1. Executive Summary & Objective
Problem: Manual playlist curation is a high-friction task that requires users to break their “flow state.”
Objective: To enable “Passive Curation” by leveraging implicit user signals (play-through vs. skip) to automatically build personalized playlists, specifically for deep-work or exploration sessions.
2. User Stories
As a focused worker, I want my music player to remember the songs I liked without me having to switch apps, so I can stay productive.
As a traveler, I want to create a “soundtrack” of my trip based on what I actually listened to, so I can reminisce later with zero manual effort.
3. Technical Requirements & Considerations
3.1 Curation Logic (The “Skip” Threshold)
To differentiate between a “liked” song and “background noise,” the system must apply a logical threshold rather than a binary 0/1 rule.
Positive Signal (Add): Song is added to the target playlist if
Play_Time >= 75%of total duration OR if the song reaches the final 30 seconds.Negative Signal (Ignore): Song is excluded if a
Manual_Skipevent occurs before the 75% threshold.Edge Case: If a song is “Repeat” played, it should not be added multiple times (logic:
If Title_ID exists in Playlist_ID, Skip Add).
3.2 Algorithm Feedback Loops (“Made for You”)
This feature should not just be a siloed list; it should act as a high-intent training signal for the Apple Music Discovery Engine.
Weighting: A song added via “Playlist Mode” should be weighted similarly to a “Love” (favorite) action in the
Recommendation_Scoremetadata.ML Integration: These playlists provide “ground truth” data for Collaborative Filtering. If users frequently add specific “Passive” tracks, the algorithm identifies these as “sticky” discovery content for similar user clusters.
3.3 System Architecture & Syncing
Offline Resilience: For travelers, the “Add” event must be cached locally in a
Pending_Curationqueue and synced to the iCloud Music Library once a stable connection (WiFi/LTE) is established.UI/UX Interaction: The “Playlist Mode” toggle should reside in the “Now Playing” ellipsis menu and the Control Center for rapid access.
4. Success Metrics (KPIs)
| Metric | Definition | Goal |
|---|---|---|
| Curation Depth | Avg. number of songs added per “Playlist Mode” session. | > 5 songs/session |
| Feature Retention | % of users who re-enable the mode within 14 days of first use. | 25% |
| Discovery Conversion | % of “Passive” songs that are later searched for or shared. | > 10% |