Agile is (in part) based on small iterations of tasks to eventually accomplish user stories, basically, right? Can someone explain how this has anything to do

Agile is (in part) based on small iterations of tasks to eventually accomplish user stories, basically, right? Can someone explain how this has anything to do with either administration work or building a physical appliance to sell to a customer who will be using it offline? Because while both have user stories, both are things that require you to not use the agile methodology. When you're doing system/network/database/etc administration, you have a specific request or task to accomplish. But there is no iteration. There is no, let's do a little work and see how it evolves. If you're practiced, you know exactly what to do and how long it takes. So you just do it. Spending time updating boards and stories and tasks is a complete waste. Yet I've seen Agile shoved down the throats of teams who used to simply get shit done as fast as they could, and now go through some mandatory red tape to justify what they've been doing on short staff and no time for years. When you have to sell someone a physical product, I just can't see Agile being very useful, mainly because of the iteration aspect. Yes, you could go through all the motions and provide full features over time and watch them grow with agile, but there's very few times I've seen a product designed from the beginning reach its target goals with any sort of a redesign. In my experience, what I have seen is the Agile method wasting everyone else's time, eventually requiring a redesign. Story time: A meniacal team lead who's trying to make a name for himself in the company sees a way to save the company tens of millions by spending only one or two million on a product. But first we need to customize the product to work for our specific needs, which means writing middleware and customizing the interfaces to give our teams what they need to use the product. Basically, turn a kit car into an actual Ferrari with the 12 cylinder engine and everything. We take six months and actually mostly accomplish it - and then the maniac decides it's time for an upgrade, while teams are trying to learn how to drive this thing. The upgrade requires massive changes, but of course we're Agile, so we iterate and iterate until the Ferrari is now a Lamborghini. It's now been a year, and we're back where we were six months ago, only with a different looking car. Only one team managed to drive the Chimera in that time. The fallacy of iterating on user stories is that the user knows what they want, the product lead knows how to get there without it becoming obsolete, and the developers will magically replace whole systems by iterating over features that they already know will not be changing before the product is done. Personally, I think anyone could come up with any methodology at all and they would have exactly the same success as with any other methodology. The only thing that makes it all work are teams who get their shit done and have the bigger picture in mind, and help each other. But you don't even need a methodology at that point. Agile just seems like so many pie charts for managers, and a buzzword to repeat to gain relevance.