Skip To Content

Quarantine Playlist: A Data-Informed Jukebox

Reading time: 8 minutes

By Jon MacMillan, Product Manager

In quarantine, most of us are consuming a lot more media than usual. There is one word that has been used repeatedly in the news to describe the sudden increase in media consumption, which is a “surge”. Netflix and Disney+ surged in the past month. Traditional media, like TV and Radio have also seen increases in consumption. Similarly, Spotify’s paid music subscribers surged to 130 million in the first quarter.

With everyone turning to services like this we wanted to offer a little insight into what we enjoy, and in a way, update an old post of ours which shared the type of music that we listen to when working.

With that in mind, I set out to collect playlists from some employees here at Rapid Insight, requesting that they cultivate a list of 10-20 songs that speak to them. In the end I got 7 submissions totaling over 120 songs. I asked the contributors to submit a paragraph that describes why they chose the music that they did.

Employee Playlist Contributions

Hear it and Feel it Playlist ▶️

by Ryan Orlando – Senior Account Manager

My playlist represents music that, when I hear it, I feel it.

For example, songs like Soundgarden’s “Black Hole Sun, take me to my senior year in high school. More specifically, to a day in early June. My friends and I skipped school. Cruising through town, we saw my mother in a car in the opposite lane. She drove by us, unaware, and we laughed and laughed at our luck. Every time I hear that song, I remember that day and feeling so free and hopeful.

Similarly, I feel “Rooster by Alice In Chains and “Say It Aint So by Weezer. These songs take me through time as well. Other songs give us permission to feel the way we do like “Short Skirt/Long Jacket by Cake and “Send the Pain Below by Chevelle. Two very different songs by very different artists, both validations of my own experience.

I’m not too wrapped up in the beat, and I don’t dance. You should be glad. I enjoy music that can be both subtle and assertive, with a preference for music like rock and eighties metal. I guess you could think of my playlist as a picture of my memories and feelings that you can hear.

Packed Playlist ▶️

by Joe Aliperti – Senior Software Developer

Hard to limit this playlist to just 20, I tried to put together a collection of tunes from various genres and eras that move or inspire me in some way. Some bring up memories of the past, others transport me to another part of the world, and a few are timeless classics.

Meandering Playlist ▶️

by James Cousins – Analyst Manager

I tried to design my playlist to transition smoothly between tracks. Rather than add my favorite songs in the order that I recall them, I thought it would be neat to try and have one flow nicely into the next.

“Standing Outside a Broken Phone Booth with Money in my Hand” is a nice, slow start to things, and with a gradual, washed out outro, the punctuated tempo of “Raining Again” is a fun contrast, but not a complete shock as far as tempo goes. I was aiming for a similar relationship between every track, but I hope that feels apparent to listeners without an exhaustive description from me.

More to the point, this is not the first time I’ve created a playlist for Jon! The last time I submitted songs, it turned out to be tame even for a library. I wanted to prove that I did sometimes listen to upbeat, major key, danceable music. I hope I can count on your vote as best playlist 2020!

Coffeehouse Playlist ▶️

by Lily Brennan – Data Analyst

While many of the songs on this playlist are favorites of mine, it is not a compilation of my all time favorite songs. Instead while making this playlist I tried to focus more on putting together a collection of songs that played nicely on the whole. That ended up being sort of a light and calm but upbeat grouping.

I think the product is something I could easily listen to while working because most of the songs are not really sing-along songs making them ideal for focus, but still catchy enough to keep you interested and occasionally leave the tune lingering in you ears for later.

I also put some attention to the order of the songs in the playlist. Since some are slower and a little more sleepy, I was hoping to lull back and forth from slower to a little more high energy to keep the listener from getting bored. I guess overall I would describe it sort of like coffeehouse music, ideal for the background and goes great with a cup of tea!

Recenter and Repeat Playlist ▶️

by Alisha Martin – Digital Strategist

My playlist is a mixture of songs that I grew up listening to on repeat, and songs that help me recenter. Being easily distracted impacts my productivity, but I’ve found that having a sense of familiarity with the music I listen to is essential, regardless of the tempo/style.

I’m often picked on for listening to “sad” music, which is ironic because the “sadder” a song is, the happier it usually makes me.

If you want to check out my playlist, make sure you start with the first song, and then play the rest on shuffle. I almost never listen to a playlist sequentially!

Mellow Playlist ▶️

by Earl Sires – Digital Content Marketer

Since I wasn’t yet with Rapid Insight to offer up a playlist when Jon did his first blog post on this topic, I created a playlist of some of the music I listen to while working.

