Study on behavioral ad targeting finds that publishers only get about 4% more revenue for an ad impression that has a cookie enabled than for one that doesn't
Keach Hagey / Wall Street Journal : Tweets: @johnwilander , @jason_kint , @robinberjon , @hoofnagle , @ashk4n , and @profcarroll . Thanks: @cremnob Tweets: John Wilander / @johnwilander : “Publishers only get about 4% more revenue for an ad impression that has a cookie enabled than for one that doesn't. The study tracked millions of ad transactions at a large U.S. media company over the course of one week.” https://www.wsj.com/... Jason Kint / @jason_kint : Huge. This research pops myth that behavioral advertising is critical to media. It also explains the tight lobby between ad tech, platforms and advertisers and how its interests likely has been screwing legit publishers all along. https://twitter.com/... Robin Berjon / @robinberjon : Behavioural advertising isn't paying publishers. It's not what brings you free stuff on the Internet. It's just how Google and the rest of adtech sells your data. What value it does have - to advertisers - can be delivered without surveillance. So let's. https://www.wsj.com/... Chris Hoofnagle / @hoofnagle : WSJ: in one of the first credible “empirical studies of the impacts of behaviorally targeted advertising on online publishers' revenue, researchers...suggest publishers only get about 4% more revenue for an ad impression that has a cookie enabled...” https://www.wsj.com/... Ashkan Soltani / @ashk4n : The house of cards that is OBA is finally starting to collapse. New study by @ssnstudy highlights that “publishers only get about 4% more revenue for an ad impression that has a cookie enabled than for one that doesn't.” (via @keachhagey)https://www.wsj.com/ ... David Carroll / @profcarroll : STUDY: Advertisers pay more than double for privacy-invasive behavioral ads but a new study shows that publishers only see 4% boost in revenue from behavioral ads. 60% of an ad dollar funds adtech. Yet no data available to measure adtech's alleged value in the supply chain https://twitter.com/... Thanks: @cremnob
Context & Ripple Effects
This study lands weeks after the New York Times bought targeted ads against 16 profile attributes to show how much advertisers think they know about readers (the NYT's own targeting experiment), and it attacks the same question from the money side: tracking millions of real transactions at a large U.S. media company, it finds a cookie-enabled impression earns only about 4% more than one without.
That number feeds a skepticism already building in the coverage — critics argue ad effectiveness benchmarks can't tell buyers-who-would-have-bought from buyers-convinced-by-ads (the measurement-benchmark critique) — and it prefigures the industry's exit from third-party data, starting with the Times building its own first-party audience segments (NYT's first-party data platform) rather than waiting for the cookie question to resolve itself.
First-order effects
- Publishers now have a quantified argument that consenting to third-party cookies costs them little in revenue, weakening the case for consent walls and for paying adtech intermediaries a premium for 'targeted' inventory.
- Adtech firms whose pitch rests on behavioral lift face direct pressure to prove value beyond that ~4% margin, since the study tracked actual transactions rather than click-through proxies.
Second-order effects
- Advertiser budgets migrate toward walled gardens with logged-in first-party data — the same dynamic later visible when Apple's ATT pushed up Meta's ad prices and some buyers shifted spend to Google, Amazon, and TikTok (advertisers rerouting around Meta's ATT price increases) — leaving the open web to compete on context instead of identity.
Third-order effects
- If behavioral lift stays marginal while privacy regulation and browser changes keep raising its cost, the structural endpoint is an open-web ad market rebuilt on first-party relationships and contextual signals, with identity-based intermediaries consolidating or shrinking.
The trend: The measured premium for behavioral targeting is small enough that publishers are trading third-party cookies for first-party data and contextual approaches, reordering who captures ad dollars across the open web and walled gardens.