The third-party cookie is on its way out, but the mirage of the universal ID persists. Marketers should be sceptical about the hype surrounding dozens of single-point identity solutions. Just as important, they should think about the advantages of a distributed identity model.

Why there’s no such thing as a universal ID

A universal ID needs to cover a lot of ground from the open web to walled gardens (such as Facebook, Google and Amazon) to mobile and connected TV. But universal ID isn’t just a technological challenge, it’s also a non-starter from a business standpoint because nobody wants to give up control of those customers. Even if you do manage to surpass those hurdles, how can a single-point solution manage both general (GDPR, CCPA, etc) and industry specific privacy laws (HIPAA, for example) legislatively line up as roadblocks to universality, identity or proxies to identity?  

According to an IAB study, fewer than 5% of consumers bother with data privacy settings. But is that because they don’t care (hard to believe given the public backlash), or is it because privacy is too personal and nuanced to fit neatly into a dashboard?

And don’t forget consumers – especially the vocal minority.  Any practical application of a common identity solution is damned to some form of a circular firing squad. The walled gardens have too much to gain from guarding their installed base, and there are enough legal billable hours in the privacy crusade to keep every constituent in the debate alive and well-fed for a good long time.

Identity resolution isn’t a noun, it actually is a verb

Currently our industry tends to view identity resolution as a product category. In reality, it is more likely a point solution that solves one problem, but also impacts the full stack of marketing activity. Being able to help marketers connect behaviors across all platforms, audiences and activity is critical to measurement – and, by extension, critical to managing cost. The universality of the user ID exposes that ID to being reused. This is why it occupies so much attention.

A more realistic approach is to think of identity resolution as the verb that connects other categories. Rather than being a category in and of itself, maybe marketers should view identity resolution as a repeatable process that turns first-party data into a destination-specific outcome that can’t be used elsewhere. What if each application of identity was in fact unique to its application? Under such a distributed model, the focus isn’t on transacting against an ID, it’s on the data owner (and only the data owner) to manage the data in a proper, identity-safe manner.

Distributed identity isn’t just achievable, it’s better

Sure, a universal ID that perfectly aligns with the plumbing of the current identity model would be easy to explain, and even easier to implement. Unfortunately, there is far too much leverage in maintaining identity dislocation between key programmatic, CTV and social providers.

Marketers have no option but to embrace a distributed identity model for two primary reasons. 

First, a universal ID is a big target, not just for hackers, but also for public opinion. In a universal model, brands that work hard to comply with the law and understand the nuances of consumer privacy sentiment will still be judged by the worst-performing members of the ecosystem. After all, if the problem is perceived to be the universal ID – it is by definition a target and everyone that uses it shares the blame. 

Second, a distributed identity model puts the identity asset firmly within the control of the data owner. While a distributed identity model is protecting that asset, it is also making attribution easier and safer. The only participant in a distributed identity is the data owner.