Spotify Blend: Product Design Tricks that Get You Hooked (on Blend, also the person 💕)

Emily Zhou
Bootcamp
Published in
6 min readNov 26, 2022

A case study on how Spotify Blend gets us hooked, what metrics it contributes to, and where the future of Music lies.

I’ve been watching The Playlist lately. It’s been mixed feelings that, in one dimension of the world, Netflix is feeding us dreams with shows like The Playlist, inspiring us entrepreneurs who still have hope, while in another dimension, tech companies are (finally) shifting from enlarging their user base and market to actually profiting. I always loved music, worked in the music industry (for a tiny bit), obsessed with the idea of working in a cool music tech company once. And now I’m sitting on my couch watching the show, with my phone in my palm screaming what is happening in this world. I started doubting the faith I always had for Music.

Blend was my getaway island.

The Magical Moment

I’ve tried Blend before with a couple of friends, didn’t work out. Then a month ago I was FaceTime with someone, I had this random idea and asked her “hey do you want to do a Blend with me?”. I admire her taste of music (she’d say me too) and already have a handful of her songs living in my Liked. I’d kill to see our Blend.

A mix as unique as the two of you. We got 89%!

That moment was magical. 89%. We have a perfect Blend, with tracks we were 100% they’d be in it. The few lines Spotify wrote about our Blend was also beautiful, musical as usual.

And then something really interesting happened.

Day 1, 2 and 3, the Blend was a fair mix of our likes. Everyday or every other day, I would save a few of her tracks in our Blend. About a week after, I noticed that the Blend, or our “baby” that we joke about, started to add more and more of her songs. The majority became music she liked. Looks like our baby resembles one parent.

Baby resembles one parent

This continues for the rest of the month, and today 10 out of the 50 tracks in our Blend are mine, 7 we both liked, and the rest, which is 66% of the tracks, are hers. And I’m staying, she is too. WE ARE HOOKED.

The Product Design Tricks, The Numbers, and The Minds

Here’s what I think that got us hooked.

#1 Updates Daily turns you into a DAU

We were fascinated by Spotify’s Discover Weekly once (some of us still do), and it made us WAUs. It was good enough then, but I am one of these music lovers who’d die for their favorite bands and reluctant to discover new loves. I tend to discover new bands recommended by bands I follow already. I stopped following Discover Weekly a few years ago. It’s this new person I met and the Blend we started that got me back to discover new music on Spotify — and this time, I was made a DAU.

Updates daily

And guess what, because the mysterious mixing logic behind, everyday I wake up thinking about what tracks our Blend is bringing to me today. If I didn’t go through the whole playlist one day, I knew it’d be gone forever and I’d never know what was there. And if I did go through the whole list very quickly, then it’s a 20+ hours wait till I see its new face the next day.

The curiosity, the longing, and the fear of missing out got me so hooked on this Blend that expires every 24 hours.

#2 A Mix so intelligent that brings up Streams, Likes, Artists Follows, and Retention of paid users

Coming back to the suspicious unbalanced track % distribution, I doubt the mix was ever designed to be fifty-fifty. I’ve got 2 assumptions.

Assumption 1: The strategy is to focus on who’s more likely to stay, invest, and pay more. Getting this person converted is considered a success.

I’m a paid user with an individual account for about 6+ years, heavily following a fair amount of artists (which means it won’t be easy for me to leave after all I have invested), while the other person is on a family plan for about 3 years, mostly relying on Discover Weekly for new music. The mix learned that I was open to stream, and more likely to save and follow music picked. Because I’m more likely to: 1. stay (and pay more), 2. save songs in Blend (be converted), the mix learned to pick tracks that are more likely to convert Me, which are tracks the other person liked. And that explains why >66% of the tracks come from her, not me.

Liking a track animates like falling for somebody

Assumption 2: Strategies vary across user groups. And I need the other groups’ side of story, please share with me as I’m dying to know.

For this trick, I think it’s us attaching whatever feelings we have for the person we Blended with to the tracks in our “baby”, the sense of togetherness (or even love) that we unconsciously apply to and romanticize these songs, that got us so hooked.

#3 The Insights that you were touched and stay for

That again, “a mix as unique as the two of you”, the moment you touches that magical bubble, you were pulled into the shared memory (and a shared future), reading beautiful words while listening to the best part of your favorite song together.

The wheel spinning, our hearts pounding

The design is so musical and romantic that it makes you dream. If you (and Spotify) is lucky enough to get a perfect match, first you’d be touched, and then before you know it, you’d be wanting more — “what if I can turn 89% to 98%, or 100%, so we are a 100% match”?

#4 The Perfect Human Match we believe in

Lastly, the smartest bit of this design goes to the perfect human match it attaches to and relies on. Of course there’s doubts how big the market is and how sustainable it is (infatuation doesn’t last, human bond does, but how many of us listen to songs shared by our loved ones daily).

I think it’s the human match that gets us into Blend at the very beginning, been sustaining it for some time so far (we are willing to stream the other person’s tracks because we believe in the connection we have musically and personally). It’s beautiful and I’m excited to see how it works out.

The human match we believe in

The Future of Music

So far there are two things that I’m most unsure of about Blend. One is sustainability, and the other is the market size.

I’ll have to stay with Blend a little longer to see what our baby grow to be. The baby resembles one parent, do I love them (and most importantly, stay with them) unconditionally, even when I don’t love that parent anymore? And speaking of market size, is 89% a common case? I had blended with other before, but only got hooked recently because of a perfect match. How many 89%+ pairs are out there?

I love Spotify’s design from the bottom of my heart, since the day I discovered it — well, not as much as I love Music, but enough for me to stay. While watching The Playlist, I repeated the words I said to my younger self many years ago: I don’t believe in the future of record labels, I believe in the future of Music. I’d like to add one more statement in it, I don’t believe in the future of one tech company, I believe in the future of Music, plus human creativity that sparks beautiful ideas which fuels the future of Music.

--

--

Emily Zhou
Bootcamp

Product design for the people and the planet. Currently with Microsoft. Proudly vegan and queer 🌱🌈.