Google’s FLEDGE lacks momentum 4 months into trials, offers ‘limited focus’

privacy sandbox

Barely four months after launch, trials of Google’s version of cookieless retargeting are stuck in first gear. Only five ad tech vendors have expressed an interest in testing the FLEDGE, or the “First Locally-Executed Decision over Groups Experiment,” to date. And of those, only RTB House, Criteo and Google itself are doing so in any meaningful way. 


  • Number of vendors testing FLEDGE: 5
  • It’s a single-browser solution
  • Lacks reach and frequency controls
  • Only handles the retargeting part of performance advertising
  • Google is the only gatekeeper

Don’t expect this to change anytime soon. 

It turns out most ad tech companies just aren’t keen on FLEDGE — which is understandable. Supporting it now is a bit like jumping into the unknown: No one can afford to be first, especially when there are so many unresolved issues. 

In short (but no particular) order, the issues include: FLEDGE only handles the retargeting part of performance advertising. So marketers won’t know if the users who saw a retargeted ad converted, as that’s a whole other work stream from Google. There are also no proper reach and frequency controls. Plus, there’s the elephant in the room: Based on the current specifications for FLEDGE, it would seem that there’s one gatekeeper — and to no one’s surprise that appears to be Google. 

Add all of this up and it’s not surprising that the specialists in network (RTB House) and retargeting (Criteo) advertising would be among the first (and only) companies to test FLEDGE. They can see a way to make it work for their businesses. But for everyone else, the downside is too sharp and the upside too ambiguous. 

As with everything else in Google’s Privacy Sandbox, FLEDGE is like the proverbial chicken and egg: Companies are trying to assess how much traction FLEDGE will get before they test it, but few companies want to test it to begin with. 

“From what we are seeing and hearing on FLEDGE there is a very limited focus and no one is showing high volumes and performance yet,” said Rob Webster, global vp of strategy at digital marketing consultancy CvE. 

That’s not to say FLEDGE trials will be stuck in the slow lane forever. There’s no doubt the interest is there, but it’s not enough to get most ad execs off the fence — not yet anyway. For now, they would rather wait and see. 

“While we are closely monitoring the progress of FLEDGE, we are not actively testing FLEDGE because we have not encountered much customer demand for testing, and it remains a single browser solution,” said a MediaMath spokesperson. “Also, FLEDGE is a remarketing use case that advertisers use to drive conversion, as opposed to awareness. As such, testing the measurement of advertiser conversions should be coupled with FLEDGE.”

Even the ones who are testing FLEDGE are taking things one step at a time. 

“So far we are pleased with our progress because we’ve seen improved effectiveness on our device deep learning algorithm for FLEDGE,” said Lukasz Włodarczyk, vp of programmatic ecosystem growth and innovation at RTB House. “We plan to add additional complexity to our tests, which means working to improve the integrations with publishers and supply-side platforms and eventually get to a point where we can support a fully programmatic auction inside the FLEDGE environment.”

We plan to add additional complexity to our tests, which means working to improve the integrations with publishers and supply-side platforms and eventually get to a point where we can support a fully programmatic auction inside the FLEDGE environment.
Lukasz Włodarczyk, vp of programmatic ecosystem growth and innovation, RTB House

It’s a similar story at Criteo, insofar as executives there are still figuring out how to get the most out of FLEDGE. In fact, they’re still evaluating the results of the test, which should be ready to share by November, said a spokesperson for the company. Until then, they’re not discussing the matter. 

Whether support for FLEDGE rallies once more information about these tests is made available remains to be seen. On the one hand, it could be good because at least other ad execs will be able to see what they’re missing. But on the other hand, FLEDGE’s issues arguably run deeper than this.

Google’s apparent influence over FLEDGE is a case in point. For FLEDGE to work, it needs to take an input of the winning bid from an auction overseen by Google Ad Manager (Publisher’s Ad server). In this scenario, Google picks the final ad. This way, publishers and their ad tech partners can’t disintermediate Google. That changes if FLEDGE API on-device top level auction picks the final ad and when Google Ad Manager participates as equivalent participant in component auctions. Put simply, FLEDGE is predicated on auction decisions happening in the Chrome browser that Google owns — something that could have a material impact on how much money SSPs make. 

“Both SSPs and publishers we talk to all say they want full control over the top level auction,” said Włodarczyk. “What publishers want (from our understanding) is they want full control over the auction in the way they have with Prebid. Every SSP including the one owned by Google could be an equivalent participant in the auction.”

This is all to say change won’t come easy for FLEDGE. That’s not an indictment, it’s an acknowledgment of one, simple fact: There’s no quick fix for this part of Google’s Privacy Sandbox. This is the same as it ever was, but it’s just clearer now. Even as recently as the summer, there were reports that its take-up had been slow among some parts of the market. And while FLEDGE’s problems are likely to continue for a while yet, the industry (and Google for that matter) remains hopeful that they will eventually be overcome.

It does, after all, offer a solution to one of publishers’ big bugbears about ad tech: data leakage. FLEDGE essentially gates off a publisher’s audience data so that advertisers and the technology they use to buy those audiences can’t study that data and buy it cheaper elsewhere.

“We’re encouraged by early integration testing from companies like RTB House and by the enthusiasm from publishers to move their inventory to a privacy-preserving supply chain,” said a spokesperson from Google. “As we ramp tester integrations and origin trial sizes, we look forward to continued industry input this year on the functionality of the APIs and next year on performance. Our new timeline accommodates for this phased approach and provides time to make adjustments to improve the privacy and effectiveness of the APIs based on the results from testing.”

More in Marketing

Cannes Briefing: Sports hold full court at Cannes Lions 2024

Sports isn’t just a sideshow at Cannes Lions 2024, it’s taking center stage.

HR and marketing are more connected than they used to be

A highly competitive job market and the rise of employer branding have necessitated a closer collaboration between HR and marketing.

Digiday Podcast at Cannes: Inside Instacart’s plans to make every surface shoppable with CMO Laura Jones

Instacart is on a mission to make every surface shoppable, pitching that to advertisers at this year’s Cannes Lions festival.