The Drum Awards for Marketing - Extended Deadline

-d -h -min -sec

Tech Machine Learning Audience Targeting

Why it’s time to finally abandon audiences

By Andrew Ahn, senior director of product management

Moloco

|

Promoted article

September 12, 2022 | 6 min read

Sponsored by:

What's this?

Sponsored content is created for and in partnership with an advertiser and produced by the Drum Studios team.

Find out more

For as long as digital advertising has been in existence, advertisers and agencies have relied on audience targeting for their campaigns. While this tried-and-true tactic has been effective for so long, its efficacy is on the wane in today’s privacy-first era. Instead of relying on pre-set expectations around audiences, advertisers today are much better off using the latest in machine learning to find their best users.

1 yellow pawn surrounded by black pawns

If audience tracking is out of the question, then how can advertisers and agencies reach and acquire high-quality consumers?

Why audience targeting no longer works

It’s easy to see the inherent logic in targeting based on audiences. For example, who else would a diaper brand want to target besides new parents?

The biggest problem with audience-based targeting now is that the data underpinning this strategy is becoming either unavailable or unpalatable. On the technology side of the equation, Apple’s AppTrackingTransparency framework has largely removed the Identifier for Advertisers (IDFAs) from the mobile in-app ecosystem, meaning that it is now much more difficult to directly connect one user to one device (at least in the all-important iOS ecosystem). And with Google beginning testing of its Privacy Sandbox for Android, it’s only a matter of time before a similar dynamic sets in with the other major mobile operating system.

This is already the reality in the world of browsers. Mozilla’s Firefox and Apple’s Safari browsers have already banned third-party cookies, and Google will follow suit eventually on Chrome.

Of course, an advertiser’s first-party data can be leveraged for targeting purposes, but getting enough and then making it usable at scale is no small feat. It will likely be a few more years at least before any of the most notable digital identity solutions are widely used – and then sharing data securely and privately is another matter entirely.

Plus, the legal landscape will make any one-to-one targeting less feasible. GDPR in Europe, Brazil’s LGPD and the California Consumer Privacy Act have already upended data sharing and tracking practices, and plenty more laws and regulations are bound to spring forth in the coming years.

And on top of it all, most consumers simply don’t want to be tracked online by advertisers. 2022 research from Permutive and The Harris Poll found that around three-fourths of consumers in the UK and the US have serious concerns about tracking and how data about them is tracked and used.

This then begs the question: if audience tracking is out of the question, then how can advertisers and agencies reach and acquire high-quality consumers? This is where machine learning comes in.

Why machine learning is better than audiences for digital advertising

The benefit of well-trained, high-quality machine learning models is that they enable advertisers to find their best audiences without needing personally-identifiable information about them. Machine learning is capable of ingesting a wide spectrum of behavioral and contextual data – think time of day, where the ad is being seen, how someone engages with an ad – to make smart predictions.

The machine learning-driven advertising approach also helps preserve privacy as data about users doesn’t get exposed explicitly to advertisers. Let’s go back to our earlier example. Instead of just targeting new parents, a machine learning-based digital advertising solution may find that certain other variables (such as location or device type) can be better predictors of whether someone is likely to buy a particular diaper brand, and this can be achieved without tagging and tracking individual users with certain attributes. After all, people besides new parents buy diapers, and not all new parents are looking to buy diapers (or a different brand).

The benefit of machine learning is that it can leverage data that will continue to be available and that consumers are OK sharing. In today’s privacy-first world, it’s the best path forward.

Of course, not all machine learning is the same. Just about every adtech company says it has machine learning, but only some companies have the resources and track record to back up their claim. But the ones that have invested in quality machine learning over the past few years are the ones that will succeed in the years to come – and are the ones to leverage to future-proof your digital advertising.

Suggested newsletters for you

Daily Briefing

Daily

Catch up on the most important stories of the day, curated by our editorial team.

Ads of the Week

Wednesday

See the best ads of the last week - all in one place.

The Drum Insider

Once a month

Learn how to pitch to our editors and get published on The Drum.

Tech Machine Learning Audience Targeting

Content by The Drum Network member:

Moloco

Machine learning for performance marketing campaigns

Find out more

More from Tech

View all

Trending

Industry insights

View all
Add your own content +