Measuring Online Advertising Viewability and Analyzing its Variability Across Different Dimensions

Abstract

Many of the current online business base completely their revenue models in earnings from online advertisement. A problematic fact is that according to Google more than half of display ads are not being seen. The International Advertising Bureau (IAB) has defined a viewable impression as an impression that at least 50% of its pixels are rendered in the viewport during at least one continuous second. Although there is agreement on this definition for measuring viewable impressions in the industry, there is no systematic methodologies on how it should be implemented or the trustworthiness of these implementations. In fact, the Media Rating Council (MRC) announced that there are inconsistencies across multiple reports attempting to measure this metric. For this reason, we select a subset of implementations to track viewable impressions and we perform a case study by implementing them in a webpage registered in the worldwide ad-network ExoClick in order to see their results on different dimensions. Our results show that the Intersection Observer API is the implementation that detects more viewable impressions and that there are significant viewability differences depending on the banner location on the website. Finally, we also propose an ensemble viewability method that proves to be able to detect a higher number of viewable impressions.