One thing I don't see in the discussion here, is how many monitors your users are using. Developing and testing on a different environment than your customers

One thing I don't see in the discussion here, is how many monitors your users are using. Developing and testing on a different environment than your customers are using is asking for problems. It's the same issue as developers who develop and test on high bandwidth connections and then end up missing issues that customers on low-bandwidth connections see. I remember working on an application that ran in a clinical environment where available screens were limited (so the user base was almost 100% single screen). The developers used two screens exclusively. There was an issue that was reported by the customers which the developers could not reproduce, which would cause a complete freeze of the system. Finally, I went through the process step by step with one of the developers. I noticed that the developer would click on a pop-up message that was appearing on the right screen. The customers did not have a right screen to click on. So the process required clicking on a pop-up appearing off-screen. The problem was simple to fix, but was never noticed because the environment was different. I stuck with a single screen for years because it was what my users were using.