A lot of ostensibly talented developers do the same thing—it's not always easy to predict in advance what is the most efficient expenditure of effort.
You'd have be stupid to disagree with that. But there's a difference between "most efficient" and "efficient at all". I've seen Excel spreadsheets used to track orders, at a rate of 100 or so a day. Then it gets too big so new spreadsheet files are used and the filenames are dated (using inconsistent date formats, making it difficult to find a spreadsheet for a specific date range, and none of the regular users thought to rename the files). But finding historical orders is a nightmare because the order numbers are not based on dates. And two people need to edit the same spreadsheet at the same time, or someone gets the bright idea that they'll be more efficient if multiple people can enter the orders, so someone makes a copy of the current master, and the changes never get merged in. So there's duplicated data or data for overlapping date ranges in multiple files and no authority on which data is valid, it needs to be looked at every time to find out which one is the "real" version. But even that's hard because columns have been added and removed and reordered on all of them. And most of these issues are not even a matter of "using the right tool for the job", even though the root problem is just that; they are caused by not spending even a half-hour to think about things before doing them, and not having enough experience to say "boy, this is a real headache to deal with". So inefficiencies get piled on top of inefficiencies, all in the name of automation. And the few clueful people who are forced to use the system end up with a bad impression of computers.
If an ostensibly talented developer creates a system like that, they can remove "ostensibly talented" from their title.
If an ostensibly talented developer creates a system like that, they can remove "ostensibly talented" from their title.