How to check if a Spotify playlist has listeners

· 7 minute read

As an artist, submitting your song to Spotify playlists is usually one of the important steps to grow your number of streams and try to find new fans and listeners.

Doing so is not an easy task: many curators ask for a submission fee to listen to your song, and then add it in their playlist if they like it. Those submissions have a cost which you have to take into account in your budget and consider these as an investment. And as every investment, you have to make sure to get something in return: real streams from real spotify users that will like your music.

Our tool Isitagoodplaylist.com provides a lot of data to help you analyze a playlist before submitting your song, to make sure it has real listeners, to make sure it's not a fake playlist full of bots, ... There is so much data to look at, it's not an easy task. Here is how I personnally look at the data to understand if a playlist is good or bad for a specific song.

Top genres of the playlist

On the first tab of the analysis, "General Infos", the first data to look at is the genres of the playlist. This list is built by taking into account the genre information from all the artists in the playlist, and it's sorted so that the first genre is the one that appears the most.

The genre of the playlist is important for two reasons:

  • It has to match the genre of your song. If that's not the case, the listeners might be tempted to skip your song as this is not what they expect. That means less streams, and a bad signal to the spotify algorithm.
  • There need to be some logic in the playlist genres. If the playlist goes from pop to hiphop to classical music to hard techno to deep house to lofi to ... The Spotify algorithm will be lost and won't be able to categorize your music and artist profile correctly. The impact will be huge for all the algorithmic playlists, radios, ...

The followers evolution

The followers count is the only public metric given by Spotify to measure how popular a playlist is. However this is a very bad metric if we just look at it without more context.

A follower does not necessarily means a listener. The curator could have bought fake followers to increase the metric, to look like his playlist is a good one. The followers could have followed the playlist a long time ago, but do not listen to it anymore. So a big follower count is not always a good sign. On the other side, a small follower count is not always a bad sign either: It's better to have 1000 followers highly engaged with the playlists, than 10,000 followers who never listen to it.

So instead of looking at the number, you have to look at the evolution. We record the followers data regularly for millions of playlists, and display this data on a graph. This way you can easily see if the followers evolution is legit: it should grow slowly, at a regular pace.

The graph above shows a followers evolution which seems legit. It went from 4k to 35k followers in a period of 18 months, with a steady growth.

On the contrary the graph below shows something you should avoid at all cost: sudden spikes and drops. It probably means the curator bought fake followers, which are sometimes removed by Spotify, then he buys again some of them and so on... You can bet that if you get streams from this playlist, most of them will be from bots, which is something to avoid as well.

Listeners data from the Discovered On

For each artist, Spotify shows a list of playlists on which they have been discovered, which actually means playlists from which they got listeners in the last 28 days. In the past, Spotify also displayed the number of listeners coming from each playlist but removed this information a while ago. Too bad!

We use this Discovered On section intensively in our analyses. Here is how:

  • If a playlist does not appear in the Discovered On section for any artist, it probably has (almost) no listeners.
  • If a playlist appears in the Discovered On section from artists, the playlist probably has listeners. To know if it has many listeners, you can look at the number of monthly listeners from the artists (the more the better), and the Discovered On position (the closest to 1 the better).

Based on this table, you can vaguely estimate an upper limit on the number of monthly listeners of the playlist. Here is a real life example:

Only two artists have our example playlist in their Discovered On section. The artists have 2,195 and 1,073 monthly listeners, and they have been in the playlist for more than 28 days. Also, they are at position 11 and 8 from 48 tracks, which is quite in the top of the playlist.

I think it's safe to say that we can be sure this playlist has less than 2,000 monthly listeners, as otherwise the artists would have a much higher monthly listeners counter.

Also, as the artist with 2,195 monthly listeners has this playlist at position 24 in his Discovered On section, it means he received more streams from 23 other playlists than this one. The math will not be 100% correct but we are just trying to guess ballpark numbers here: if all the listeners from the artist come from those 24 playlists, the 24th would bring him less than 91 monthly listeners (2,195 / 24). Doing the same exercise with the other artists gives us a maximum of 97 monthly listeners.

Now let's confirm this very gross estimation by looking at the artists that are in the playlist but do not have the playlist in the Discovered On section. I only kept the last entries for our example as I will use only one artists data but of course you should do the work for more than one artist.

We see that one artist has been in the playlist for a long time, more than 28 days, and at a very good position (3 out of 48 tracks). Based on that, most of the listeners of the playlist from the last 28 days would have listened to his track, and would therefore be counted as Monthly listeners for this artist. Now you see the point, this artist has only 308 listeners so it means the playlist has less than 308 monthly listeners.

At the end of the day if you put everything together, you can easily guess that this playlist has not much listeners, less than 100 (maybe even less than 10, as our estimations used very optimistic hypothesis) even though it has thousands of followers.

Streams data

Thanks to our community our artists, we have stream data for thousands of playlists. You can get access as well to our stream estimation, it's completely free.

Based on the collected data, we are able to compute a stream estimation for the playlist, giving you and indication of how many streams you get over one day, a week or a month if you are added to the playlist. Again, this is an estimation. Even if it's based on real data, many parameters will influence how many streams a song gets from a playlist, such as the position in the playlist.

While this streams estimation is a great number to look at, it's no silver bullet. Streams could be coming from bots, or from a very limited set of users. That's why this is important to look at all the parameters together and not only the stream count.

It's also great to compare the number of streams with your estimation of listeners you did previsouly. Too many streams compared to the number of unique listeners? It could be bots, or maybe hardcore listeners playing it on replay.

And some other parameters as well...

I like to look at the track popularity distribution. It indicates if the playlist contains popular tracks or more underground tracks. I believe it's best if you target a playlist that will match your popularity. For example, imagine if someone is listening to a "Top hit" playlist with only popular tracks, but then an underground track he never heard starts playing. What is the probability that he will skip it in less than 30 seconds?

The track age is also a great data, not that important compared to the others, but again if you're listening to a playlist with only old tracks (that we could consider as classics), wouldn't it be weird if a brand new song comes up?

Finally the Audio tab of the analysis shows you the audio features of the tracks in the playlist. You can easily see if the playlist contains tracks with a high danceability, energy, if these are more instrumental tracks, live recordings, high or low tempo, acoustic tracks, ...

The listeners tab also displays one other interesting data: the Top city for each artist of the playlist. Sometimes you can spot that the playlist has bot listeners thanks to this information. If you see a lot of the artists have the same Top city, that's weird. If they all have pretty much the same amount of listeners from this same Top city, that's even weirder. If this top city happens to be close to a datacenter from amazon where bots could easily be hosted, that's annoying. Or if the number of listeners from this top city is much larger the the population of the city, boy that's not good at all.

Conclusion

There is so much data to process, let's recap the checkbox you have to tick before submitting your song to a playlist:

  • Does it make sense for the listener if my song is in this playlist? Look at the genres, popularity, audio features, ...
  • Is the playlist legit (no bots)? Look at the followers evolution and Listeners data. Maybe use Stream data and Top city information as well.
  • Has the playlist listeners ? Look at the Streams data if available, and the Listeners data.






Looking for good playlists accepting submissions?

Sign up for our newsletter to get a list in your inbox every week.

© Isitagoodplaylist.com All rights reserved.