Not at all. There are plenty of big networks out there with 10G interconnect with a fair amount of overhead. Adding QoS in for future protection may be a good idea, but there are a lot of networks out there that don't *need* it, per se. I think it boils down to your mean time before refresh, knowing your baselines and anticipating the capacity planning.
What I'm hearing is that there are a lot of folks using it and it seems to be a very manual process of implementation and testing. I suspect that this will need to be rolled into any pre-post testing done with code upgrades as well.