Two songs can have the same number of streams and be in completely different situations. One is played by people who saved it and come back to it every week. The other is riding a single playlist that could drop it tomorrow. The difference shows up in the source of streams: the breakdown of where each play started. Learning to read it is one of the most useful habits an independent artist can build.
Where to find source data
Your distributor's reports tell you how many streams you had on each platform and in which countries, per reporting period. In the CULT dashboard, for example, you see that breakdown together with the source and period of each report. What these reports don't show is the path inside the app: whether the listener came from a playlist, your profile or a recommendation.
That level lives in each platform's own artist tools, such as Spotify for Artists and Apple Music for Artists. Look for the sections about stream sources and playlists. Names and categories differ between platforms and change over time, but the logic is similar everywhere.
The main sources and what they mean
Most artist tools group streams into a few families. Here's how to read each one.
- Your profile and catalogue. People opened your artist page or one of your releases and pressed play. They came for you on purpose. Small numbers here are normal early on; growth here means your name is starting to mean something.
- Listeners' own library and playlists. People saved your song or added it to their own playlist and played it from there. This is the strongest sign of all, because it repeats without any push from you.
- Algorithmic and personalised sources. Personalised playlists, radio, autoplay and recommendation feeds. This is the platform testing your song on people who don't know you yet.
- Editorial playlists. Lists made by the platform's editors. Big exposure, usually for a limited time.
- Other listeners' playlists. Playlists made by users, from a friend's party list to independent curators with real followings.
- Other. External links, search, embedded players and anything that doesn't fit elsewhere.
Reading the playlist list itself
Below the source breakdown, artist tools usually list the specific playlists your song appeared in, with the streams each one brought. A few habits help:
- Compare followers with streams. A playlist with a huge following that sends you few streams probably has your song deep in the list, or its listeners skip it. A small playlist that sends steady streams for months may be worth more.
- Watch the shape over time. Algorithmic playlists tend to rise and fall in waves as the platform tests the song. A slow, steady line from listeners' own libraries is what you want to see after the waves.
- Be suspicious of sudden giants. If an unknown playlist suddenly sends a flood of streams with no saves and no new followers, be careful. Artificial streams can get a release taken down, and you are responsible for any promotion you pay for.
Common patterns and what to do about them
Release week is mostly profile and library. Your existing fans showed up. Good. The question is whether algorithmic sources follow in the next weeks. If they don't, look at saves and skips before blaming the platform.
An algorithmic spike, then a drop. The platform tested the song with new listeners and the test ended. If saves were low during the spike, the song didn't convert. If saves were healthy, your next release starts from a better place, because those listeners now know you.
One user playlist carries everything. Thank the curator, but don't build plans on it. If that playlist drops the song, your numbers drop with it. Use the time to move those listeners towards your profile: follow-up content, your other songs, the next release.
An editorial add with little effect afterwards. Editorial exposure brings plays, not automatically fans. Check how many listeners saved the song or followed you during that period. That number is what stays.
Say Lalə Nova sees that her song "Gecə küləyi" gets most of its streams from listeners' libraries, with a slow but steady algorithmic share. That's a healthy song. Her next step is not to chase playlists but to give those listeners more: a live version, a video, a second single in the same mood.
Mistakes to avoid
- Judging a song in its first 48 hours, before algorithmic sources have had time to react.
- Comparing your source mix with an artist at a completely different stage.
- Treating "other" as noise. It often includes your own social links, so it shows how well your promotion outside the app works.
- Looking at streams without saves. A playlist that brings plays but no saves is renting you attention.
Check the source data once a week rather than every hour. Patterns appear over weeks, and decisions made over weeks are the ones that pay off.