Future States

• Audience & Media Measurement
• Article

Explainable AI (XAI) Helps Minimize the Impact of Errors

• MSI

The use of AI-based voice assistants is becoming ubiquitous, and this technology continues to develop at a rapid pace. Now, Explainable AI (XAI) can help customers understand things better and help mitigate certain kinds of errors. This research discovered that XAI can help consumers forgive minor social faux pas (violations of social norms) but not minor technical errors (failures of the algorithm interface). The series of studies also found that XAI helps consumers overcome their reluctance to use such AI-assisted technologies.

• Research & Data Quality
• Article

The ARF Attention Measurement Validation Initiative: The How & Why

Today, there is much discussion and debate on the definition of attention, how to measure it properly and whether we should be measuring it at all, or if inattention instead should be what we assess. Plus, is the highest form of attention always desired, or do we want low levels of attention for certain kinds of ads, such as emotional ones for TV? As a result of all these questions, the ARF has announced the Attention Validation Project. Its goals are to help define attention, evaluate the methods available to us and what their rightful application is, whether it be analyzing advertising creative or the media environment.

• Targets & Segments
• Article

How AI and Deepfakes Could Reshape the Advertising Industry

• JOURNAL OF ADVERTISING RESEARCH

Artificial intelligence is already changing advertising, but how will it reshape it moving forward? Researchers unpack the potential effects of three different forms of AI—analytic, interpretive and creative—on seven categories of advertising industry stakeholders. Can those who embrace the evolving creative AI mechanisms get ahead of competitors?

• Research & Data Quality
• Article

Can Observational Methods Assess the Causal Effect of Ads?

• MSI

While Randomized Controlled Trials (RCTs) are ideal for advertising measurement, it is not always feasible to conduct one. Until advertising platforms are willing to provide more information about how they deliver targeted ads or implement auctions, alternative observational methods are unlikely to reliably estimate causal effects. That includes utilizing newer machine learning techniques.

• Analytics & Data Science
• Article

Optimizing Interventions Along the Customer Journey

• MSI

Random controlled experiments for A/B testing help improve things like a company's marketing or customer service. However, individually optimizing interventions may not always capture interactions across the entire purchase decision journey. To optimize interventions more holistically, use a Bayesian reinforcement learning model. It can integrate multiple historical experiments, which can improve both current impact as well as future learning.

• Audience & Media Measurement
• Article

The ARF Attention Report

• ARF & THE ATTENTION COUNCIL

The ARF partnered with the UK’s Attention Council on this project, whose aim was to gauge the level of interest in attention metrics among ARF members and the Wonks community. Sixty-three participants took the survey in January of this year (2022). The qualitative study focuses on variations among the industry’s sectors in their outlook on attention metrics. The differences in outlook between researchers and the buy-side were particularly interesting.

• Research & Data Quality
• Article

Does the Hierarchy of Ad Effects Still Work?

• MSI

Advertising can trigger cognition, affect or experience, according to the influential Hierarchy of Effects (HoE) model. The following study valides the basic integrated HoE model. However, it also finds that operating sequence varies across brands. The predominant one is affect → cognition → experience (ACE).

• Research & Data Quality
• Article

ARF Academic Journal Highlighted for Industry Authors, Expansive Themes

• JOURNAL OF ADVERTISING RESEARCH

In December 2020, the Journal of Advertising Research published an independent seminal bibliographic study that identified key publishing trends since the academic journal’s inception in 1960. Now, a more granular overview expands on those findings to reveal the prominence of practitioner authors amid a wide range of themes, while encouraging future submissions.

• Analytics & Data Science
• Article

Improving Product Sales Predictions Is Brain Science

• MSI

It’s no surprise that new product launches often fail to meet their targets. The trick for managers is to improve their predictions for such products. They must balance the costs and benefits of many different data sources and analytic techniques in order to improve forecasting. To enhance the accuracy of predicting the market-level sales of new products, researchers Marton Varga, Anita Tusche, Paulo Albuquerque, Nadine Gier, Bernd Weber, and Hilke Plassmann, analyzed the added value of different data types. Their conclusions are illuminating.

• Article

JAR: Moment-to-Moment EEG Metrics Enhance Ad Effectiveness Analysis

A European research team is on a mission to advance the development and use of electroencephalogram (EEG)-based methods for evaluating advertising effectiveness. Although their research leaves open questions, it suggests that recent developments in this area—specifically moment-to-moment EEG-based indicators—provide a clearer view into emotional response and attention to ads than previous EEG methods. Read the article.