Integrated escort works relatively well in simpler environments or under human supervision. But when you get into complex environments of heavy jamming, modern integrated defense networks, peer or greater-level large potential CAP, I start to use more escort missions that, while indirectly connected to the strike missions, work independently for maximum flexibility on ingress, engagement, and egress.
The short of it that in more complex scenarios, one-size-fits-all mission planning is not the optimum solution. Its why some people want the AMP. Theoretically, the AMP will let you plan each individual component within the plan for its own mission, but all linked to the ultimate plan. Current mission planning that includes strike, tanker, and escort is a very generic and rigid approach to mission planning.
The thing that is great about CMO is that its a game that gives you back what you put into it. If you skimp on planning, you'll find all kinds of issues that some perceive as game issues. But the game gives you the tools you need beyond anyone's expectations 7-8 years ago. If you want to jump into action immediately, you'll get what you planned for.
Its also why I think that the AMP is some amorphous ideal that has expectations around it that it is some magic button you hit to plan complex missions. I think the people calling most passionately for some kind of AMP are going to be sorely disappointed no matter how much effort the devs put into it. In large scenarios, there is still going to have to be a s**t-ton of work put into OOB planning and adjusting missions paths, ROEs, ordinance plans, recon plans, intelligence reviews, tanker management, etc. That will not get done magically. Will it be easier to get a baseline to work with...probably. But if you aren't putting in the planning work today for these big complex scenarios that some people like, the AMP won't change that.