Agencies Concerned They Are Giving Too Much Control to Meta’s AI and Algorithms

Meta, the parent company of Facebook and Instagram, has invested in artificial intelligence and machine learning to enhance its advertising technology, which has received mixed reviews from marketers. 

The algorithms test and select the most effective ads, with the option to modify text and images automatically. Tinuiti data shows that brands that tested Advantage+ Shopping Campaigns (ASC) in October 2022 experienced significant cost savings, with a minimum of 20% reduction in costs per thousand impressions (CPMs) and a 15% decrease in costs per click.

Advantage+ has also resulted in improved advertising campaign performance and lucrative returns, with iProspect clients reportedly earning $7 for every $1 spent on website advertising campaigns using the service. According to a senior staffer, Meta has allocated more resources to upgrade its AI technology to address the effects of Apple’s privacy changes than to CEO Mark Zuckerberg’s metaverse goals.

However, some marketers are worried that Meta’s algorithms are taking too much control over their campaigns, leading to several companies withdrawing from Advantage+. According to The Financial Times, multiple companies have cited the need to relinquish too much control as their reason for withdrawal. For instance, a UK-based games company withdrew from the program after it recommended an advertisement that attracted inflammatory and hateful comments, which the company blamed on the tool’s lack of human emotion and common sense. Similarly, a UK-based retailer admitted that while the CPM prices were significantly better, the company had to cede too much control over their campaigns.

After Apple implemented its privacy changes in April 2021, Meta experienced a loss of approximately $10 billion in revenue within nine months. To adapt to the changes, Meta now relies more on first-party data, such as a user’s interactions with brands, such as liking, commenting, tagging, or mentioning them.

According to senior analyst Max Willems, “Advertisers would do well to remember that Meta has a history of inflating metrics for its own benefit.” Willems advises advertisers and agencies to approach Meta’s advertising platforms with caution.

Our analysis suggests that while agencies currently see ASC as a valuable addition to their advertising tactics, the platform’s highly automated and opaque nature could have a significant impact on agencies’ roles in Meta’s advertising business. As advertisers become increasingly reliant on ASC, they may find themselves needing agencies less.

Meta’s extensive agency network is vital to its ability to persuade brands to invest in its advertising services. However, leaning heavily on AI and automation could disrupt this alliance and jeopardize the company’s relationship with agencies, which it has long relied on for support.

Nonetheless, Meta could argue that by using ASC, advertisers can save money on asset development and deployment and then redirect those savings into ad buys, resulting in ad revenue growth. This argument is particularly advantageous for Meta since the company stands to benefit from that ad revenue growth.