Beyond the Recommendation: How Prompt Language Changes How AI Frames Your Brand

  • Psychology of Gen AI
  • ARF; MSI

Generative AI tools are not just recommending products—they are shaping how consumers perceive them. This latest issue in the Psychology of Gen AI series, the third phase of the seventh study on AI product recommendations, examines how small changes in prompt wording alter the explanations AI systems generate around the same brand. Using ChatGPT descriptions of Arm & Hammer Advance White toothpaste across multiple, shopping-related prompts, the study reveals how AI recommendations construct different product narratives, emphasize different benefits and even introduce different caveats depending on the consumer’s wording of their question.

Member Only Access

What Marketers Need to Know About Retail Media Measurement

  • ARF ORIGINAL RESEARCH

Retail Media Networks (RMNs) have become a cornerstone of modern advertising, but measurement practices across networks remain inconsistent and difficult to compare. This new ARF study examines 16 major RMNs to uncover how advertising assets, attribution systems, incrementality testing and closed-loop measurement are actually implemented in practice. The findings reveal that while reporting dashboards may appear unified, the underlying measurement systems often rely on disparate sources of retailer data, platform reporting, modeling and privacy-safe integrations. The report offers practical guidance for advertisers evaluating retail media investments and highlights the growing need for industry-wide measurement transparency and standards.

Member Only Access

The Co-Viewing Advantage: Maximizing Attention Across TV Platforms

  • JOURNAL OF ADVERTISING RESEARCH

Watching television is often a shared experience, but how does co-viewing affect audience attention? An analysis of nearly 25 billion seconds of TV viewing data reveals that watching with others decreases attention to advertisements while increasing attention to programming. The effects vary by platform, with linear TV outperforming CTV in sustaining attention during co-viewing situations. The study also finds that longer co-viewing sessions strengthen engagement, while longer ads can undermine attention. These findings offer important guidance for advertisers, content creators and media planners seeking to maximize audience engagement in an increasingly fragmented television landscape.

Member Only Access

Attention in Context: What Real Campaigns Reveal About Media Performance

  • ARF Attention Measurement Validation Initiative

What does “attention” really mean in today’s fragmented media landscape—and does more attention actually drive better results? In phase three of the ARF Attention Measurement Validation Initiative, real-world campaign data reveals where attention metrics align, where they diverge and how they relate to brand lift. The findings challenge assumptions and offer a more nuanced view of how attention works across channels, platforms and outcomes.

Member Only Access

Striking the Right Balance in AI-Personalized Advertising

  • JOURNAL OF ADVERTISING RESEARCH

Generative AI is opening new possibilities for hyper-personalized advertising, including the ability to create AI-generated faces that closely resemble individual consumers. But how similar is too similar? This Journal of Advertising Research study finds that while moderate facial resemblance can improve advertising effectiveness, excessive similarity may backfire. The research introduces a new framework for measuring facial similarity and identifies an optimal personalization threshold that maximizes purchase intentions while avoiding consumer discomfort and resistance.

Member Only Access

Scaling Causal Measurement: A New Path Beyond Attribution

  • ARF; MSI

Randomized controlled trials remain the gold standard for measuring advertising effectiveness, but they are often too costly and complex to deploy across every campaign. This MSI Working Paper introduces Predicted Incrementality by Experimentation (PIE), a new framework that uses data from a limited set of experiments to predict causal advertising effects for campaigns that are not directly tested. The research demonstrates that PIE can dramatically outperform traditional attribution metrics, offering marketers a practical way to scale causal measurement without scaling experimentation itself.

Member Only Access

Newest Members

Everyday Health Ibotta Merck Uber Walmart