When it comes to data-driven advertising, measurement will dictate the future of identity

And if the industry wants to continue to grow in 2023, it will need to efficiently reach and measure audiences outside of third-party cookies with fewer identifiers.

Digital advertising has made huge strides to operate without third-party cookies, but I’d be lying if I were to say we just have a few minor kinks to work out before we’re in the clear. To evolve into a sustainable ecosystem outside of third-party cookies, there’s a major obstacle we need to tackle if we truly want to move towards a privacy-centric open and free web: identity measurement.

It's quite easy and inexpensive to target a user on Safari today; however, since buyers are less likely to buy what they can’t measure, programmatic’s free market system discounts premium Safari users. Until there’s an acceptable identifier that can measure outcomes, Safari users will remain on programmatic’s sale rack.

Programmatic advertising has been an efficient way to reach audiences at scale and as we race to rebuild without third-party cookies – and this allows us to curb the downsides of the cookie such as overstepping consumer privacy, fraud risk, layers of fees and latency.

Here’s what needs to happen to access the entire open web.

Rethinking identity

Today, we often define identity as having the ability to identify an individual, device or browser using a variety of online signals. But if the goal is to move to a privacy-centric future, we need to rethink the meaning of identity as it relates to digital advertising.

The next iteration of identity should entail replicable signals that allow marketers to compare outcomes across their digital media buys. Alternative identifiers need to capture and pass privacy-centric signals that can be directly inserted or inferred by programmatic demand; once that occurs we’ll see more support from buyers and, in turn, an increase in publisher monetization.

Since enhanced consumer privacy and choice is the supposed culmination of a world without third-party cookies, these principles need to be at the forefront of any cookie alternative framework and how we measure it. We have to be mindful of the consumer and provide simple and clear controls to empower the consumer to make their own choices about their data. Luckily, most consumers can appreciate and understand a fair value exchange to access content, and will therefore often be happy to share their information.

Supply and identity partners need to align with how demand partners measure. Creating a set of parameters on what performance standards we need to capture will curb the need for custom development work and encourage supply chain efficiency. Making measurement a priority now allows for testing against the third-party cookie and a better understanding of performance.

Creating sustainable, cookieless measurement methods

To prepare, the industry will need to adopt new approaches to measurement. Some viable options include the following:

Record authenticated conversions: A one-to-one measurement is typically anchored by an authenticated event on a publisher page – like email registration – by opted-in users. Though authenticated conversations are extremely accurate, this kind of solution tends to be very low-scale and will require a boost from a probabilistic solution.

Modeled conversions: Most media folks will be familiar with buy-side audience extensions and lookalike modeling to expand audience scale using third-party cookies. To broaden the impact of authenticated conversions, such an approach will have to act as the seed for probabilistic conversion expansion. A modeled expansion will use signals of a known authenticated customer journey and infer a conversion without user-level identifiers.

Develop a bring-your-own-ID foundation: Think of this as the next iteration of identity resolution. Buyers will rely on trusted partners to convert and translate privacy-safe signals, events and other identifiers to their ID.

Eyeing our privacy-first future

Buyers can choose which methodology provides the best value based on their preferred strategy. Any sound model requires plenty of incoming data, but it’s important to ensure that it's being fed from different sources since there are many paths to get to the same solution. Let’s be real: completely excluding probabilistic solutions won’t be sustainable and will be up to the buyer’s discretion if they want to expand their coverage with a known signal or use it to create their own.

As the old saying goes, there’s no better time to start than now. While third-party cookies remain on approximately 50% of browsers, running modeled measurements using alternative identifiers in tandem will allow advertisers to gain a better understanding of this new paradigm of measurement.

The sooner we can establish parameters that account for privacy, the sooner Safari and other forgotten golden eggs of the open web become digestible.

Paul Bell is president at 33Across. To read more from The Drum’s latest Deep Dive, where we’ll be demystifying data & privacy for marketers in 2023, head over to our special hub.