Spotify's Summer Breeze
An exercise on intuition from minimal data, some product knowledge, and a reasonable understanding of human behavior
Pictured above is a Spotify playlist of Summer Breeze by Seals and Crofts, a soft rock duo from the 1970s. Spotify is a large music streaming platform. The playlist's tracks mirror the original vinyl format, without each side noted. Each track includes a track length and a play count. It takes at least thirty seconds to count as a play. Tracks appearing on several albums may have their plays combined.
"Summer Breeze", "Hummingbird", "East of Ginger Trees", "Yellow Dirt", and "The Euphrates" combine plays with other albums1. Ostensibly, this inflates their plays compared to other tracks.
Let's begin.
The other evening I decided to listen to the LP2 of Seals and Crofts' *Summer Breeze*. I've been practicing the KonMari Method3 on my vinyl collection. My ground rules are simple, I listen to each record in its entirety. If the record doesn't spark joy, it leaves my home.
After finishing the album, I went to Spotify to play a song I'd enjoyed. I didn't know its name, but I knew it was on the B-Side. I went to the album on Spotify to find the song and I thought I'd use the play counts as a guide post. The one with the most plays must be the track I thought was best? Right? Examine the playlist above and take a guess. It'll be fun.
The only audio-marker I had was "Summer Breeze". I knew it was on the A-Side. From there I couldn't pinpoint the track. But it led me to notice that the number of plays indicated something curious. Many other people had taken the same journey I had, only they'd done it on Spotify. There is muddiness in the data, but seeing how many people finished the album after starting it—like I had—was interesting. I'm competitive, so what.
First we must establish how many people started the journey. Since "Hummingbird" shares plays with _Greatest Hits_, it's difficult to tell. We know approximately 300,000 played the next track, "Funny Little Man."" That gives us a more information, but the potential is greater due to how plays count.
Remember, it takes thirty seconds to register a play. There's a chance people didn't make it past the first thirty seconds of "Hummingbird." Even if they did, there's a chance they didn't finish it at all. With all due respect for Seals and Crofts, it's not that far fetched to imagine. It's unlikely people jumped to "Funny Little Man" directly. Again, no disrespect.
"Hummingbird" has a play count buoyed by following "Diamond Girl" on _Greatest Hits_. I'll give humanity the benefit of the doubt. People starting the album with the intention to play it through finished the first song and the first thirty seconds of "Funny Little Man" before giving up. Let's approximate 400,000 intentional listeners.
Disclaimer. I am not a music critic. I know what's good to my ear—that's it. But it's more fun if I offer my opinion. The first two tracks suffer for the same reason. They start poor and end poor. Was it Seals or Crofts decision to do this? They should put all their middles together. Just saying.
At least 250,000 people stuck around to play the third track, "Say." It doesn't mean they finished it. Even if the next track is the title. The track was a struggle. I bet a percentage of listeners decided even "Summer Breeze" wasn't worth it and put something else on—or nothing at all. Only 60% of attempted full-listeners remain.
Which is saying something, because "Summer Breeze" is wonderful—a true pleasure.
The massive difference in the plays of "Summer Breeze" versus its auditory siblings doesn't tip many hats. About 250,000 entered from the previous track, and most of the other plays were from folks that searched the song or heard it on another album. A small number opened the album and began playing the track. I'm confident this is the case, because the fifth track "East of the Ginger Trees" has more plays than "Say."
A quick break to discuss the difference in mediums and how it may affect human behavior. "Summer Breeze" was the second to last track on a black vinyl disc. If you put the record on to play only the fourth track, you had to drop the needle in place. It's feasible, but the friction involved far exceeds what it takes to play the track—and only that track—on Spotify. The plays say little about modern tastes and more of how low-friction experiences affect our behavior. On Spotify there is scant cost to change our mind.
