Publishers Face 0.3% Click-Through from AI Search—Here's Why

Digital advertising publisher alliance Ozone has launched R&D Labs, a technology sandbox that lets publishers model how their content is processed, retrieved, and cited within AI answer engines such as ChatGPT, Google AI Overviews, and Perplexity.

Ozone's Simulation Platform Addresses Publisher Blind Spot

Eight publishers participated in a three-month pilot program before the public announcement. Ozone offers the platform free of charge as a research initiative.

The platform simulates AI responses using data from participating publishers rather than analyzing actual outputs from large language models (LLMs), which would raise proprietary concerns. This allows publishers to understand how their content is structured and cited in AI-generated answers.

"There is complete information asymmetry between what the publisher knows, how their content is being consumed, and what the LLMs know," said Damon Reeve, CEO of Ozone. Publishers are "completely in the dark" regarding content usage and performance within AI systems, he added, even when formal licensing agreements are in place.

Ozone's publisher network includes the Wall Street Journal, New York Post, BBC (U.S.), and CNN. The company plans to publish research findings within one month to benefit the broader publishing community.

A Near-Zero Click Rate Is Forcing Publishers to Rethink Content Strategy

The urgency behind this tool is rooted in a sharp traffic problem. Publisher click-through rates from AI search tools currently sit at approximately 0.3%, according to publishing executives cited by Digiday. Where traditional search once delivered meaningful referral traffic, AI answer engines now summarize content without sending users to source pages.

Zero-Click Searches Now Drive 60% of Queries; APAC PR Teams Rethink ROI

AI Overviews slash referral traffic by 47%, forcing Asian PR teams to abandon click-based ROI metrics entirely. Earned media citations now define visibility.

Mission MediaMission Media

McKinsey projects a 20 to 50% drop in referral traffic to publishers as AI tools displace conventional search. For Asian media houses and brand publishers already navigating fragmented digital audiences, this trajectory adds significant pressure.

Some publishing executives remain cautious. One anonymous executive described their current stance as a "watch-and-learn phase," citing the 0.3% click-through rate as making large resource commitments difficult to justify.

Early Commercial Products Show GEO Is Already a Revenue Category

Generative Engine Optimization (GEO), the practice of structuring content so AI systems are more likely to cite it, is moving from experimental to commercial. Future Publishing deployed Future Optic, a GEO product across its portfolio of brands including TechRadar and Tom's Guide, delivering a 33% uplift in AI citation performance for Samsung.

Ahrefs independently named TechRadar the most-cited domain globally on ChatGPT, giving Future Publishing a credible authority base from which to sell AI visibility as a product to brand advertisers.

Affiliate marketing platform CJ (Commission Junction) has also entered the space with a managed service that audits brand presence across five AI platforms simultaneously: ChatGPT, Gemini, Perplexity, Claude, and Microsoft Copilot. This signals that AI content visibility is becoming a managed service category, not just an editorial discipline.

What This Means for Asian Publishers and Brand Marketers

Measurement remains fragmented. Ahrefs ranks TechRadar as the top-cited domain on ChatGPT, while a separate analysis by Profound ranks Forbes first. The divergence across tools highlights the absence of standardized metrics for AI content visibility, a gap Ozone's simulation platform is directly positioned to address.

A Tow Center analysis of ChatGPT Search across 20 publishers found citation inaccuracies including misquotes, misattributions, and opinion-based claims presented as fact. This reinforces the case for publisher-side monitoring tools.

OpenAI provides confidential partner reports to select publishers containing granular metrics on URL displays and click-through rates. However, access is not universal, reinforcing the information asymmetry Ozone is working to counteract.

Ozone plans to publish its research findings openly, giving publishers outside its direct network access to early intelligence on how AI systems source and cite content.