• by Cory Treffiletti , Featured Contributor, April 24, 2024

Google launched Performance Max, its AI ads optimization platform, back in November of 2021.  Since that time, the world has advanced exponentially in terms of the opportunities to use AI in advertising. So I’m curious if the platform has taken hold the way Google hoped.

I turned over some stones to try and find out.  At a high level, it would appear people are interested in Performance Max, and many advertisers are very happy with the solution.  It seems to deliver on its promise of performance and the ability to unlock new audiences across the entire landscape of Google ad inventory.

Like any advertising technology, it has some areas of concern, but nothing is ever perfect.

I’m not here to point a finger, but I am interested to call out how brilliant a move this solution was for Google the publisher because it more effectively monetizes what I would consider its “b-grade” inventory.

Google’s most important ad solution is search, likely followed by YouTube and then by the broader display network.  It has Gmail and Maps, plus the Discovery platform, but all of these are lower cost, with almost infinite inventory available, and less likely to generate clicks or engagement.  These are often branding vehicles, or a way to drive reach and frequency of exposure, hopefully increasing the effectiveness of other campaign elements.  I know there are ways to make each one work, but ultimately, it’s all about search, YouTube and display.

Performance Max is a brilliant way to position all that inventory as “optimized by AI” and enables Google to drive budget to these channels without having to specifically call it out.  This aligns with one of the only real major gripes that media buyers seem to have about Pmax -- that it is not transparent, and you don’t really know when and where your ads will run.  This translates to a lack of control, and we all know that advertisers hate to lack control.  Big brand advertisers probably view this as a hurdle, while performance-based advertisers are OK with that fact and lean in on the cost per action they are driving.

This boils down to a trend I see rearing again in digital advertising:  transparency.  The topic keeps coming up, and if you look at the use of AI and tools to optimize delivery, optimize creative, and more, advertisers are going to be forced to make choices.  What are they most interested in --  control of the brand placements or control of the performance?  You may not easily have both.

In my opinion, this creates a strategic gap between brand and performance advertising, or at least widens the gap that was already there.  Agency planners and brand managers are going to have to work together to determine where AI can be used in their campaigns, and to what degree they will allow AI to optimize without full transparency on when and where their ads will run.  Essentially, is it more important to know where or to what effect your ads ran? 

This becomes very interesting as cookies degrade/expire and targeting becomes more difficult.  Are you willing to give the keys to your kingdom to one vendor like Google, because it has the tools, both manual and automated, the inventory -- and ultimately, the data? The same question can be asked about Amazon and the retail media networks.

Once again, we have lots of questions to ask and lots of different ways to answer them.