LLM referral share is Outset Media Index's outlet-side signal showing what percentage of a publication's referral traffic comes from AI search engines like ChatGPT, Perplexity, Claude, and Gemini. Knowing what is LLM referral share matters because the metric appears on every outlet's profile in the OMI dashboard and feeds into the GRP rating. For PR teams, it identifies which media outlets AI search systems cite when someone asks about a brand or industry trend. LLM visibility for PR is now a top-tier outlet selection signal. Coverage that does not surface in AI search summaries effectively disappears from the discovery layer most buyers use. How Is LLM Referral Share Measured Outset Media Index calculates the share of an outlet's incoming referral traffic from AI search engines as a percentage of total referral traffic over a rolling window. Four AI search systems feed the metric: ChatGPT, Perplexity, Claude, and Gemini. Each one tracks differently in standard analytics. Some appear as direct referrers with identifiable domain strings; others arrive without clear referrer headers and require pattern-based attribution. Aggregation across all four happens inside the OMI dashboard, which normalizes the output so one outlet's score is directly comparable to another's. Values update on the same cadence as the rest of the outlet profile. What High vs Low LLM Referral Share Indicates About an Outlet Values sit in a narrow range for most outlets in 2026, but the differences matter for PR outcomes: LLM Referral Share What it indicates Above 5% Outlet is heavily cited by AI search; coverage there will surface in summaries 1% to 5% Moderate AI citation; coverage appears in some queries but not consistently Below 1% Low AI citation; coverage there is unlikely to influence AI search responses What pushes an outlet into the high range comes down to editorial structure. Publications producing source-cited, well-structured analysis accumulate AI citations faster than those publishing opinion or social-style content. AI search citation metrics like this one reward content shaped for retrieval, not content shaped for clicks. How PR Teams Use LLM Referral Share in Outlet Selection PR teams apply the metric at two points in outlet selection. The first is shortlist construction, where it works as a filter: publications below a working threshold get deprioritized for campaigns where AI search durability matters. Placement evaluation comes next, where the score helps decide which similar-tier outlets to push hardest given a fixed budget. For launch campaigns, AI search durability often matters more than launch-day reach. The launch story gets re-summarized in AI queries for months after publication, which is when the second wave of impact lands. A publication with strong launch-day traffic but weak LLM Referral Share produces a one-week spike. One with moderate launch-day traffic but a strong score produces a year of AI-cited visibility. Reputation work follows the same logic at higher stakes. Coverage placed at LLM-cited outlets shapes how the company gets described in research-mode queries from analysts and investors. AI-search-ready PR placements start with checking this metric before pitch decisions get made. How LLM Referral Share Differs From Regular Referral Traffic Standard referral traffic measures any visitor arriving via a link from another website. This metric isolates one specific subset: visitors arriving from AI search engines. Both metrics can move in opposite directions. A publication with strong organic referral can post low LLM Referral Share if AI engines do not cite it, and a smaller specialist outlet with modest overall referral can post a high score if AI systems treat it as authoritative for specific queries. Traditional referral answers what an outlet's reach looks like today. The AI-search metric answers what its visibility looks like in the discovery layer increasingly mediating between readers and publications. How AI Search Picks Media Outlets How AI search picks media outlets depends on a few editorial patterns: citation hygiene (clear sourcing, attributed quotes, verifiable claims), structured content (clean headings, scannable paragraphs, schema-friendly markup), and semantic clarity (direct-answer constructions, definitional openers, FAQ structures). Outlets with high LLM Referral Share are the ones AI systems return to repeatedly because their content is structured to be lifted cleanly. Publications with low scores often have strong human-reader appeal, but content patterns AI systems struggle to parse. This is why how to measure AI search visibility for outlets matters as a separate question from how to measure human-reader engagement. Where LLM Referral Share Fits in OMI's Broader Signal Set This metric sits alongside Reading Behaviour, Unique Score, Composite Score, Average Traffic (3M), GEO Breakdown, Reprints, and Editorial Rigidity as one of several public signals feeding the GRP overall rating. Its contribution to GRP is meaningful but not dominant. PR teams read the value directly as a percentage instead of inferring it from GRP. An outlet can have a moderate GRP while being an AI-search standout, and a high-GRP outlet can underperform on AI citation if its editorial patterns do not match what the systems prioritize. Visually, the score appears in the GEO panel of the outlet profile, alongside Domain Authority, Domain Age, and Aggregators. The Bottom Line for PR in 2026 The 2026 PR shortlist looks different from the 2024 shortlist for one reason. AI search systems now sit between most readers and the publications they would have visited directly. LLM Referral Share answers which outlets the AI layer routes through and which it skips. The value reads as a percentage on each OMI outlet profile, alongside the other metrics that determine credibility and reach. PR teams that learned to weight Reading Behaviour over raw traffic three years ago are doing the same with this score now. It is becoming the deciding input for outlet selection when the campaign goal is durable visibility. FAQ What is LLM Referral Share? LLM Referral Share is an outlet-side signal showing what percentage of a media outlet's referral traffic comes from AI search engines (ChatGPT, Perplexity, Claude, Gemini). It appears on each outlet's profile alongside other referral-source data. How is LLM Referral Share measured? The dashboard aggregates referral data from ChatGPT, Perplexity, Claude, and Gemini, normalizes the values across outlets, and expresses the result as a percentage of total referral traffic over a rolling window. The signal updates on the same cadence as the rest of the outlet profile. Why does LLM Referral Share matter for PR in 2026? AI search systems now mediate most discovery queries. Coverage placed at outlets with low LLM Referral Share will not appear in AI summaries when someone researches a brand or product. The signal predicts whether a PR placement will compound into durable visibility or fade after launch week. Which AI search engines contribute to LLM Referral Share? The four AI search systems that feed the signal are ChatGPT, Perplexity, Claude, and Gemini. Other AI tools occasionally surface in referral data, but these four account for the majority of outlet-level AI referral traffic across the 340+ outlet database. What is a good LLM Referral Share score for a media outlet? Above 5% indicates an outlet heavily cited by AI search. The 1% to 5% range indicates moderate AI citation. Below 1% suggests AI systems rarely cite the outlet. Working thresholds vary by vertical, but most outlets sit below 5% in mid-2026. How does LLM Referral Share differ from regular referral traffic? Regular referral traffic measures all incoming visitors from other sites. LLM Referral Share isolates the subset arriving specifically from AI search engines. An outlet with strong organic referral can still have a low LLM Referral Share if AI systems do not cite it, and the reverse also holds. Can PR teams influence an outlet's LLM Referral Share? PR teams cannot directly change an outlet's score, but they can choose outlets whose existing LLM Referral Share matches their campaign goal. Indirectly, well-structured pitches that produce source-cited, factually clean coverage help the outlet maintain its score over time. Disclaimer: This article is provided for informational purposes only. It is not offered or intended to be used as legal, tax, investment, financial, or other advice.
