You can train coders, increase your interrater reliability, and make improvements to instruments. Research teams can compare coding of the same or different videos or even discuss specific events in their data.