People always like to like immunize whatever they are trying to do and data. They spend a lot on Markov change and heat maps what what not you can throw at that problem just to get that point across. If people don't trust you based on your statement then maybe there's a different problem behind it, he says. "I'm a hundred percent sure there is somebody smarter in the room who's going to call me on it" John Defterios: I will absolutely freeze up did I say we should do this thing? He was talking about his experience with an ad agency which ran a B test on a campaign but presented their results as statistics.
Do analysts make things more complicated than they need to be, or is the data representing a complex world, so that is just the nature of the beast? Or is it both? Stakeholders yearn for simple answers to simple questions, but the road to delivering meaningful results seems paved with potholes of statistical complexity, data nuances, and messy tooling. What is a business to do? Frederik Werner from DHL joined Michael and Tim for a discussion that definitively determined that, well, the topic is…complicated! For complete show notes, including links to items mentioned in this episode and a transcript of the show, visit the show page.