Effect of Disabling Third-Party Cookies on Publisher Revenue
Do you know how recent changes in digital advertising could impact your revenue? Google recently experimented to assess the effects of deactivating third-party cookies on programmatic ad income. By analyzing revenue and ad effectiveness with cookies removed for a segment of users, the study revealed a significant drop in income. As digital advertising keeps changing, publishers and advertisers need to be aware of the implications of such changes. Google recently experimented to investigate the impact of deactivating third-party cookies on programmatic ad income. The study evaluated the effects of deleting cookies for a portion of the user base on revenue and ad effectiveness in the real world. The results show a sharp drop in income and provide important context for understanding the crucial function cookies play in the ad ecosystem and the difficulties associated with navigating a world without them.ÂGoal
To empirically quantify the impact that disabling access to third-party cookies would have on the programmatic ad revenue of web publishers.Executive Summary
Using Google Ad Manager's programmatic component, we did a randomized controlled trial with publishers. This service displays advertisements on websites other than Google's. We restricted access to cookies for a subset of users (the treatment group) who were chosen at random for the trial. The treatment group's average revenue fell by 52% for the top 500 worldwide publications, with a median reduction of 64% per publisher.ÂMethodology
Google uses a powerful experimental framework to do A/B testing to randomized. We took advantage of this technology to essentially disable access to third-party cookies by limiting the availability of tailored data for a subset of randomly chosen users using Google Ad Manager's programmatic arm. Only non-personalized programmatic advertisements, such as contextual advertising that doesn't rely on third-party cookies, were displayed to users in the treatment group. We compared the treatment group's earnings to that of the control group, which received tailored advertising as usual. From May to August 2019, a total of 96 days were dedicated to gathering data.ÂResults & Statistical Analysis
The average revenue for the top 500 worldwide publishers fell by 52% in the treatment group, with a median per-publisher fall of 64%, according to the statistics, which showed a considerable impact. Publishers suffered varying percentages of revenue loss; some lost more than 50%, while some lost more than 75%. The confidence intervals for lesser publications are broader, notwithstanding the substantial results. Of the publishers studied, seven (1.5%) did not exhibit any loss; nevertheless, this discrepancy is explained by statistical noise. All but one of the top 200 publishers saw a revenue decline of more than 10%. The News vertical had a median revenue loss of 60% and an average revenue loss of 62% for publishers.Â
Related Work
This is the first publicly described randomized controlled experiment measuring the effect of disabling third-party cookies on publisher revenue. Previous studies based on observational data provide similar findings:- Johnson et al. reported a 52% revenue decrease from users opting out of online behavioral advertising.
- Beales and Eisenach observed revenue losses ranging from 37.5% to 66% for users without cookies compared to those with newer or longer-lived cookies.
Additional Reflections
We examined first-order impacts in our investigation. If third-party cookies are disabled, the following second-order consequences could occur: Client spending on advertisements has decreased as a result of non-personalized ads' worse return on investment; funds may have been diverted to other channels.- higher operating expenses when publishers modify their business plans to account for the lack of third-party cookies.
- There are several difficulties in quantifying these second-order effects, and this work was not designed to address them.
