of a byline attribution, structure sentences so the subject of the finding is your company. Engines summarizing the sentence tend to preserve the subject and drop the framing.

Should we still chase AI citations if half strip the brand name?

Yes, but weight your goals by engine. Gemini, Grok, and Copilot name brands in roughly 75% to 81% of citations, so they deliver stronger recall per citation. Perplexity and Google AI Mode require you to earn the mention through claim structure, not just source authority.

How does this change AI visibility reporting for CMOs?

Split your dashboard into two metrics: citation rate and mention rate. Report the gap between them as a distinct KPI. Our view on building an AEO measurement framework covers the metrics stack your team should adopt this planning cycle.