I can’t focus if a song has lyrics (unless it’s a song I’ve heard a million times, like the Talking Heads song on this playlist), so most of these are instrumental tracks with either a calming or invigorating vibe. When vocals do appear in these songs, it’s usually as an instrument: either harmonically, or else chopped and spliced to the point where it’s recognizable as voice, but without comprehensible lyrics.

Electronic music with a consistent rhythm helps me get in the zone while I work, especially songs that incorporate physical instruments, which tend to be comfortable and familiar to listen to.

Eclectic Playlist ▶️

by Jon MacMillan – Product Manager

I think what is evident in my musical selection is that I like a lot of variety. My playlist consists of acoustic, instrumental, blue grass, electronic rock, rap and more.

As opposed to some of the other submissions, I didn’t cultivate this playlist for any particular mood or listener. These songs make me happy and serve different purposes for me.

Some songs like “Your Smiling Face” by James Taylor and “We’re Going To Be Friends” by the White Stripes are songs that my daughter and I enjoy together. Other songs like “Everlong” by the Foo Fighters are songs that bring me back to high school and remind me specifically of a close friend.

A lot of the songs are nostalgic, but many more of them are just easy listening that I enjoy having on in the background while working. I hope you enjoy them as well.

The Data Behind The Music

As I mentioned, we are consuming a lot of media at this point and I just unloaded 7 new playlists on you, containing over 120 songs and over 9 hours of music. That’s an entire work day’s worth of tunes. You can listen to all of the songs here if you prefer that approach, but if you want a more selective process, don’t worry. Just like Spotify does with its recommended playlists, I have gone through the data to help identify the playlist that nearest matches your particular style. I won’t spend too much time on the inner workings here, but the data is really cool (if you are into that stuff) so I will share a little of what I did to put this together.

Spotify allows users to access song data through their API. Using Charlie Thompson’s well documented spotifyr package, I was able to pull down the track information from all of our playlists to get metrics like the positivity (valence) or energy of songs. If you are interested in more about the measures you can check that out here. But the data is confusing and trying to keep track of what all of these measures mean is difficult. I found myself referencing the definitions countless times to remember exactly what the speechiness of a song is. With that in mind, I wanted to create a way for you to sift through these playlists and find the one most aligned with your musical taste, assuming that it is unlikely you know what your ideal valence level is.

Pick The Playlist For You

Rather than having you set filters for energy, tempo, valence, and other settings, I came up with an alternative approach. In the dashboard below, you can select the song you are most interested in listening to using the album images at the top (hover over them for artist and song name). None of these songs match any playlist perfectly, but instead help to guide you to a playlist that may be quite different, but consists of similar musical features. To determine those pairings, I used a formula in Construct to match each song’s attributes to the average ratings of songs in each individual’s playlist to identify the nearest match. The formula relies on 6 different attributes (check beneath the dashboard for a detailed definition of each).

These measures help to identify the best option for you among the plethora of songs we have offered up here.

Use the dashboard below to find the playlist best for you. You can also use the dashboard to navigate to the individual playlists or play individual songs from each playlist. If you want a similar analysis done of your musical preferences, send us a link to one of your playlists to find out which artist you most closely align with.

Spotify Attributes

  • Acousticness – a confidence measure from 0.0 to 1.0 of whether the track is acoustic. 1.0 represents high confidence the track is acoustic.
  • Danceability – describes how suitable a track is for dancing based on a combination of musical elements including tempo, rhythm stability, beat strength, and overall regularity. A value of 0.0 is least danceable and 1.0 is most danceable.
  • Energy – a measure from 0.0 to 1.0 and represents a perceptual measure of intensity and activity. Typically, energetic tracks feel fast, loud, and noisy. For example, death metal has high energy, while a Bach prelude scores low on the scale. Perceptual features contributing to this attribute include dynamic range, perceived loudness, timbre, onset rate, and general entropy.
  • Liveness – detects the presence of an audience in the recording. Higher liveness values represent an increased probability that the track was performed live. A value above 0.8 provides strong likelihood that the track is live. Liveness here is the average liveness value for all songs on the individual’s playlist.
  • Tempo – the overall estimated tempo of a track in beats per minute (BPM). In musical terminology, tempo is the speed or pace of a given piece and derives directly from the average beat duration. In this case Tempo was normalized to a scale of 0 to 1 where 1 is the fastest song, in terms of BPM in the sample.
  • Valence – a measure from 0.0 to 1.0 describing the musical positiveness conveyed by a track. Tracks with high valence sound more positive (e.g. happy, cheerful, euphoric), while tracks with low valence sound more negative (e.g. sad, depressed, angry).

Stay up to date
Subscribe to our blog

Notify of
Inline Feedbacks
View all comments