Turns out we're just running our refresh functions one after the other which isn't ideal but we can fix that separately. As it stands this wait group isn't doing anything.