I hear what you're saying, but sometimes (e.g., trading/finance) it pays to be closer to the data rather than the things that operate on that data. It has to do with accountability. If you're being asked to put your job on the line every time that you (for example) run a model based on some data that's provided to you, then you're going to want to personally inspect the data for correctness before running your model. Often-times a user of an Excel-based financial model in the Banking industry will perform numerous ad-hoc checks (charts, pivot tables, random calculations, etc.) on the data before officially running their model. That's tough to do when the data isn't sitting in a software package like Excel. PS. I've never posted before -- your blog is excellent