There were a lot of reasons the old start menu was removed. I listed some technical reasons why "just leave it in as an option" isn't as easy as it sounds.
It is not a straw man to say that investing in one area means you have to sacrifice in others. This is what we call software development.
Features are not added in service packs, and I don't even think service packs exist anymore (service packs are just collections of security and reliability updates, created and packaged together by a special "sustained engineering" team, not the product team).
The product team is building a new start menu option for a future version of Windows. It's a significant undertaking, not something they're throwing together willy nilly. They don't have a time machine, so they can't go back to 2012 and ship it then.
Still, given that the missing start menu is most folks' #1 complaint about Windows 8, wouldn't you have to say that (at least in hindsight) it was a mistake to leave it out? Yes, software engineering is always about compromises, but isn't it also about making the right compromises?
It's a straw man to suggest that putting in the effort to update a fancy menu would have had so much impact that the product would have been ruined. But I think I misread your clause about 'disjointed' at first.