Case study · Feature proposal
One interaction, proposed as a change to a product I use every day.
Spotify already has multi-select. It lives on desktop only, and even there it selects everything between the first song and the last. This proposes the mobile version: press and hold a song, tap the others you want, and move only those. The scope is one interaction inside an app that already works.
The product
As a long-time Spotify user I noticed a gap between two halves of the same app. Discovery is fast. The app surfaces things constantly, and saving a song you like takes one tap.
The organising half is much slower. Building a playlist on your phone means adding one track at a time, and that repetition is enough that a playlist you thought of on the bus usually never gets made.
The gap
Spotify's own community moderators answer this question directly: multi-select works in the desktop app and it does not work on mobile. What desktop gives you is shift-select, so when you pick two songs that sit far apart, everything between them comes along too. If the tracks you actually want are scattered through a thousand-song library, that hands you hundreds you never asked for.
Every person I spoke to described both their listening and their playlist building as something they do on their phone, which is the one place the feature is missing.
What people said
I interviewed Spotify users about how they actually build playlists, then sorted what came back into what I learned and what it was costing them.
What I heard
The cost
Discovery feels effortless. Organising songs into playlists is tedious.
Adding songs individually takes too many steps on mobile.
Most people rarely create or update playlists, despite wanting to.
The repetition discourages spontaneous use.
People prefer making their own playlists over using premade ones.
No bulk selection means less personalisation and less control.
People want a faster, more intuitive way to manage several songs at once.
The current design has no efficient way to add, remove or reorder in bulk.
The current flow
The complaint above is a general one, so this is the specific version of it. These are all the steps the app asks for today, from opening your library to getting the first song into a new playlist.
Your Library, then tap Create.
Choose Playlist from the menu.
Name it, then tap Create.
The playlist is empty. Tap Add to this playlist.
Tap the plus on one song. Then the next. Then the next.
Steps one to four only happen once, but step five happens once for every song you add, and that is the whole problem. The app gives you a way to say this one over and over again, and no way to say these five.
Two directions
The complaints sorted themselves into two distinct moments. One is picking several songs out of a list you are already looking at, and the other is reaching into a playlist you already have to pull songs out of it. They are different problems that happen to share one gesture, so I sketched each of them against the interview notes before I built either.
Both directions went into the prototype. The first defines the gesture, since it is where someone would meet selection for the first time. The second is where it returns the most work, because pulling from an existing playlist is the case where people move the largest number of songs at once.
The feature
The header becomes the toolbar. A live count, plus add and remove.
Three picked out of nine.
The rows in between stay untouched, and dim. Desktop shift-select cannot do this.
The same header, the same count.
The response
I put the prototype in front of ten people and asked two questions.
8 of 10 said 5, extremely intuitive. 2 said 4, very intuitive.
9 of 10 said 5, extremely likely. 1 said 4, very likely.
If it shipped
None of this proves the feature would work at Spotify's scale. If it did go out, there are three numbers that would tell me whether the argument held up.
If people never find the gesture, the problem stays exactly where it was.
Takeaways
I learned the importance of aligning usability with user intent. Good design should feel invisible. Users should not have to think about how to complete a task. It should just make sense intuitively.
Even small inefficiencies can discourage users from taking action. At times I overrefine my work for the sake of presentation rather than clarity. It is often better to go simpler and focus on making the simple action as close to perfect as I can get it.
Designing something holistically starts with deeply understanding the problem, the psychology behind user actions, and the small details that shape behavior. Even minor inefficiencies in a flow can lead to larger drops in engagement over time.
