How do you roll out R to hundreds of colleagues, ensuring that the version you’re providing is tested, qualified and well managed? Is it possible to ensure that everyone is using the same version of R and packages? How do you account for differences between departments and groups in which R packages they need? How do you minimise effort so that you’re not perpetually in the process of qualifying and deploying R and packages? How do you balance the need to be flexible and allow colleagues to try new R packages and methods, versus having R under strict version and change control to ensure that their results are submission ready and reproducible in the long term? In this talk we’ll discuss how we’re trying to achieve this at Pfizer.