I get that they have built around a monolithic database structure, but with proper site architecture, there should NEVER be a need to take a full, extended outage.
At a bare minimum, this should be the LAST one, and it should be because the work they are doing is transitioning to a new architecture which will never require this.
I know the site is big, but when was the last time you saw Amazon go down for planned maintenance (for any period of time)? Or eBay? It doesn't happen, because it isn't necessary if you build correctly, with site availability as a priority.
Is there any assurance that this outage will lead to never having to take the site down again?