The news: AI chatbot mentions and overviews now handle most shopping research that was once done on company websites, per MarTech, but traditional web analytics cannot see or measure any of it. Marketers who still measure performance by page views might be focusing on the wrong metrics.

As AI mediates product discovery, analytics remain trapped in a click-based past. Buyers now evaluate brands inside ChatGPT, Perplexity, and Google AI Overviews without needing to visit a website. 

AI citations are building branded search before referral traffic, per DemandLocal, which suggests brands track citation share, mention frequency, branded search queries, direct visits, and multi-touch assist rates—or the percentage of conversions where a channel appeared in the journey but didn’t receive last-click credit.

Why it’s worth watching: 60% of US marketers name AI-driven discovery platforms as their toughest attribution challenge, per EMARKETER and Partnerize. That dwarfs connected TV (CTV) at 39% and organic social at 37%. 

• Direct search volume offers the clearest signal. When users encounter a brand in an AI response, they often search for it by name minutes or days later. 
• A 30- or 90-day multi-touch window catches assisted conversions that single-touch reporting writes off, MarTech suggests, meaning budgets currently labeled “underperforming” may be doing more work than dashboards show.

Recommendations for brands: Faced with signal fragmentation, brands can seek out the metrics that are most relevant to their goals rather than waiting for a single standardized reporting model to emerge. 

• Monitor AI visibility continuously rather than with periodic spot-checks. Only 30% of brands remain visible from one AI answer to another, AirOps found, and just 20% were visible over five consecutive runs.
• Manage brand mentions across the surfaces marketers can directly influence—such as owned content (blogs and explainer videos), review sites, and community platforms like Reddit, YouTube, and LinkedIn.

These surfaces and channels are ones brands can actively shape through participation and attribution and where they can engage customers directly rather than being at the mercy of genAI outputs.