Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.getcited.in/llms.txt

Use this file to discover all available pages before exploring further.

Perplexity is the only major AI platform that provides explicit, numbered citations on every response. When a user asks a question, Perplexity retrieves web sources in real time, synthesizes an answer, and displays each cited source with a clickable link. This makes Perplexity the cleanest platform for measuring citation rate — and the most important one to optimize for if citation-driven traffic is your goal.

How Perplexity generates answers

Perplexity is a retrieval-first platform. Every response starts with a live web search, then synthesizes the retrieved content into an answer with explicit source attribution. Unlike ChatGPT (which blends training data with optional search), Perplexity’s answers are always grounded in real-time retrieval. This makes its behavior more deterministic than training-data-heavy models — the same query will tend to surface similar sources across runs, though the synthesis still varies. Perplexity offers multiple modes: standard Sonar (faster, cheaper) and Sonar Pro (more sophisticated reasoning). The Cited Index uses standard Sonar for monitoring, which provides an 83% cost reduction compared to Sonar Pro with comparable accuracy for mention rate measurement.

How sources are selected and ranked

Perplexity’s citation list is ordered — the first-cited source typically receives the most weight in the synthesized answer. Five factors appear to drive source selection based on observed behavior.
FactorObserved weightDescription
RecencyHighRecently updated pages are strongly favored over stale content
Topical authorityHighPages that comprehensively address the specific query outperform general pages
Editorial qualityHighWell-structured content from established publications ranks above user-generated content
Technical accessibilityMediumClean HTML, no bot-blocking, fast page loads — sites that block crawlers are invisible
Domain authorityMediumEstablished domains with a publication history tend to be cited more frequently
Perplexity consistently favors editorial publications and structured reference sites. Brand-owned content is cited but at lower frequency than on ChatGPT — earning Perplexity citations typically requires third-party editorial coverage in the publications Perplexity retrieves.

What Cited’s data shows

Perplexity shows the strictest mention behavior of the tracked platforms in the Cited Index: p25=0.0%, p50=5.0%, p75=12.0%. A full 25% of brands receive zero mentions on Perplexity. This reflects the retrieval-first approach — if your content does not surface in real-time search for the query, you are invisible on Perplexity regardless of how well-known your brand is. The flipside is quality: brands that DO appear on Perplexity tend to get cited (linked) rather than just mentioned, making Perplexity the highest-quality citation signal among the tracked platforms. A Perplexity citation means the platform both found your content via retrieval AND considered it authoritative enough to cite explicitly.

What brands should optimize for Perplexity

Ensure crawlability. Verify that robots.txt allows PerplexityBot. Many sites block AI crawlers by default. Also check whether your hosting provider is IP-blocking Perplexity’s crawlers at the infrastructure level — this is common with Cloudflare and similar CDNs. Publish frequently. Perplexity favors recent content more heavily than other platforms. Pages with “last updated” dates from 2-3 years ago are systematically passed over in favor of recently-updated content on the same topic. Regular content refreshes are high-leverage for Perplexity visibility. Structure content for retrieval. Clear headings, concise answer paragraphs, schema markup, and focused topical pages outperform long, unfocused articles. A dedicated page answering one specific question will outperform a comprehensive guide that buries the answer in section 7. Use llms.txt. Perplexity is one of the platforms most likely to use llms.txt for site discovery. A well-structured llms.txt file at your domain root helps Perplexity understand what your most important pages are. Earn editorial coverage. The publications Perplexity cites most frequently are your best PR targets. Run a baseline scan to identify which publications appear as cited sources for your category queries, then pursue editorial coverage on those specific publications.

Why Perplexity matters most for citation measurement

Perplexity is the only platform where citation rate is reliably measurable across queries. ChatGPT cites inconsistently. Claude rarely surfaces URLs. Gemini has variable citation behavior. For this reason, Cited uses Perplexity as the primary source of citation rate data. When Cited reports a brand’s citation rate, it is predominantly reflecting Perplexity behavior. This also means Perplexity is the platform where the distinction between citations and mentions is most operationally meaningful. On other platforms, most brand references are mentions without links. On Perplexity, a reference is typically a citation with a link — making it a direct traffic signal.

Frequently asked questions

Different retrieval mechanisms. ChatGPT blends training data with optional Bing search. Perplexity relies entirely on real-time retrieval. If your content does not rank well in Perplexity’s search index, you will not appear — even if the model “knows” your brand from training data.
Check your robots.txt for PerplexityBot directives. Many sites block AI crawlers by default. Also verify your hosting provider is not IP-blocking Perplexity’s crawlers at the infrastructure level — this is common with Cloudflare and similar CDNs.
Yes. Sources cited first (position 1-2) receive disproportionate attention from readers. Perplexity’s citation ordering reflects the platform’s assessment of source relevance. Cited tracks average position across platforms, which captures this effect.