Research Talk by Garrett Johnson, Boston University (US)

As part of the Marketing Research Seminar Series, Garrett Johnson presented joint work with Kobayashi and Gu on a topic that sits right at the intersection of advertising, privacy, and policy: how does retargeting hold up when the underlying tracking infrastructure changes? Partnering with an advertiser intermediary, the team ran an industry-wide field experiment across more than 2,000 advertisers, comparing traditional retargeting with Google's Privacy Sandbox, which lets advertisers reach past site visitors without tracking cross-site browsing by moving the ad selling logic onto the user's device. The headline numbers are striking: retargeting lifts baseline conversions by 4.6%, and while removing third-party cookies cuts ad clicks and click-through conversions sharply, Privacy Sandbox recovers about 46% of what was lost. Once you adjust for ad spend, the gap narrows even further, with Sandbox reaching around 86% of traditional retargeting on click per dollar. The team also makes a convincing case that part of the remaining gap likely reflects limited supply-side adoption rather than a fundamental weakness of the technology itself.

What I really enjoyed about Garrett's talk was how different it felt from many of the other MRSS sessions, both in methodology and in subject matter. I think this is one of the most valuable things about the series for PhD students: getting exposed to researchers who approach marketing from very different angles, whether that is behavioral lab work, structural modeling, or, in this case, a large-scale industry collaboration with real policy stakes. Each style of work asks different questions and accepts different kinds of evidence, and sitting through a talk that is far from your own methods is honestly one of the better ways to figure out what kind of researcher you want to become.