The loss of addressability within digital media will continue as third-party cookies and mobile advertising IDs (MAIDs) vanish across the landscape. But that doesn’t have to mean a return to spray-and-pray ad tactics.
As necessary consumer privacy protections and policies go into effect to safeguard personally identifiable information (PII), technology advances are yet again stepping in and preserving the ability to customize messages to targeted audiences and cohorts — with precision and at scale.
Let’s take a look at how advances in predictive modeling and machine learning can be leveraged to extend panels of consented consumer data into highly targeted cohorts.
Overcoming challenges facing today’s targeting methodologies
There’s been no shortage of solutions coming to the market that purport to solve for targeting in a privacy-first world. But the reality is, many of these solutions lack the scale needed to serve as an adequate replacement for the addressability being lost in this new era of digital advertising.
Predictive modeling, which can leverage a pool of consented user data to identify patterns and predict audience characteristics and behaviors among users, offers a privacy-compliant path to addressability 2.0. But not all the solutions that purport to deliver on this possibility truly can.
Most predictive audience solutions aren’t being fed by sufficient seed data that’s representative of the overall U.S. population. If there’s not enough seed data going into a solution or that data isn’t high quality, it simply can’t be extrapolated to reliably predict who might fall into a specific cohort or persona.
At the same time, there are a number of walled gardens in the market that are vying for ad dollars based on the scale of their logged-in user base. However, in selecting partners, marketers should consider the type of data that these platforms are using to target audiences.
For example, while some of today’s retail media networks might have access to large user bases, they are often limited in their holistic understanding of those users. While these networks might be able to target ads based on purchase history, they often have few insights into who their audiences are from a demographic or broader behavioral perspective — beyond what may be inferred from a subset of purchases. This thereby limits the ability of advertisers to target in a personalized way.
Harnessing the benefits of carrier-level engagement data
Within the modern privacy-first advertising landscape, carrier-level mobile and in-app engagement data offer benefits to advertisers that can’t be found elsewhere. These data sets, based on app ownership and app usage across highly representative populations, can be transformed via predictive modeling and machine learning into highly targeted advertising executions at scale.
Carrier-level data is one of the only sources of insights that can provide the addressability today’s advertisers seek, while avoiding the challenges related to scale, fidelity and complete audience understanding that face so many of the other audience solutions in the market today.
By applying predictive modeling to comprehensive app insights — data around the apps people use, how often, how long and when they use them — mobile carriers are able to understand which users are likely to fall into certain desirable cohorts with a privacy-forward approach. At T-Mobile Advertising Solutions, we’re able to create these cohorts at scale and enable marketers to engage with them across multiple channels and screens via predictive models — without the reliance on addressable identifiers across all ad opportunities.
Predictive audience modeling that’s built on a strong, consented, deterministic seed data set represents the true path forward for advertisers in a landscape where MAIDs and third-party cookies are going by the wayside. If your advertising partners aren’t delivering robust, scalable and privacy-compliant addressability solutions in a future-proof way, it’s time to be looking for ones that can.
Campaignliveus
