I think it's pretty common in tech jobs as well. A frequent setup is that you have a longer-term project that's really interesting but not deadline-pressing, so you can only work on it when you've gotten other immediate things out of the way first. There are attempts to try to explicitly make schedule space for the interesting long-term work (like Google's 20% time), but it still often ends up that you have to get a bunch of other less-interesting stuff off your plate each day or week, before you can get to the best stuff.
Yeah, I've got an interesting website reflow project that's on my todo list and will make a big difference, but it's not a priority I can block time out for, and the daily stream of tasks that need to be done "now" means I rarely get to it, but simply adding more hours to my work day wont actually produce the desired/best work.
Not to mention that quite a few small companies don't have the ability to hire a dedicated sysadmin, IT staff, and/or customer support rep. These tasks falls to the engineers.
Doing these tasks is almost certainly a suboptimal use of any single product engineer's job, but if they don't get done by someone the business won't be a business for long.
