Group by and sub-setting with dplyr

Question

I need to calculate mean by account_id for (2) scenarios:

1. Scenairio 1 ALL observations
2. Scenairio 2 Primary owners

Scenario 2