How Lookalike Audiences Can Support Your Marketing Efforts in the Post-IDFA World

May 18, 2021

With Apple’s recent roll-out of IOS 14.5, the digital marketing industry has finally faced what it fretted for many months – the depreciation of Apple’s identifier for advertisers (IDFA).

The advertising industry has long relied on the IDFA, a.k.a. identifier for advertisers, which is a type of mobile advertising ID used to gain anonymous insights into user activity. However, as Apple has shifted its focus to consumer privacy, its new App Tracking Transparency feature allows users to opt-out of tracking, preventing advertisers from collecting information that is essential for personalized ad targeting. 

The main question advertisers and marketers now have to answer is how they will pivot their practices given that many of them depended primarily on the IDFA to target consumers on mobile devices. We know that consumers are still interested in being offered the same level of ad personalization, so companies need to engage with alternative solutions.

One such solution is lookalike audience modeling powered by machine learning. This solution does not rely on a single identifier, so it’s more ‘future-proof’ and can help companies reach the right audiences at the scale they seek. 

Lookalike audience models are based on companies’ first-party data, which includes information about their existing customers. Based on this data, machine learning algorithms assemble a highly accurate, scalable model of ideal target consumers who share the same characteristics as the company’s current customers. Once this model is created, it is tested for accuracy using the company’s original list of customers to confirm that the model would have selected them based on their demographic and behavioral attributes. 

Not only is a look-a-like audience an optimized and effective marketing tool to reach consumers in a targeted and personalized manner, but it also most importantly is privacy compliant, making it the ideal solution for marketers looking to break free from their dependence on big tech, especially as identifiers and cookies are phased out. 

The important thing to note, however, is that lookalike audience models are only as good as the data on which they are based. So, while many AdTech companies are now offering these solutions, not all lookalike audiences are created equal. To make the most of their marketing dollars, companies need to make sure that the data they use is accurate and up to date. Without solid data, companies risk diminishing their ROAS by targeting non-ideal customers.

While many marketers are understandably concerned about the depreciation of cookies and the IDFA, our industry needs to look at this shift as an opportunity to innovate through independent solutions like lookalike audience modeling. This is the future of digital marketing. And as our industry practices continue to get disrupted with other anticipated identifier clampdowns and consumer data policy changes, it’s imperative that we recognize that privacy and personalized marketing do not have to be exclusive of one another. Lookalike modeling shows us how they can complement each other perfectly to ensure that businesses can expand their reach and grow.

David Finkelstein is an internet pioneer, tech entrepreneur and founder of numerous internet companies dating back to the earliest days of the internet in 1994. He currently serves as the Co-Founder and CEO of BDEX, an Inc 5000 company, the first and the largest consumer data exchange platform in the U.S.