The IAB Tech Lab’s Privacy Sandbox Task Force just unleashed its final Fit Gap Analysis, and let me tell you, it’s like watching a train wreck in slow motion. The conclusions are no more reassuring than a clown at a funeral. For anyone hoping the February report was a fluke, bad news: the Sandbox still sucks. Big time.
Picture this: it’s August 2023, and the IAB Tech Lab assembles a brain trust of senior leaders with more experience in digital media than most of us have in avoiding spam calls.
These folks, many of whom practically invented OpenRTB, came from 65 companies to dissect the Chrome Privacy Sandbox APIs like a high school biology project. The goal? Identify crucial digital advertising use cases and see if they stand a chance within Google’s so-called “Protected Audience” auction.
Spoiler alert: They don’t.
The task force’s mission was ambitious—think of it as trying to climb Everest with a broken leg. First, they wanted to give the ad tech industry a reference point compared to today’s standards. Next, they aimed to rally the industry for some serious testing. Finally, they hoped to hand Google’s Chrome Sandbox team a roadmap of feedback, illuminating the product and operational nightmares of the PAAPI.
These valiant souls spent countless hours poring over Sandbox documentation, trying to figure out if their essential use cases could be met. The result was a 106-page tome released to the public for comment in February 2024. And boy, did the industry have things to say. The feedback poured in, like complaints at a cable company’s call center, forcing the task force to wade through hundreds of comments.
Google’s engineers even joined the fray for a robust discussion, which sounds like code for a polite but heated argument. The focus was to pinpoint technical blunders or overlooked APIs, resulting in some updates to the final draft, which, drumroll please, is being released today. There’s a handy table on page 9 summarizing these changes, with a fully detailed change log available for those with insomnia.
Two use cases were unceremoniously dumped: Second Price Auctions, axed due to low adoption. One case, “Multi-Touch Attribution,” got a promotion from “Not Supported” to “Impractical”—an improvement akin to moving from the basement to the attic in a burning house. “Exclusion Targeting” jumped from “Not Supported” to “Degraded,” a two-step upgrade, while “Publisher Revenue and Accrual Validation” took a nosedive from “Temporarily Supported” to “Not Supported.” Five other use cases had minor tweaks, akin to rearranging deck chairs on the Titanic. And a couple of cases, “Bid Using a Deal ID” and “Receive a No Bid Response from a DSP,” didn’t get updates but were deemed complex enough to warrant further discussion.
The task force’s meticulous efforts over two months of weekly meetings led to today’s grand finale. Now, they’re pivoting to work with Google’s product and engineering teams to hammer out a wish list of features that could—fingers crossed—make the Privacy Sandbox marginally useful. They’ll publish this list in Q3 2024.
But let’s cut to the chase: the task force’s verdict on the Privacy Sandbox remains as bleak as ever. They maintain that it’s woefully inadequate for supporting a vibrant open web, balancing advertising utility for brands against the media companies’ ability to make a buck. In its current form, the Privacy Sandbox threatens to choke the digital media industry, hampering the delivery of relevant, effective advertising and putting smaller media companies and brands at significant risk. It’s like trying to drive a car with square wheels: you’re not going anywhere fast, and you’re damaging everything in the process.
Amid this digital dystopia, Criteo offered a sliver of hope—or at least, a slightly less depressing outlook. Based on their latest tests, they delivered some good news and some bad news about the Chrome Privacy Sandbox. Brace yourselves.
The bad news? If Google killed third-party cookies today and rolled out the current version of the Privacy Sandbox, publishers would see their ad revenue on Chrome nosedive by around 60%. Imagine your paycheck shrinking by more than half overnight. Not a pretty picture.
The slightly better news is that the Privacy Sandbox could, theoretically, become a workable alternative to third-party cookies. But—and it’s a big but—Google needs to make some significant tweaks. Todd Parsons, Criteo’s chief product officer, emphasized that these changes aren’t about adding new features but refining the existing ones. It’s like asking for your burger to be cooked medium instead of well-done—you’re not changing the menu, just making it palatable.
Parsons’ optimism stems from Criteo’s rigorous testing. They’ve been poking and prodding the Privacy Sandbox since 2021, back when the APIs had cutesy bird names. Their latest round of testing, spanning mid-March to mid-May, involved 18,000 advertisers, 1,200 publishers, and 100 million weekly ad impressions. The results? Alarming, to say the least.
Without third-party cookies, publishers’ revenue on Chrome would tank by 60%. Ads would take twice as long to load, and Google’s own ad business would gain an unfair advantage, potentially boosting its market share from 23% to a staggering 83%. It’s like playing Monopoly with someone who already owns Boardwalk and Park Place.
Parsons pointed out that broader, more rigorous testing is essential. Many ad tech vendors have only dipped their toes in the water, and publisher adoption of the Privacy Sandbox remains below 55%. This half-hearted engagement is partly due to Google’s repeated delays and the daunting task of overhauling existing advertising infrastructure.
Criteo’s approach involved rebuilding their performance pipeline from scratch—not just the auction and bidding components, but also targeting, product recommendations, creative, and measurement. It’s akin to renovating an entire house while still living in it, an exhausting and monumental task.
So, where does this leave us? The Privacy Sandbox, in its current state, is a mess. It’s a well-intentioned idea bogged down by practical shortcomings and technical hurdles. The industry needs to keep the pressure on Google to refine and improve the APIs, or risk watching the digital advertising landscape crumble.
For now, the Privacy Sandbox remains a work in progress—a flawed, frustrating, and sometimes farcical attempt to navigate the post-cookie world. Stay tuned, folks. The saga is far from over.
