Discover
Conducted in-depth interviews + Analyzed Deezer onboarding + industry best practices
A project focused on improving the onboarding experience for users switching from other music streaming platforms to Deezer.

Context
With an influx of users leaving other platforms, Deezer needs to create a smooth onboarding experience that captures each user's unique music tastes.
How can we ensure new users feel at home on Deezer by identifying their tastes with minimal effort?
Deezer is one of Europe's leading music streaming platforms, serving millions of users across global markets, :
Role & approach
As part of a 2-week product sprint led in collaboration with Deezer's internal product design team, I led the end-to-end exploration of how to improve the actual onboarding with two others designers.

Conducted in-depth interviews + Analyzed Deezer onboarding + industry best practices
Insights collect + Problem framing
→ Problem statement
Ideate → Shape new onboarding + prototyping
Usability testing + iteration + impact
Problem
After signing up, new arrivals known as “Switchers” are required to choose at least 10 artists, which are supposed to represent their musical tastes.

Objectives
Eliminate
the feeling of loss
Accelerate the understanding of preferences
Create a 'wow' effect right from onboarding
User research
Understand the arrival experience and the perceived effort during onboarding
Understand the initial usage moments until the definitive adoption of the app
How users define themselves musically and feel a sense of belonging?
Google Meet 45 min
Users who recently switched or are considering switching music platforms.






Users choose a platform based on their values, as their musical tastes involve the artists and their remuneration, which contributes to a sense of "home".
Since I’m starting to produce music and we have artist friends, I think it's nice that artist compensation is fairer.
Users find that the current onboarding process is cumbersome and time-consuming, failing to reflect the true diversity of their musical tastes.
I really don't like this thing of being guided too much step by step, being led by the hand.
Users do not expect the onboarding to provide a selection of their tastes, which they consider counterproductive as it may skew the algorithm during the transfer.
I think I would be a bit stressed about choosing the wrong artists and that it wouldn't give me a good algorithm.
The transfer is seen as a key feature for successful onboarding.
In terms of mental load, it has relieved me immensely.

So many are aware of playlist transfers but do not have enough information to take action.
Problem statement
Make library transfer the heart ♥️ of a simple, quick onboarding process that aligns with users' values.
Also known as the IKEA effect, positive friction involves enhancing the perceived value of a product. A piece of furniture assembled by oneself will have more value than one that is pre-assembled.
The aim is to find the right balance between an overly light onboarding process that diminishes the perceived value of the service and an overly burdensome onboarding process that hinders access.
Ideation
The solution already exists. All we need to do is present it and improve the other approach.
Onboarding Benchmark
| Criterion | Spotify | Apple Music | Amazon Music | YouTube Music | Tidal |
|---|---|---|---|---|---|
| Number of screens | ~16 | ~9 | ~9 | ~7 | ~17 |
| Personal account before | ✓ | ✓ | — | — | ✓ |
| Artist selection | ✓ 3 min. | ✗ | ✓ 3 min. | ✓ 5 min. | ✓ 3 min. |
| Artist format | Grid, photos | — | Grid, photos | Grid, photos | Grid, sections |
| Artist search | ✓ | — | ✓ | ✓ | ✓ |
| Genre filters | ✓ (tabs) | — | ✓ (tabs) | ✗ | ✓ (sections) |
| Explicit genres | ✗ | ✗ | ✗ | ✗ | ✗ |
| Podcasts | ✓ | ✗ | ✓ (topics) | ✗ | ✗ |
| Loading personal | ✗ | ✗ | ✓ | ✗ | ✗ |
| Upsell premium | ✗ | ✓ (from screen 2) | ✗ | ✓ (from screen 1) | ✓ (end of flow) |
| Notifications | ✓ | ✓ | ✓ | ✗ | ✗ |
| Music transfer | ✗ | ✗ | ✓ | ✗ | ✗ |

from Dice
Sync your music with clear named button

from Amazon Music
Personalized experience

from Tumblr
Sub-topic
Prototyping
→ 100% free for all your playlist → Simple = find your ex-app, connect, transfert → All your music (*almost)

Deezer offers a new onboarding process that begins with the selection of at least one music genre, which will be linked to a selection of reference artists in the chosen genre(s).
Limited the choice of genres to the most well-known, listing sub-genres in the 'Other' modal. We have ruled out the Tumblr option, considering it too gimmicky.
The selection of artists remains optional, allowing each user the freedom to either contribute to the algorithm or not.
A warm welcome for newcomers, with reassurance that highlights their musical taste
→ For people who won’t transfer: Deezer is their first music streaming platform
→ Or for those who prefer to wait and see what Deezer’s algorithm can do before bringing their playlists

We add the transfer bloc from Library to Home, because the transfert is the more efficient tool to bring your taste in your new music home
+ this is the heart of our purpose.
Iteration
→ Quicker (x3) → Simplier (no noise) → Clearer (bigger subtitles)

+ Sometimes users take time understanding they can find more type of music → change ‘Other’ button style
+ Go back CTA → Obvious missing in first prototyping
+ Add explainations to avoid misunderstanding
▲ Expected impact
x2
→ Number of switchers using the music library transfer
▲ Proxy Success criteria (2nd path)
In the early days, the new incomers uses the app from the home screen with Deezer recommendations + likes for the recommended tracks.
Provide a loader or even a transfer percentage and explore app in waiting time
Integrate TMM directly into the app to avoid any disruption in the user experience
The observation regarding values is a starting point for the change in application but is ultimately not a subject related to our issue → transfer to the marketing team
It might be useful to analyse where switchers are coming from, in order to target the call-to-action buttons leading to these platforms more effectively. However, as TMM is an external service, we are unable to track cta to these platforms. A form could therefore be useful for identifying Deezer’s cta links to TMM. Identified button (label: Spotify...) could also be a good solution.