On 3rd March this year, Google made quite a splash when it announced plans to chart a new course with the aim of creating “a new privacy-first web.” Why change course when Google itself profited handsomely from the whole consumer-data economy over the past 20 years?
Well, the company read the writing on the wall. Consumers are unhappy with many aspects of the data collection industry, and their displeasure is well past the tipping point to act. An alarming 72% of people say they believe that most of what they do online is tracked by advertisers, technology firms or other companies, and 81% say they’re likely to see more risks than benefits come out of that data snooping.
We knew that Google was getting serious about privacy. Last year the company announced that by 2022, Chrome will stop using third-party cookies. Given that the vast majority of users (about 75%) use Chrome, the move will bring data-driven advertising as we know it to a screeching halt. Ditto for the way much of the world does attribution and measurement.
Of course, Firefox and Apple had already eliminated cookie tracking in their browsers, but they don’t have the same level reach, and therefore the impact, on the market as Google.
Many marketers were counting on Google to come up with an alternative to the cookie. After all, companies all over the world invested in data management platforms (DMPs), third-party data sets, and even DSP licenses to put all this data they collected to good use. But in the March, Google dashed those hopes by announcing it would not, in fact, build an alternate identifier to track individuals, and even if someone did, Google wouldn’t use them in their products. Google products, they said, “will be powered by privacy-preserving APIs which prevent individual tracking while still delivering results for advertisers and publishers.”
Instead of cookies, Google will leverage a federated learning of cohorts (FLoCs) API, an idea the company proposed as part of its Chrome Privacy Sandbox last year. The idea is that an unsupervised machine learning model will group people together based on their interests – aka cohorts – based on their browsing behavior. To preserve privacy, cohorts can be targeted, but not individual users.
Chetna Bindra, also of Google, says that FloCs are nearly as effective as third-party cookie targeting. Citing research conducted by Google’s Ads Teams, she says that advertisers “can expect to see at least 95% of the conversions per dollar spent when compared to cookie-based advertising.”
This raises a lot of thorny issues for marketers. Cookies aren’t just used for targeting; marketers use them to measure campaign performance and to evaluate the efficacy of channels, tactics and partners via attribution. If you no longer have that data at your disposal, how do you know where to place your media spend so that you get the most bang for your buck?
Clean rooms
There’s one solution to the measurement and attribution challenge that’s received a lot of attention – clean rooms. Clean rooms are offered by LiveRamp, InfoSum, Snowflake, Habu and many other companies.
Sometimes called “walled gardens,” clean rooms are secure environments in which data is anonymized and processed in some manner for a multitude of purposes, including measurement and attribution.
Let’s say you’re a brand and you ran a two month campaign on the New York Times. Was it successful? As of 2022 you will no longer be able to rely on cookies to track users who saw your ad on the New York Times, visited your site and then converted. But a clean room will allow both you and the New York Times to match users. In this scenario, the New York Times allows the clean room to see the list of its users who saw your ad, and you allow the clean room to see the list of users who converted. The delta allows you to assess the efficacy of the campaign.
Clean rooms are touted as “privacy-first,” because you don’t get to see the New York Time’s data, and the publisher doesn’t get to see yours. Companies like InfoSum refer to this as the non-movement of data.
While clean rooms can be very privacy compliant, GDPR grants consumers some rates as to how their data is processed if you plan to use a clean room for marketing purposes. Let’s say you’re a marketer for a brand that specializes shirts and tops for women and you want to know if it makes sense to enter into a joint marketing arrangement with a brand that sells women’s shoes. A clean room can help you identify whether or not you have a lot of customers in common, and even if those common customers tend to be high spenders. If you see that there is significant overlap, a joint campaign may make a lot of sense.
If you intend to go the next step, however, and send ads for your shirts to shoe-customers of your partner, you may need to obtain consent. GDPR regulates data processing, and guarantees EU citizens the right to be informed of how their data will be used, in a “concise, transparent, intelligible and easily accessible form, using clear and plain language.”
GDPR gives consumers the right to opt out of automated decision making and profiling. In other words, Sally Jonas may not want the clean room algorithms to profile her as a likely candidate for your shirts based on her shoe-buying history.
A new future
We are still very much in the early days of a cookie-free world, but the digital ad-tech sector has been hard at work coming up with solutions to allow brands reach and engage consumers as they go about their digital lives in ways that respect their privacy. I doubt that there will be a single approach going forward, and the right solution will depend very much on the brand’s customers, goals, and a host of other factors.
The right partnership
Working with an offshore ad operations provider that has the resources and knowhow to navigate the cookie-free world makes good business sense. Get in touch today.
Related Post
If you enjoyed reading this then please explore our other articles below
