Personalized Workout Playlists That Adapt to Your Gym Session
Prioritising eight ways to make YouTube Music aware that the listener is in the middle of a workout
✳The problem they went after
People who train at least 15 days a month get recommendations that ignore what they are doing. YouTube Music has no idea when a workout starts, what kind of exercise it is, or how hard the user is pushing. The result is skipping mid-set, switching playlists by hand, and shorter listening sessions overall.
✳What they actually did
A longlist of eight solutions was scored with RICE. The winner is a Personalized Workout Playlist Generator that builds only from tracks the user already plays frequently during workouts or has saved, so an unfamiliar song never breaks focus mid-set. Second is Adaptive Intelligence, a learning layer that watches skip and completion behaviour specifically inside workout sessions. Third is a dedicated Workout Mode triggered manually, by location, or by smartwatch, capturing intent before the session starts. All three got wireframes.
✳What you can take from it
Worth stealing
- Eight-solution longlist scored with RICE before anything was designed
- Playlist built only from songs the user already trusts
- Workout detection via manual trigger, location, or smartwatch
- Continuous listening minutes as the headline metric
Read this if
PMs who need to turn a long solution list into a defensible shortlist
The transferable bitRICE does real work here. The top three ideas stack into one product rather than competing, with the playlist generator delivering value immediately and the adaptive layer improving it over time. The metric choice is equally disciplined: average workout listening session duration measured in continuous minutes, with skip rate and feature adoption underneath.
✳The submission itself
Open the PDF →Reading on a phone? The embedded viewer is cramped — open the PDF in a new tab instead.
