Don’t Split the Streams — Part 4 — Making sense of Spotify Wrapped
What’s missing in Spotify for Artists? Data Triangulation
A look at music marketing campaigns — including song exclusivity to specific music streaming platforms, and how music metadata fields are handled by such platforms can make stream counts vary wildly
In part 2 of this series about making sense of a musical artist’s stream counts on major platforms like Apple Music, Amazon Music, YouTube etc I explained how Spotify’s year in review artist stats program — Wrapped — operates. It’s an annual review of a musical artist’s performance on Spotify.
The artist stats in the Spotify Wrapped package are great — but they leave questions unanswered that are up to the musician to figure out. There will soon be a new Spotify report for all you musicians — Wrapped 2020 — probably coming out early December. In part 3 of the series I detailed what would drive a Spotify artist to be able to net 1 million streams annually.
Spotify does not do any of the “data triangulation” in it’s Wrapped program, but if you are a musical artist or manager who’s been logging into the Spotify for Artists platform regularly to view your music consumption stats, you will probably able to do such triangulation quite easily.
For part 4 of this series, Don’t Split The Streams, I dig in on how to make sense of the Spotify Wrapped for Artists stats via data triangulation.
What do I mean by triangulation? It means tying together the various data points Spotify offers to specifically figure out what’s driving the consumption of your music.
Let’s look at what big data points Spotify Wrapped 2019 offered up — I highlight each one at the top of each section following in the post. Then using my music group FSQ, and our own Spotify Wrapped 2019 report, I attempt to understand the drivers behind the consumption of FSQ music on Spotify in 2019.
I’m trying to answer questions that are left unanswered by the presentation of the Spotify Wrapped data points.
You are going to want to do the same when you’re 2020 Wrapped drops, so the idea of this post is to help you understand what to look for in advance.
Some of this data triangulation work pulls in music data sources external to Spotify.
Here are the main data points that Spotify offered in past Wrapped reports. It will be interesting to see if they expand the scope of the year end stats offering for the 2020 report.
Ok let’s break down each Spotify Wrapped data point, given what they offered in 2019.
Most streamed songs for the year (released by the artist), and what day of the year these songs had their greatest streaming peak
Here, you would want to ask WHY?! are these songs the most streamed out of all the other songs in your catalog and why were they streamed the most on one particular day?!
Ok let’s do some data triangulation. First, using more data available to us within Spotify’s Artist data platform.
In my case for FSQ, the most streamed song of the year in our Spotify catalog was our FSQ remix of Life On Planets “Cold Front” with 23,900+ streams on Spotify in 2019.
Again, I have been keeping track year around — usually a check-in each week, if not 2x-5x a week looking at the Spotify for Artists to see stats for FSQ.
The power of playlists — keeping track of their impact on your overall stream counts
So before the Spotify Wrapped year end review was released, I was already aware FSQ’s remix of “Cold Front” was having some success versus my other songs on Spotify in 2019. I knew the majority of our Spotify song plays came from Bon Entendeur, a French DJ and production collective who added this song to their artist curated Spotify playlist.
To be exact, 15.6K of the 23.9k streams of the song in 2019 came from Bon Entendeur’s playlist. That means 65% of the streams of FSQ’s “Cold Front” remix in 2019 were driven by people playing the tune within Bon Entendeur’s playlist.
Using the other Spotify for Artists data, which includes a chart of playlists your songs have been added to, I was able to point back to the Bon Entendeur playlist addition as being the main driver of our FSQ “Cold Front” remix being our most streamed song in 2019.
Spotify for Artists also tells you the date any playlist curator first added your songs to their playlist. In this case, February 21st, 2019. So this data point does not explain why on March 8th, we received the most plays of this song.
Now let’s triangulate the Spotify Wrapped data with some external data sets.
SOUNDCHARTS and CHARTMETRIC:
Soundcharts and Chartmetric are subscription services that roll up artist analytics from Apple Music, Spotify, Amazon Music, Deezer, other social media channels. Furthermore they are exciting analytic platforms because they also keep track of broadcast radio airplay beyond streaming. They are paid subscription services more geared to artist managers and record label marketing teams. Chartmetric starts at $140 a month, while Soundcharts is about $60 a month. Using these offerings, one thing a manager or label marketer could do is triangulate Apple Music and Spotify events — like playlist additions — to see if behavior on one service drives fans to also consume music on another platform.
