Predictive maintenance, big data, AI, Industry 4.0, IoT, digital twins...it's a technological jungle out there. But the good news for the wind industry is that along with (or despite?) all this hype, data analytics is being used to create significant and tangible value. There are differences in opinion on which methods we should use: neural networks, genetic algorithms, physical models or maybe something fuzzy? In the end I think there is no 'one size fits all' solution. We need to understand the domain, work out how our equipment functions and why it sometimes doesn't. Then we check what information is relevant and available, and choose the best tools for the specific task at hand. If there was only one way to do things, there would only be one organism in the jungle.