Does optimising for Perplexity citations improve AI search visibility?

Yes: high confidence, with four platform-official sources confirming the core mechanisms.

Perplexity operates as a live-web retrieval system. PerplexityBot actively crawls and indexes content for Perplexity search results. The platform officially confirms that allowing PerplexityBot in robots.txt is required for Perplexity visibility: sites blocking the crawler are excluded from all results regardless of content quality.

Freshness is the dominant signal

A Seer Interactive analysis of 5,000+ URLs found approximately 50% of Perplexity citations come from 2025 content alone; 80% from 2023–2025. An arXiv study found median cited content age is 148 days for AI search versus 493 days for Google: AI search is 2–3× fresher across platforms, with Perplexity showing the most extreme freshness preference.

For content strategy: consistent publishing cadence creates a direct lever on Perplexity citation rates. Content older than 12–18 months is structurally disadvantaged regardless of quality.

Quality thresholds and the GEO-16 framework

Content achieving a GEO-16 quality score of 0.70 or higher with at least 12 of 16 on-page quality pillars satisfied achieves a 78% cross-engine citation rate across Brave, Google AIO, and Perplexity. Pages below that threshold are cited inconsistently regardless of content quality or freshness.

The ghost citation problem

A Growth Memo study of citation patterns found 61.7% of Perplexity citations are ghost citations: the domain gets a source link but zero brand name recognition in the answer text. The brand is technically cited but users don't see its name. Getting from ghost citation to visible citation requires content that directly influences the answer, not just content that appears in the retrieval context window.

Perplexity is platform-specific

Only 11% of cited domains appear on more than one AI platform. Optimising for Perplexity does not transfer to ChatGPT or Google AIO. Perplexity citation patterns skew heavily toward Reddit (6.6% of results vs 1.8% for ChatGPT) and toward B2B review platforms (Reddit, LinkedIn, G2) for professional queries.

Performance benchmarks

A citation rate of 15–20% across relevant queries represents solid performance for new AEO programs. 30–40% represents excellent performance across ChatGPT, Perplexity, Google AIO, and Copilot (Leading Minds). Use these as reference points for evaluating Perplexity-specific citation programs.

What the evidence doesn't prove

The freshness correlation doesn't establish cause directly: recently published content is also more likely to cover fast-moving topics, which receive freshness weighting for topic reasons rather than publication date alone.

The 78% cross-engine citation rate for GEO-16 compliant pages is practitioner research. It cannot separate whether quality signals caused the citations or whether consistently cited pages are also the pages most likely to implement those signals.