Measured’s Post

Meta announced last week that it’s working on “AI-powered optimization updates to its ads system so advertisers can customize business objectives and measure incrementality.” This has obviously received a lot of attention, I wanted to weigh in with what this means for marketers who are increasingly focused on optimizing their media mix with incrementality. First let’s take a look at what’s happening with this change. These measurement vendors are sending user-level attributed orders back to Meta. Meta then uses that signal to optimize the algorithm, essentially resulting in “1-day click on steroids.” This may optimize these campaigns to even more non-incremental clicks from high-intent users. As far as the separate enhancements to optimize media to incremental orders, that process will be entirely determined by what Meta deems incremental and it's not yet clear how they will do that. So what should marketers know? ✅ Optimizing Meta's algorithm to your multi-touch attribution could be outright dangerous and damaging to your business by training your campaigns to chase non-incremental user level signal. ✅ The option to optimize your campaigns to incrementality is very interesting, but also currently a black box. We're encouraging brands to experiment with it. ✅ Whatever you do, make sure you independently validate whether these changes are providing incremental lift via a test and control methodology. Of course your MTA model will tell you your ROAS is improving, but is that actually driving real business lift? Measured is in conversations with Meta to explore how brands can use incrementality to directly impact Meta's optimization algorithm. We're thrilled that Meta is opening up these opportunities for advertisers to ensure that their campaigns are delivering the most business impact possible. Looking forward this is very exciting and potentially a huge opportunity, for now we encourage brands to explore these opportunities but also to ensure they have the tools to independently validate their impact. #Analytics #MediaOptimization #PerformanceMarketing