MicrosoftRebuilding of all indexes and using the same amount of freespace for all indexes is way overkill and only serves to make such maintenance a whole lot longer than it needs to be and can seriously waste the precious resource known as "Buffer Memory" for any tables that are mostly static.
The correct way to maintain Random GUID indexes is to assign a Fill Factor of 71 or 81 only to those indexes that suffer from fragmentation. All the rest should be assigned a Fill Factor of 100 (or left at 0, which is the same thing in all other respects). The "1" in the Fill Factor is to remind you that you MUST do a rebuild (NEVER A REORGANIZE ON RANDOM GUIDS!!!) as soon as you hit 1% fragmentation. Your instincts on how often that will cause a rebuild are incorrect. I've tested a system where I inserted 100,000 rows per day and near the end of just the first year of the simulation, the Random GUID Clustered Index on the 123 byte wide test table went 58 days with <1% fragmentation and no index maintenance using a 71% Fill Factor and 32 days using an 81% Fill Factor. Do the math... that's 5.8 and 3.2 MILLION rows respectively,
Check out https://eightkb.online/ for my 28 July 2021 presentation (it'll be live that day and available as a recording after that) for how it all works.
