Yeah, not surprised at all here. What is surprising is that they actually found this out and acknowledged it. Used to work in the performance marketing team fo

Yeah, not surprised at all here. What is surprising is that they actually found this out and acknowledged it. Used to work in the performance marketing team for one of the top 10 brands as an analyst/data scientist, and the reality is that the company politics and individuals' career ambitions top the science. Most of the time the work that gets visibility is spent on "creating stories around data for the leadership": marketers do not understand at all the data they are fed to by the advertising platforms. Also, running controlled A/B studies tended to show between 0 and 10% (this was rare, closer to 0% more common) causal impact on campaings making none of the marketing campaigns profitable in the sense that was used for performance evaluation. There is one key visicious loop keeping the game going: 1. The marketers want to show great numbers back to the leadership (internal storytelling). Their employment/bonuses depend on the number reported by the ad platform. 2. The ad platforms want more of your ad spend which can be achieved by showing better ROI in their platform. 3. Both the marketers and the ad platforms want to show as good numbers as possible, leading to a situation that nobody is incentiviced to measure true performance / correctly measuring performance. None of the A/B tests nor analyses about ad fraud went beyond the marketing director. One cool case was analyzing Google's DoubleClick ad logs for a bunch of campaigns that showed there were few cookies that received a significant % portion of the total ad spend - meaning that few "insividuals" received tens or hundreds of thousands of ad impressions. Nobody cared and this still goes on. It is such a toxic environment to work in I ended up changing career a bit.