Audience & Media Measurement

As the Cookie Crumbles: Perspectives on Cross-Platform Data from Every Corner of the Industry

As the Cookie Crumbles: Perspectives on Cross-Platform Data from Every Corner of the Industry

  • By Margaret Felger (Cint), Young Pros Officer
  • CROSS-PLATFORM MEASUREMENT COUNCIL

On February 5, 2025, the Offline-Online Metrics Working Group of the ARF Cross-Platform Measurement Council hosted a panel of measurement experts from different sectors of the industry to discuss cross-platform measurement challenges and opportunities in today’s evolving data landscape. The session kicked off with a presentation from Rishi Saxena (World Federation of Advertisers) on the WFA’s Findings for Cross Media Measurement and Advertising Needs, which covered issues that marketers face around media fragmentation, frequency, data challenges, and need for new solutions. Following the presentation, the panel members discussed how their respective companies are facing these challenges and how they are preparing for the future. Working Group Chair Charles Buchwalter moderated the engaging conversation with Karen Chisolm (Pernod Ricard), Lee Doyle (Empower Media), Neil Napolitano (DotDash Meredith), and Working Group member Rishi Saxena (WFA).

Member Only Access
7th Annual (2024) Privacy Study

7th Annual (2024) Privacy Study

  • ARF Original Research

The ARF's 7th Annual Privacy Study surveyed 1,242 American consumers to understand their attitudes towards online privacy, data sharing and trust in institutions. This impactful perennial survey for the first time this year even gauged people’s feelings on AI. The study revealed a decline in perceived knowledge about online privacy, with only 40% of respondents feeling well-informed, down from 46% in 2023. Trust in media and brands also declined, particularly among younger demographics, while medical and financial institutions retained higher trust levels.

The study also highlighted increased resistance to data collection, even when tied to personalization or improved ad experiences. Consumers showed a growing aversion to sharing sensitive information and a heightened sensitivity to data breaches. Emerging concerns about AI and its impact on privacy were also noted, with AI platforms ranking among the least trusted institutions.

Member Only Access
An Introduction to Robyn’s Open-Source Approach to Media Mix Modeling

An Introduction to Robyn’s Open-Source Approach to Media Mix Modeling

  • MSI

As privacy-centric changes reshape the digital advertising landscape, deterministic attribution and measurement of advertising-related user behavior are increasingly constrained. In response, there has been a resurgence in the use of traditional probabilistic measurement techniques, such as media and marketing mix modeling (m/MMM), particularly among digital-first advertisers. To address the gap for small and midsize businesses, marketing data scientists at Meta have developed the open-source computational package Robyn, designed to facilitate the adoption of m/MMM for digital advertising measurement.

Robyn is a widely adopted and actively maintained open-source tool that continually evolves. This article explores the computational components and design choices that underpin Robyn, emphasizing how it “packages up” m/MMM to promote organizational acceptance and mitigate common biases. The solutions described are not definitive but outline the pathways that the Robyn community has embarked on.

Member Only Access
Evaluate Identity Resolution Effectively with this Council Guide

Evaluate Identity Resolution Effectively with this Council Guide

  • ARF ORIGINAL RESEARCH

Identity resolution (IDR) is crucial in media measurement and advertising, connecting media messaging to individuals. This guide, produced by the ARF Identity Resolution Working Group (of the ARF Cross-Platform Measurement Council), explores different units of analysis in IDR beyond individuals, such as households, geography and cohorts and their implications for matching quality, targeting and marketing success.  

Member Only Access
Learn How to Train LLMs to Identify Implicit Consumer Needs

Learn How to Train LLMs to Identify Implicit Consumer Needs

This study explores the potential of large language models (LLMs) to revolutionize marketing research. By partnering with a Fortune 500 food company, the authors replicated qualitative and quantitative studies using GPT-4. The findings indicate that LLMs can effectively generate synthetic respondents, moderate in-depth interviews and perform data analysis tasks, matching or even surpassing human performance in certain aspects. The study highlights the benefits of a Human-LLM hybrid approach, where LLMs assist in various stages of the research process, from study design to data analysis. This approach not only enhances efficiency but also uncovers new insights that might be overlooked by human researchers alone.

Member Only Access
LOLA: Revolutionizing Content Experiments with LLM-Assisted Online Learning

LOLA: Revolutionizing Content Experiments with LLM-Assisted Online Learning

In the rapidly evolving digital content landscape, media firms and news publishers require automated and efficient methods to enhance user engagement. This study introduces the LLM-Assisted Online Learning Algorithm (LOLA), a novel framework that integrates Large Language Models (LLMs) with adaptive experimentation to optimize content delivery. Leveraging a large-scale dataset from Upworthy, which includes 17,681 headline A/B tests, the study investigates three pure-LLM approaches and finds that prompt-based methods perform poorly, while embedding-based classification models and fine-tuned open-source LLMs achieve higher accuracy.


LOLA combines the best pure-LLM approach with the Upper Confidence Bound (UCB) algorithm to allocate traffic and maximize clicks adaptively. Numerical experiments on data from the website Upworthy show that LOLA outperforms the standard A/B test method, pure bandit algorithms and pure-LLM approaches, particularly in scenarios with limited experimental traffic. This scalable approach is applicable to content experiments across various settings where firms seek to optimize user engagement, including digital advertising and social media recommendations.

Member Only Access
 Next »