It only gives you that clarity if you're getting the data unfiltered, though. As I read the article, the problem P&G was running into is that it was outsourcing the placement of its ads to third parties, meaning they didn't get access to the raw data on how those ads performed -- they only got whatever data the third party chose to share with them, packaged up however the third party chose to package it. That kind of relationship opens up all sorts of opportunities for the third party to fudge the numbers to serve their own interests, which is what it sounds like P&G decided was happening.