The Five Numbers to Check Every Week
Published · updated · by Loopin
Checking your streams every day tells you nothing and costs you an hour a week. Five ratios, checked once, tell you what to do next.
1. Save rate — the only leading indicator
Saves divided by streams. Under 3% means listeners heard it and moved on. Above 8% means the algorithm has a reason to keep serving it after you stop pushing. Between the two is normal.
This is the one number that predicts next week rather than describing last week. A track with 2,000 streams at 10% saves is in better shape than one with 8,000 at 1%, and the second week will show it.
2. Listeners divided by streams
Streams per listener under 1.3 means people played it once and left. Above 2 means they came back. If this number is low across everything you release, the problem is the song, not the promotion — more spend on ads will buy you more single plays and nothing else.
Check it per track, not across your whole catalogue. One repeat-heavy track hides in an average and it is exactly the one you want to find, because that is the sound to make more of.
3. Where the streams came from
Spotify for Artists splits this into your own profile, editorial playlists, algorithmic (Radio, Discover Weekly, Release Radar), listener playlists, and other. The number to watch is algorithmic as a share of the total.
Rising algorithmic share means the platform is doing work you are not paying for. Falling share while your own posting stays constant means the track is cooling — that is the moment to change the clip, not to post the same one harder.
4. Follower growth per 1,000 streams
Divide new followers by streams, per thousand. This tells you whether people who hear the song want more of you or just liked that song. Under about 2 per 1,000 and you are renting attention; above 10 and you are building something.
This is the number that decides whether to spend on ads. Ads applied to a track with a low follow rate buy streams that evaporate. Applied to a high one, they buy an audience for the next release.
5. Which city moved
Not the top city — the one that changed most since last week. A city that jumps is almost always a playlist add, a share by someone with local reach, or a scene picking the track up, and it is the single most actionable thing in the whole dashboard.
When one moves, do something with it within days: post something referencing that place, check whether a local playlist added you, look for a support slot. The window where that spike is still warm is about two weeks.
What to ignore
Daily stream counts, total all-time streams, and follower count as an absolute number. None of them change a decision — they only change your mood, which is why they are the ones everyone checks.
Set one recurring ten-minute slot a week and look at nothing in between. The data does not get more useful when observed more often, and the hours you save are the hours the next song needs.
Frequently asked questions
What is a good save rate on Spotify?
Under 3% is weak, 3–8% is normal, above 8% is strong. It is the best leading indicator you have, because saves are what keep the algorithm serving a track after your own promotion stops.
How often should I check my music analytics?
Once a week, for about ten minutes. Daily checking does not produce better decisions — the numbers are too noisy at that resolution to act on.
Which metric tells me whether to run ads?
Followers gained per 1,000 streams. Below about 2 the streams are not converting into an audience, so ads will buy plays that vanish. Above 10, ads are buying listeners for your next release too.