In the latest update on its plans to replace third-party cookies with ads, Google said tests on a particular proposal look promising.
Google planned to share some new findings showing the effectiveness of its “Federated Learning of Cohorts” proposal that is part of the Chrome browser’s “Privacy Sandbox” in a blog post running Monday. The “Sandbox” is an initiative launched in 2019 to find alternatives to the cookie while mitigating the impact on publishers and other players. In Google’s words, it was about finding a solution that both protects users’ privacy and ensures that content remains available for free on the open web.
Not long after announcing the initiative, Google said it would end support for third-party cookies, which power much of the digital ad ecosystem, in its Chrome browser within two years of January 2020.
Chrome engineers have worked with the wider industry, including web standards organization W3C, on ideas in the Sandbox suggested by Google and other ad technology players. What will likely result is some of these ideas moving forward, Google says.
“This is one proposal,” Chetna Bindra, group product manager for user trust and privacy at Google, told CNBC about “FLoC” progress. “It is by no means the last or the unique proposal to replace third-party cookies … There will not be one final API that continues, it will be a collection that enables things like interest-based advertising, and for the use of measurements, where it is essential to ensure that advertisers can measure the effectiveness of their ads. ”
Bindra said the company has “extraordinary confidence” in the progress of the proposals and testing so far.
Google’s post on Monday says that test results show that FLoC (pronounced like a flock of birds, in line with some bird-themed proposals such as’ Turtledove ‘and’ Swallow ‘)’ is an effective privacy-focused substitute signal for third-party cookies. It says advertisers can expect at least 95% of the conversions per dollar spent compared to cookie-based ads.
FLoC would essentially put people into groups based on similar browsing behavior, meaning that only “cohort IDs” and not individual user IDs would be used to target them. Web history and input to the algorithm would be kept in the browser, with the browser only exposing a “cohort” containing thousands of people.
“We really see that one of these first Sandbox technologies for interest-based advertising is literally nearly as effective as third-party cookies,” said Bindra. “There is definitely a lot more testing to come. We want advertisers and ad technologies to participate directly.”
Bindra said these cohorts, including people with behaviors such as an interest in gardening or rock music, would still allow targeting those interests. However, instead of targeting at the individual level, this would target groups.
“The difference will really only be that now they no longer track every user on the Internet. There’s really that idea of privacy for those users who are now clustered in a cohort,” Bindra said.
She added that the numbers from FLoC’s tests should be reassuring to publishers. Chrome will then make the cohorts available for public testing with the next release in March, and Chrome expects to begin testing FLoC-based cohorts with advertisers in Google Ads in the second quarter, the blog post said.
Myles Younger, a senior director in global data practice at MightyHive, said the Sandbox proposals are all about “ how can we build new features into the Chrome web browser to simultaneously conserve user privacy and the death of the third-party cookie. while preserving the ability of brands to advertise effectively. He spoke before Google’s latest findings were released.
One question is whether players will actually use it.
“I’m not sure it’s something Google is able to just flip a switch and turn on,” he said. “Publishers have to use it. People have to use this system. [Google] must prove that it works. ”
Paul Bannister, chief strategy officer at CafeMedia, said advertisers and publishers have some fear of the unknown as it relates to what comes next.
“I think we all want to believe that this will be good and we all want to move to a place where users have more privacy and the internet works better,” he said. But given how complicated and technical the process is, it’s unclear what’s really going to happen.
He said there is some concern that these types of actions could benefit the “walled gardens” of companies like Facebook and away from open web advertising.
The UK antitrust authorities are also aware of the plans and are investigating whether the plan to remove third-party cookies from Chrome could harm online advertising competition. The Competition and Markets Authority is investigating whether Google’s plans could cause advertisers to shift their spending to Google’s own tools at the expense of their competitors.
Commenting by email, Bindra said, “The Privacy Sandbox has been an open initiative from the start and we welcome the involvement of the CMA in developing new proposals to support a healthy, ad-supported web without third-party cookies . “
Some privacy advocates are also skeptical of the “FLoC” approach. The Electronic Frontier Foundation wrote in 2019 that these cohorts could be used in malicious ways, allowing discriminatory advertisers to identify and filter groups representing vulnerable populations.
“A herd name would essentially be a behavioral credit score: a tattoo on your digital forehead that provides a concise summary of who you are, what you like, where you go, what you buy and who you hang out with EFF staff technologist Bennett Cyphers wrote in the blog post. “The herd names are likely to be inscrutable to users, but can reveal incredibly sensitive information to third parties.”
Whether the machine learning would create cohorts based on health problems or low income status or other sensitive characteristics remains open to some.
“It has the potential to do very creepy and arguably illegal things,” Bannister said. “How does Chrome protect itself against this?”
Google said in documents that its analysis evaluates whether a cohort can be sensitive without knowing why it is sensitive, and that cohorts that reveal ‘sensitive categories’ such as race, sexuality, or personal hardship were blocked or the cluster algorithms were reconfigured to match the correlation.
Google added that it violates its policy to show personalized ads for these sensitive categories.
Nominations are open for 2021 CNBC disruptor 50, a list of private start-ups using breakthrough technology to become the next generation of large public companies. To send until Friday, Feb. 12 at 3:00 p.m. EST.