Like Dug the talking dog, too many eDiscovery practitioners are easily distracted by the latest bright shiny object, and there have been so many over the years! Latent semantic indexing, clustering, deep learning… each promoted as the next revolution in the practice of eDiscovery. But seldom, if ever, do you see empirical data that supports on-the-ground improvements in review efficiency or effectiveness. We will show you what the data actually says about some of these “squirrels”, and why you always need to look critically at the latest developments.