Meta’s recent announcement that its platform will be able to fully automate ad-campaign creation and media duties has the advertising industry both scared and skeptical.  

“Our goal is to make it so that any business can basically tell us what objective they’re trying to achieve, like selling something or getting a new customer, and how much they’re willing to pay for each result, and then we just do the rest,” Zuckerberg said on Meta’s Q1 earnings call.  

Mark Zuckerberg’s planned “redefinition of advertising” is anything but surprising. It’s a leap, but an expected one, from products like Advantage+ and Google’s PMax that have been designed to operate in a “black box” capacity like this for the last several years. We’ve already had to live through the painful development of these products’ early days.  

What we like about some of this automation is that it takes some of the guess work and grunt work out of our processes, and we can shift our human time and attention into higher-value places. Also, we believe it’s a great way for smaller, local advertisers to get into advertising more easily. But there are some considerations. 

AI-driven creative work can sometimes suck. Think about early PMax video creative that was translated from statics. Rough, right? Machine learning and AI will struggle to be accountable for carrying the “soul” of what a brand stands for. There are cultural nuances, audience nuances, and a brand’s POV that are central to some campaigns.  

Initial success for brands that let Meta build their creative entirely will depend on what they need it to do. If a brand needs to tell more people that it is offering $3 off Smoothies next week, AI can likely create that and various versions appropriately. If the brand wants to reposition itself or needs to be introduced to a new audience entirely, marketers will want to be very careful about the inputs and outputs. 

Think about the waste. Fundamentally, cross-network products are a way for brands to monetize distressed inventory, and that becomes a cost consideration. For example, if you really only want to be on Instagram, can Meta make it lucrative enough for you to opt in to their placements more broadly? Both Meta and Google have been through some serious price editing behind the scenes, which has made these products substantially more attractive over time.  

What about measurement? Meta can only see and optimize what happens inside its “walls,” which might be good enough for many small advertisers that don’t have extensive budgets across channels. But for the mid-market and above, third-party measurement has never been more important. Let Meta optimize itself to the be best it can be, but brands must still hold a singular, objective measuring stick. Understanding incrementality has been the focus for us in these AI-led and optimized products.  

The existential question becomes if Meta (and Google etc.) have every advertiser’s budget, goals, bidding control, audience control, and creative control, how will they manage the inherent competitive conflicts? Specifically, if the perfectly primed swimsuit consumer is about to receive an impression, how will they determine which brand’s ad gets served? 

Influencers will continue to shine. As the market has been flooded with influencers of all sizes and niches, it can feel overwhelming from a consumer’s point of view. But we expect this trend to stay, especially in a heavy machine-generated future. Their content will feel fresher than ever, relatively speaking. It’ll continue to be a way for brands to get in front of audiences that’s not so “manufactured.”  

The best agencies will adapt. The job of agencies is to help clients navigate the entire ecosystem of advertising – where and how should your brand show up to maximize your investment? There are thousands of choices you can make within that scenario. It was never about our excellence in creating Meta carousels in the first place.