To decide what to eat for dinner, we need to define our goals as well as any constraints.
Some variables we could include as either goals or constraints:
Nutrition
Cost
Taste according to the people eating
Taste according to the person cooking
Time to prepare food
Skill to prepare food
Time to eat food
Usage of specific ingredients
Availability of required ingredients
The meta-problem of deciding what to eat for dinner is about navigating these criteria and defining the tradeoffs between them. Is it worth paying a little more for a tastier dinner? Is it worth taking a hit on nutrition to reduce the time to prepare the food?
Depending on your preferences, this problem might be easier or harder to navigate. If you happen to enjoy the taste of some cheap healthy foods that are easy to prepare, you might have a winning choice. If you are a skilled cook you may have more options, but only at the cost of more time to prepare them.
Assuming we do not know exactly how to balance the different criteria, we have an incompletely defined problem to pick dinner based on the criteria. If we knew exactly how to balance everything, the problem would be completely defined with a single right answer. But since it is an incompletely defined problem there could be multiple right answers depending on our priorities.
And so to solve the meta-problem we can try a set of tradeoffs and see what the optimal dinner is for those tradeoffs. Depending on the result, we can update the tradeoffs in the interest of finding a better solution to the meta-problem.