GOOGLE:
Google is free. So using it to get some ideas of why we see trends is a great way to figure out more about why musical artists are gaining streams.
Example: Through some Googling with the terms “Bon Entendeur” and “FSQ”, I learned that also on February 21st, 2019, Bon Entendeur broadcast a radio show including that FSQ “Cold Front” remix.
APPLE MUSIC FOR ARTISTS:
It’s a good practice to compare your Spotify for Artists artist analytics against your Apple Music for Artists analytics. Both are free offerings once you’ve registered as an artist.
Via Google, we already know that the playlist addition of FSQ’s Cold Front remix to the Bon Entendeur Spotify playlist also coincided with them playing the song on their online radio program. So did the radio show play impact the plays of our song over at Apple as well?
Yes — Bon Entendeur’s radio play of FSQ on Thursday, February 21st 2019 also led to a nice bump of play on Apple Music that very same day.
Comparing: 43 plays of FSQ “Cold Front” remix on 2/21/2019 on Apple Music and over at Spotify we see 156 streams for that day for the same song.
Meanwhile, Bon Entendeur did NOT add us to their one and only Apple Music artist driven playlist.
We also received 5 Shazams (mobile app music listeners use to identify unknown songs) of this song, the day Bon Entendeur added the song to the Spotify playlist and played it on their radio show. The Shazam data is available via Apple Music for Artists, as Shazam is now owned by Apple Music.
I can not explain the bump in Spotify plays of this FSQ Cold Front remix on March 8th, but let’s say it was still within a few weeks of Bon Entendeur adding us to their playlist and it was getting a bit of extra action that day.
The specific day and hour of the year an artist had the most # of Spotify users streaming the artist’s music
Apparently on March 27th of 2019 we had the biggest volume of streams of FSQ music in a single hour — 33 streams per hour at 4PM on Spotify. Not sure if this means we got 33 streams at 4PM exactly, or between 3PM and 4PM or between 4PM or 5PM.
Why is this Spotify Wrapped data point that offering important? Well obviously I did something right at that specific moment last year to make it the most volume of streams of FSQ music. I would love to repeat the feat. I have not been able to find any third party data that would explain to me why we did so well at this specific hour and this specific day.
Without any context on why your music streamed the most at a particular point in the year, this stat could be considered irrelevant.
Many music artist marketers will know — maybe the “most streamed hour / day of the year” was when they launched a new release for their band or solo artist. Maybe they ran an Instagram live release video pointing fans to the stream on Spotify on that day. Maybe a targeted email went out to an engaged fan base, pointing them towards a particular song or playlist on Spotify.
In the case of FSQ, our 2019 EP release came a week before March 27th. We checked our records: there were no FSQ Facebook or Instagram posts, no tweets, and no email campaigns sent on this day. We did have a pretty big DJ gig on the 29th of March 2019 in Miami. Maybe there was more interest in FSQ on Spotify in the days running up to the gig. In this case, we can’t any triangulate data to figure out what drove our most “most streamed hour / day of the year”.
The specific country / nation in the world where an artist had the biggest increase in audience size year over year
Triangulating this data point isn’t exact science but its fair to say if you had a really big increase in streaming from a single country you would know the reason why.
Maybe you would expect an artist would grow the most in terms of Spotify streaming in their own city and country. Especially if the artist is getting local radio play or performing regularly live in the area. Both activities can drive streaming.
What if you see tremendous streaming growth in a country far away? Sometimes the language of an artist’s songs can help growth abroad. Say maybe you or your artist released a Spanish language song that gained some popularity in a few Latin American countries. Or for instance there’s quite a bit of music listening connections between the Portuguese speaking countries, also known as the Lusophone market of Brazil, Portugal, Angola, Cabo Verde.
I keep tabs monthly tabs on countries and cities where our music is streamed. Looking at the month of March 2020 I noticed something unusual — I can NOT tell you why our 2nd top city for where our music was streamed that month is of all places Santiago, Chile! That city also never showed up again in the the top 10 city stats. In September 2020, Santiago was the 25th most popular city for FSQ.
New York and Los Angeles seem to make sense — large US cities with people into dance music, and we are a US based artist. It’s nice to see Amsterdam and London in the top 5 too when we took this screen shot in March.
In March 2020, FSQ did not run any marketing campaigns in Chile either ;) Nor did we run one in Paris, France. But Paris has been our top city for streaming for a while. At the time of writing this — September 2020 — Paris is still the top city for FSQ streaming and France is the top country.
