Meta Shares Notes on the Development of its AI-Based Ad Targeting Systems

Looking to get a better understanding of Meta’s evolving, AI-driven ad tools, and how machine learning is helping to drive better results for its ad partners?

This week Meta held a roundtable discussion on AI strategy in marketing, and where advertisers are focusing their efforts in 2024. The session included Meta ad execs, as well as an interview with Cody Plofker, the CMO of Jones Road Beauty, to provide specific, actionable insights into the development of Meta’s ad tools, and the trends that it’s seeing on this front.

Among the key notes:

Meta’s AI content generation features for advertisers, including text variation and image expansion, have seen strong initial adoption, and early tests have shown promising results in driving better performance for brand partners
Click to message ads continue to grow, with more brands looking to lean into messaging as a key connective tool, in line with broader engagement trends
In terms of AI content recommendations, 40% of the content that people see on Instagram now comes via AI recommendations. Meta also notes that over the past year, its AI recommendations have driven a 7% increase in time spent on Facebook, and a 6% increase in the same on IG
Video content remains the key to this. Daily watch time across all video types grew by more than 25% year-over-year in Q4 2023, while people are now resharing Reels 3.5 billion times every single day

In terms of specific ad trends, Meta also noted that more brands are driving better results by leaning into its AI targeting tools, like Advantage campaigns, and moving away from interest targeting.

Meta says that there has been some hesitance from brands on this, as it can feel like you’re giving up control by not selecting specific audiences to reach. But by using Meta’s AI-powered targeting systems, which rely on probabilistic modeling, brands are then able to commit more time to their creative, and in combination, this has led to better results overall.

Meta further notes that creative is the most important aspect of the ad process, impacting over half of the decisions that its ad auctions make.

AI has become a key focus for Meta, both externally and internally, as the company works to implement new systems that can help to improve performance and experience.

And as noted, Meta’s Advantage automated campaigns, in particular, have helped brand partners generate better results from their campaigns. That’s helped Meta’s broader ad business recover, following the impacts of Apple’s iOS 14 update, which saw new prompts pushed to all iOS users asking them if they want to allow Meta’s apps to use their data. And given the many negative reports around how Meta has used data in the past, many did indeed choose to opt-out, which, at one stage, Meta estimated would cost it around $10 billion in annual losses.

But in 2023, Meta’s ad business was back on track, driving an overall 25% increase in revenue, while ad impressions were up 28%, as the average price per ad decreased by 9%.

AI has played a key role in this recovery, and it’s worth considering Meta’s notes around increased performance based on ad targeting, and moving away from interest focus as a key driver of your strategy.

It could feel risky, but it may be worth experimenting with Meta’s more automated processes in ad creation and delivery.