That said, Ice really needs something defensively. It's weak to three of the best attacking types (and one of the former worst...) and resists only itself. I feel like this murders a lot of its viability, and by extension, makes hail strategies SUPER weak.
Depends on the rules, I find. Resistances play the biggest role in switching, but in a format where switching is rare to nonexistent, most notably VGC-style double battling, I see Ice-types come up there fairly often. Currently, the Japanese metagame for Ranked
Sword & Shield battling often uses Lapras, Eiscue, Vanilluxe, Arctozolt, Glaceon, Glastrier, and Alolan Ninetales. (Oddly, the usual suspects for the Ice-type in western metagames--Galarian Darmanitan, Mamoswine, and Weavile--are almost completely absent from the Japanese one.)
I think the lasting presence of Ice-types used by Japanese players is the reason why nothing has been done to increase the number of resistances--because Japanese players aren't really seeing it as a problem. More pressing are the number of weaknesses, as that makes it more likely the opponent has something that can hit them for super-effective damage. To that extent, types with five weaknesses, like Rock and Grass (but not together), type combinations with more than five weaknesses (with the exception of Tyranitar), and type combinations with multiple x4 weaknesses (with the exceptions of Coalossal and Stakataka, though Aggron is rising up), see far less representation from these same players.