And looking back at our Spotify Wrapped 2019 we also learned that France was huge for us for the year.
This should be no surprise: The tremendous growth in France — our FSQ audience grew 557% in the nation from 2018 to 2019 — it’s tied back to the French group Bon Entendeur adding our “Cold Front” remix to their Spotify artist playlist in February 2019.
Now how did I figure this out? As I mentioned Bon Entendeur’s playlist netted us 15K streams in 2019 of the FSQ “Cold Front” remix, and with 60K overall streams for 2019 that means that one song, on that one playlist account for one quarter of all of our streams. It turns out, voilà! The majority of Bon Entendeur’s listeners are French. For music groups / artists with a sizable following, Spotify publicly displays the geography of their listeners.
Now Bon Entendeur “playlist listeners” aren’t the same as listeners to the Bon Entendeur song productions, so I’m only inferring the followers of their playlist are French. Still with such a high volume of our streams coming from one playlist, and the incredible growth of our audience in France in 2019, it’s safe to say confidently the addition of FSQ to the Bon Entendeur playlist drove our almost 600% year over year audience growth in France.
In 2020 and beyond we are doubling down on marketing to French audiences to continue to support our success there.
Can Spotify playlists drive broadcast radio adoption of songs?
I was not using Soundcharts and Chartmetric at the time that the Spotify Wrapped 2019 report was released and I wish I had been because I would have been better able to triangulate how the Bon Entendeur playlist addition of FSQ drove broadcast radio airplay of FSQ in France, or vice a versa.
Soundcharts does not let me look at 2019 historically, but looking at the past 12 months rewound September 2020 to September 2019, I see that the majority of radio airplay of FSQ is from France.
At the time of writing this piece, September 2020, UK radio airplay of FSQ is up 50% year over year. The British radio success is due to the fact we released an album in July with a UK based PR campaign led by Shine Music PR. Shine pitched the FSQ album to DJs at radio UK based outlets like NTS Radio and Worldwide FM. The pitches were successful as you see by the data. Still while French radio play of FSQ is down 27% year over year it’s at this point still 3 times as much airplay versus the UK.
Playlist inclusion benefits are tempered by “Playlist Decay”
While FSQ’s Cold Front remix today is still included on Bon Entendeur’s playlist — that playlist continues to add songs and therefore our position in the playlist is further and further down from the top of the list. When they added us in February 2019, FSQ Cold Front was at the top of the playlist — in the first 10 songs. By September 2019, it was the 30th song and in September 2020 it’s the 92nd song of 191 songs on the playlist. This is what we would call “playlist decay”. The curator keeps adding songs and your song keeps falling in the position in the playlist. The reason this matters is that if listeners stream the playlist in track order (vs using shuffle mode) there’s less of a chance your song will get streamed and heard.
For instance, if each song is average 4 minutes long, at position 10 in the playlist it would take a listener about 40 minutes of streaming before they get to hear your song. At position 30, it’s now two hours of listening before they get to your song.
Now at position 92 — where FSQ’s Cold Front stands now — it will take you over 6 hours of listening to the Bon Entendeur playlist before you would encounter our song.
I’m bringing up the topic of playlist decay to indicate that I am aware as we come up on Spotify Wrapped 2020 that the FSQ Cold Front remix will may not account for the majority of our streams in 2020 as it did in 2019. If it does count for the majority of our streams, it will certainly be less than the mark of 25% of our 2019 streams driven by that one song on the Bon Entendeur playlist.
The % increase of followers an artist gained year over year
FSQ gained 89 new Spotify followers in 2019 ; that’s not much. And the growth was pretty steady throughout the year, slightly climbing each month. I asked friends to follow us but the data shows that I didn’t put much effort beyond 1:1 requests to get people to follow us on Spotify.
Do Spotify followers drive stream counts?
My take on the whole “Spotify Follow” button on artist profiles is that most fans don’t know what “following” a Spotify artist profile really means. So it’s an uphill battle to get new fans or listeners to click the follow button. Also the position of the follow button on artist profiles changed in 2019 on the Spotify mobile app ; so it’s not as clearly visible as it was before. Furthermore, I do not ever recall Spotify doing any marketing of the follow feature, reminding listeners to “follow” their favorite artists.
For artists on Spotify, increasing the follower number is critical, especially before dropping a new release. All your Spotify followers will be served your new release both via email alerts (if they have them turned on) and in the weekly Spotify playlist known as “Release Radar”. The “Release Radar” playlist can wind up being a big driver of your Spotify streams.
