This might trend toward controversial territory, given that Prebid.js has been the foundation of how the open internet (certainly the web portion) has monetized itself for nearly a decade. But it’s worth re-examining what actually made Prebid successful — because the answer isn’t as simple as I (and perhaps others) once thought.
When Prebid was first released, two things were undeniably true:
- Prebid made publishers more money.
- No one trusted Google.
At the time, most people assumed those two facts were connected — that Prebid made publishers more money because Google’s ad server (then called DFP, now called GAM) wasn’t fairly allocating impressions. That assumption was true. The DOJ antitrust trial has recently revealed that Google AdX’s take rate was roughly 5 percentage points higher than the market average (20% vs. 15%). So, in effect, Prebid erased that hidden “Google tax” and put (at least) 5% back in publishers’ pockets (and prevented Google from influencing the market in other ways as well).
But there was something else happening under the hood.
Before header bidding, the browser would render a page, call one ad server (almost always DFP), and then that ad server would sequentially request bids from SSPs or networks. Prebid flipped that logic on its head: it let the browser call all SSPs simultaneously before the page loaded, then after all bids came back, an ad was selected.
That subtle change — disassociating impression requests from actual impressions on the page — created a powerful new behavior in the ecosystem: bid request duplication (which I wrote about last week).
And here’s the controversial bit:
Prebid made publishers more money for two reasons:
- It removed Google’s bias.
- It multiplied bid requests.
The question that no one can fully answer is: which factor mattered more?
Did publishers’ revenue increase primarily because Google’s bias was eliminated — or because the same impression opportunity was duplicated across parallel bid requests?
It’s difficult to prove this objectively, but there’s a strong case that bid request duplication was the larger driver. It created a massive short-term revenue spike for publishers — and a powerful incentive for everyone in ad tech to adopt Prebid as fast as possible.
Some people recognized this even at the time. Jonathan Bellack (then at Google) debated Tom Shields about the long-term consequences of header bidding. I (and perhaps many of you?) was in the audience. Bellack’s argument — that Prebid’s efficiency gains were offset by waste and duplication — was rational, even prophetic. But it didn’t matter.
Google had already lost the trust of the industry. And once trust is gone, logic doesn’t matter.