Editor's Note: This is the fourth installment of the Living With Data series exploring how our online data is tracked, collected and used. Do you have questions about how your personal data is being used? Curious to learn more about your daily encounters with algorithms? Email the Decoder at email@example.com or submit your question via the form here. Screen shots and links are helpful clues!
Cookies work like the digital crumbs of your experience. They follow you around, leaving a trail of your online activity like Hansel and Gretel. Advertisers guess what kind of person you are from the kinds of cookie crumbs you leave and where you leave them. But there is a lot of guessing involved with cookies. Once scooped up, crumbs are hard to tell apart.
And we’ve developed lots of ways of making cookies less effective, with browser settings and plug-ins that block and clear cookies. The cookie is failing the advertising industry.
Last week Facebook introduced Atlas, its “people-based marketing” answer to the failures of the cookie. Instead of cookies, Facebook Atlas pulls all your activity across devices — Web, mobile, tablet — together with one unique, anonymous ID, which works more like a fingerprint no matter where you touch than like a mess of crumbs.
Atlas solves a problem that marketers have had for a long time: It’s hard to know what ads are effective. But what changes for us as consumers when marketing is targeted to us as identified individuals, not just via cookies that stand in for our behavior profiles?
On the Web and mobile, Facebook Atlas combines cookies with the unique Facebook ID to track individual users’ exposure to advertising across sites and apps. As long as you are logged in to Facebook on your device, Atlas tracks activity even in apps that don’t use a Facebook login.
The Atlas system will be used to report back to marketers about the efficacy of the ads they buy. Atlas does this by noting the ads that Facebook users see. The marketer then goes to Atlas with a list of details about people they know bought their product. Atlas matches those two lists together to report anonymously how Facebook ads were involved in influencing those purchases.
Facebook promises that this process is anonymous, and marketers and advertisers won’t get access to the list of names, aside from the purchasing information that they already have. No human sees that your individual purchase is tied to the successful ads. And for now, the resolution of your purchase to the ads doesn’t necessarily tie back to the new ads you see; it’s just reporting back to the advertisers about how successful their targeted ads were overall in a population.
Atlas is Facebook’s approach to an emerging industry tactic in data brokering called onboarding, which attempts to match up data that originates in offline places — like purchasing history, direct mail responses and demographically important details — with the rest of our online activity and behavior profiles.
Onboarding blurs the distinctions between our online and offline lives. Our data is commingled, and companies benefit from pooling a more complete view of our experience.
There are some problems with Atlas, namely that it presumes a one-to-one relationship with our devices. It works best because we tend to stay logged in to our accounts. It doesn’t take into account a family-owned device, on which someone might forget to log out. And Facebook representatives confirmed that there is no relationship between Atlas ID and Facebook’s recent troubles enforcing its real-name policies.
Onboarding is just one example of the ways that our data profiles are evolving. And it’s important to explore how things like this work so that we can expand our understanding of not only what is technically possible but also what should be possible as it matches up with our social norms.
The consolidated Atlas view of us as users is perhaps more accessible and understandable than previous ad technologies. One consistent Facebook ID, a digital fingerprint, makes more sense to us as individuals than a million cookies if the goal is “people-based marketing.” Facebook is a complete experience, after all, across all our devices.
Facebook still wants to know us as a singular individual. They adjusted their real name policies, but they still want to match up your name to the one you use in "real life," whether that's a nickname or drag stage persona. One name, one ID to rule them all.
But we are still different data profiles across different platforms. Facebook is building out its walled garden. It isn’t selling our data directly to advertisers or marketers. It is protecting its precious data assets within the Atlas system. So that adds to just one of the many different siloed versions of us across the Web. Google has a version of me, and Facebook, Acxiom and the NSA all are developing and expanding their unique views.
So onboarding is making it possible for some data-driven companies to develop a consolidated and consistent view of consumers. But these companies are maintaining asymmetric control over these profiles, both from us as individuals and from the rest of the ecosystem.
What would you do if you could see a profile that matched up your credit card purchases with a report on the ads you saw? Would that change your spending behavior? Would it change your research process?
We ought to be able to see what these profiles say about us so we can understand their influence over us and so we can correct them when they inevitably get us wrong.
The Living With Data series explores how our data is tracked, collected and used online. Do you have any questions about how your personal data is being used? Curious to learn more about your daily encounters with algorithms? Email the Decoder at firstname.lastname@example.org, tweet me @smwat, or submit your issue or question via the form here. Screen shots and links are helpful clues!