I think styles that introduce large changes to campaigns like domain-building need to be optional, but to do them well the structure for using them in a campaign should be baked-in. Shortly after 5e was announced I suggested in similar threads that "campaign themes" analogous to character themes might be baked into the basic system. That way there is some consistency of structure for the strategic elements of (say) god-killing, domain-building, or purely-mercenary campaigns and their interaction with the party. This would also be a perfect way to handle things like adventure paths or specific game settings with some specialized rules.
Obviously such things can be added to a campaign without this additional structure, but I think the results wouldn't be as clean. For example, in a game that focuses heavily on domain-building the in-game responsibilities and authority of individual PCs could differ quite a bit, so there is an advantage to defining a broad spectrum of character abilities that one could take (kingly authority to call loyal knights, a spymaster's access to intelligence, the high priest's influence over the local church and its laity, and the unique abilities of the king's trusted bodyguard) that interact with resources introduced by the campaign theme (the order of knights, a spy network, the church treasury and prayers of the faithful, and humbler figures in the court that lack formal influence but are rich in privileged knowledge). Having a structured way to hitch player abilities with these strategic resources introduced by the theme means one can more easily prevent system bloat or unbalancing the party, which happens more readily if all these elements are layered on top independently.
As another example, consider a campaign ending in the rise or fall of a pantheon. Here the campaign resources might be something like the favor of the gods themselves. Sometimes this affects the party in a more strategic fashion (say the creation of a demiplane from which they coordinate against opposing forces) as well as smaller boons that can be taken instead of those from normal themes (say calling forth in battle the divine-fragment that preserves and sustains the demiplane).
Finally, one might even have a system for "leveling up" the campaign so that the strategic elements closely reflect what the PCs actually do and have, and not necessarily what they "should" have based on the theme. Lots of campaigns don't have a single pursuit from beginning to end, but evolve organically. An example might be a campaign that starts with the party as duplicitous members of a thieves' guild who eventually insinuate themselves into the official power structure of the land. In that campaign much of the actual power of the PCs might come not from their ties to the nobility (as in a more typical kingdom-building game) but from their ties to the shadowy underworld. They might seek to use those resources to impoverish the nobility, in fact, while keeping the appearance that their own hands are clean. Completely unrelated systems for running the thieves' guild and kingdom-building might not introduce conflict between the two elements, but without some shared language they probably can't complement each other very well either.