Marketing to gain Spotify followers ..
There is a little bump in the chart above— I circled it in red. The increase from 270 to 290 followers the first week of November 2019 — I know exactly why that happened. FSQ had an DJ booking — but in order to RSVP to the party — we used the Toneden.io digital marketing platform — to require people to follow us on Spotify before they would receive a RSVP confirmation.
I think it’s important to run marketing campaigns that do drive up your follower number on Spotify and the 2019 data tells me that without such campaigns our follower count will grow only slightly year over year.
With the help of a digital marketing expert friend, just before our new FSQ album was released in late July 2020, we set up a Facebook advertising campaign. The ad campaign was designed so that if someone clicked through from the Facebook ad to the “follow” FSQ landing page, and then clicked “follow”, that we would be charged for that click. We are paying Facebook advertising for the result of gaining followers.
You’ll need a Feature.fm or Toneden.io subscription or alternative technology to create these “follow / subscribe” landing pages. You can also make an embeddable Spotify “follow” button to add to your own website.
The reason why I used Feature.fm landing page versus my own website to drive new Spotify followers is that Feature.fm landing pages can pass a code back to Facebook ads when people actually click “follow”.
The Facebook ad campaign set at a budget of $125 added about another 30 FSQ Spotify followers. That works out about a cost of $4 per new follower. It’s actually a very expensive campaign.
In the days where we were still having live DJ or band gigs, I would be working the room after we got off stage to get people who said they liked the music we played to follow us. “Give me your phone, let me add FSQ to your Spotify”. Today digital marketing campaigns are one of our only avenues to increase our Spotify followers.
Spotify Followers vs Spotify Monthly Listeners
Having a greater number of followers should drive more listening of an artist’s music, especially when a new release from that artist comes out. That’s why Daniel Ek, CEO of Spotify says that artists will need to release music more frequently to reach their audience. To Mr. Ek’s point — FSQ’s monthly listenership was close to 7,000 at the time of our album release in July 2020 and now has fallen to 3,000 in September just two months after the release.
It’s interesting though because having a greater number of followers doesn’t always equal a larger listener base. Remember David Marston, an artist I profiled earlier in this “Don’t Split The Streams” series, for his ability to get to get to one million streams annually on Spotify? His follower number is only 1.3% of his monthly listenership.
Meanwhile, Soul Clap — a production team and DJ duo — you find that their follower number is 23% of their monthly listenership of about 150,000.
At the end of the day, it’s better to have a higher number of followers in relation to monthly listeners because someone that is following you demonstrates that they really do like your music. Still, I do not think Spotify has done enough to educate users about the follow button.
Year over year % increase of an artists’ overall streams, total listeners, and brand new listeners
Ok let’s break down the rest of these FSQ 2019 stats pictured —some of them we already covered. Again the exercise here is so that you will be ready to make sense of your Spotify Wrapped for Artists 2020 report when it is released in early December.
Playlist adds up 7%
I talked about how the Bon Entendeur playlist drove 25% of our overall Spotify streams in 2019. Seeing that playlist addition for FSQ was only up 7% year over year — combined with the stat that Bon Entendeur had such an impact on our streams — makes us realize we need to do much more to get playlist curators to add our songs to their playlists.
New listeners up 14%
That seems very low for 2019. The lack of new listener increase speaks to us needing to spend more on digital marketing to potential new fans.
We had one FSQ EP release in 2019, on March 18th and we got a small bump of listeners in late March as a result, up to about 350 listeners per day from an average of 100 listeners per day during that time period.
Again, Spotify CEO Daniel Ek’s statement about artists needing to increase the amount of music they release to gain listenership is relevant. You can see the truth of that statement reflected in this chart which FSQ spans monthly listeners from 2015 until 2020. Our highest listenership came on July 24th, 2020 when we got 793 listeners on the release day for our FSQ album, “Reprise Tonight”.
Looking back beyond the recent data, I see we also got 776 listeners on a single day in September 8th, 2018. We had an album release on June 12th 2018 but that didn’t result in much of a bump. It’s very difficult to go back two years to triangulate data sources to understand why that particular day popped. We missed the spike that year (Sep 2018) because we weren’t consistently checking our Spotify for Artists stats ; our miss is a good reminder to be dialed into them each week so you don’t miss a big event like that and the opportunity to figure out what drove the growth.