"Fiddle in the Sky" is an oddity. It doesn't combine plays with any other album, it's two tracks after the title track, and it's the third most played song on the album. That means that enough people found this song somewhere in life and went direct to it on Spotify. It says a great deal. I'd be curious of the demographics behind the plays. Decades ago, it might have also been the third played song on the record due to it being the first track on the B-Side.
One third of the plays for "Fiddle in the Sky" appear to be those that aimed to complete the full album. The plays of the next three tracks are almost equal. That consistency says something about the quality of "Fiddle in the Sky." When I reached the track I felt vindicated that I discovered another song I enjoyed. It was that point I resolved to finish the album. Others agreed.
Now if you guessed that I finished the album and went on Spotify to play "Fiddle in the Sky," you'd be both right and wrong. You'd be right because it is the track I played first, wrong because it wasn't what I was looking for. But I'm positive I'm one of the few people who opened _Summer Breeze_ on Spotify and jumped two tracks ahead of the title. It's not a persona4 they thought to cover.
The reason I came—the track that prompted the search—was "The Boy Down the Road." It's not earth shattering, but it has a unique sound and composition. Based on the plays, few people agree with me. I tend to play songs I enjoy on repeat until they've lost their luster. Sort of like how my daughter wears holes through her favorite pants. If I do that to this track, I still won't put a meaningful dent in the plays. Are there other diehards out there?
Let's recap. A bunch of people, myself included, put on _Summer Breeze_ with the intention to listen in full. We were all well on our way, and we liked one song enough to close it out and see what remained. Three tracks left.
The plays of "The Euphrates" don't tell us much. It's a different scenario with "Advanced Guards." What can we intuit? We know roughly the same amount of people listened at least thirty seconds of "Advanced Guards" and "The Euphrates." What we don't know is how many people finished it. Based on the 90,000 play difference between it and "Yellow Dirt" it indicates one of two things. Either a fair number of people bailed on the penultimate song or the final track was bad enough they didn't last thirty seconds. Did they bury their worst tracks at the end of the album? Would you?
I have a confession.
As I write this paragraph, I still haven't listened to "Yellow Dirt" either. My journey remains incomplete. To be fair, I did think the penultimate track was the last. I noticed my mistake when I looked at the record and saw another light band of vinyl remaining. The dip in plays for the final track was one of the first things that jumped out at me when I went to Spotify. Here's the deal, it's actually worse than it looks. "Yellow Dirt" shares plays with another album, so the minimal plays are also inflated
There are four types of people in the world. Those that never played "Yellow Dirt" on Spotify, those that did but didn't last thirty seconds, those that did but didn't finish it, and those that completed the track. Which will I be? Let's find out.
Ok, it doesn't help that it's over five minutes long and it's not exactly a song I would describe as getting better as it goes. I'll be real with you, I'm guessing very few people made it to the end of the track. I don't blame them, time is precious. Yet a life without regret is a life dulled by certainty.
My best guess is that less than a quarter of those that started finished. Would more have finished if they'd chosen a different medium for the journey? There is friction involved in buying a record, bringing it home, putting it on the turntable, tuning your receiver, listening, flipping the record over, and listening once again. But I choose to think of the friction as a ritual. With acts that require intention and enhance the experience.
While *Summer Breeze* didn't spark joy, the experience did.
"Summer Breeze" has a single, but their plays aren't combined. "Hummingbird" and "Summer Breeze" appear on Greatest Hits, Get Closer, and Traces, but only combine plays with Greatest Hits. "East of Ginger Trees" is on Greatest Hits and Get Closer, but doesn't combine plays with either. "Yellow Dirt" is on Get Closer and combines plays with it. "The Euphrates" is on Traces but it doesn't combine plays.
Long play.
Marie Kondo created the KonMari Method. In its essence, she says you should pick up every object you own and only keep those that "spark joy."
Personas are a technique practiced by user-centric product designers used to represent a specific type of user. In my case, nobody sat in a room and agreed to design a Spotify feature that met the needs of a person who played Summer Breeze on LP and then came looking for a song they liked on the B-Side.