Total listeners up 60% ; Streams up 26%
I put these two stats together because they should be compared. I wish Spotify had an overlay graphic where you could see streams and listeners at once. Here they are side by side for FSQ for the 5 year period from 2015–2020.
You want to understand your streams to listener ratio.
Looking at 2015–2019 if 76.5k listeners drove 221.9k streams — that means on average each listener played 2.90 — almost 3 FSQ songs. This is known as your streams to listeners ratio.
If our listenership was up 60% for 2019, but our streams were up only up 26% that means that ratio likely went down.
Indeed with 22.3k listeners driving 59.9k our stream to listener ratio went down in 2019 — with now each listener streaming about 2.6 FSQ songs. The higher the number goes the better because it means you’re getting repeat (and potentially more loyal listeners).
Jake Udell of the site Art of a Manager writes that “repeat listenership may be the single most important metric to determine whether cult fandom is taking place.” He concludes that top hits get about 5–12 streams per listener with most #1 hits leveling out at 10 streams per listener.
You’ll want to be paying attention to your stream to listener number in Spotify Wrapped 2020. If you have a high number of streams, but lower number of listeners it means your songs are really connecting with your audience.
Getting ready for Spotify Wrapped for Artists 2020
Thank you for reading this breakdown of how to make sense of Spotify Wrapped for Artists data points. In advance of the 2020 Spotify Wrapped for Artists release— you’ll want to be keeping track of the following so you can make sense of the report quickly.
- Monitor for any weekly spikes in your Spotify stream #’s and listener #’s and try to identify the causes behind them which could be because:
— Your songs got added to big playlist
— You dropped a new release
— Radio airplay drove people to check out your songs
— A social media post — e.g. a TikTok, or Instagram story — mentioned your songs
- Check Spotify for Artist Analytics against Apple Music for Artists data weekly, especially the Shazam data that is included with Apple ; Shazam data indicates people trying to identify your songs being played in public places or via radio broadcasts
- Measure Radio airplay of your songs via Soundcharts or WARM Airplay ; Chartmetric has radio airplay data to a smaller extent. See if there’s any link between radio airplay in particular cities and your Spotify streaming stats
- Keep track of what countries and cities are driving streams of your music on Spotify and try to uncover the cause (for instance, if your music is doing well in Japan! You should try to figure that out)
- If you want to do well on Spotify for the year, make sure you’re investing in your growth of followers — use a marketing platform like Feature.fm or Toneden.io in combination with Facebook advertisers to grow your Spotify following
- If you have music ready to release, release it during the year and really push the streams towards Spotify if you want to see your stats go up there e.g. “Don’t Split the Streams” across too many streaming platforms
Comparing your Spotify Wrapped 2020 to other artists performance
I covered in part 3 of the series what independent artists should be seeing in terms of performance with stream counts on Spotify. Are one million streams for a Spotify Artist common?
Don’t Split the Streams
One of the things I wanted readers of this series to think about it is the audience for your music is split up across the various streaming and retail music platforms to which you distribute your music. How successful you are on any one of those specific platforms like Spotify depends on the following:
- How that streaming service or retail service counts up streams and sales (for instance, do you get a stream count if you are a featured or remix artist)?
- How much of your listening audience is actually on each one of these platforms. For instance, on Apple Music we only get 1/10th of the streams we get on Spotify each week.
- How much do you invest in marketing to potential and existing audiences on the platform where you want to have success. Are you driving fans to the platform where you want to have the most success? Or is your audience “split” up among many platforms
- Related to the last point — what marketing relationships you or your record label may have with particular music retail sites and streaming services. Which of them can feature your music? If any of them can, you should make that platform the primary focus of your marketing efforts and audience management, listenership growth
Looking at our overall listener # presented in last graph displayed, it says FSQ netted 76,500 listeners on Spotify from January 1st, 2015 until September 20th, 2020. Compare that stat to SoundCloud, where ONE song we have — our FSQ disco remix of Nick Monaco’s Babyface — has 75K streams over the same period.
Now mind you, streams aren’t the same as unique listeners. One person could stream a single song 10 times. But you get the point I am trying to make — you’re going to have successes in some areas (SoundCloud, YouTube) and you may struggle in other areas (for us, it’s Apple Music and a lesser extent Spotify).
Coming soon — In Part 5 of “Don’t Split The Streams”, I will detail how music marketers are focusing on Spotify because of the streaming service’s attention on playlists.