19 Comments
Aug 8Liked by Felix Lee

I believe it was 2 things:

1. Version B had artists pictures/videos, that grabbed user attention and distracted from the USP (million songs).

2. there’s a content mismatch, artists or songs? Maybe if the USP was “more than a million of artists, for free” would connect better

One more thing, we can call version A “boring”, or we can call it “distraction-free”. I thought it was a clever use of white space 😁

Expand full comment
author

Interesting take, love it. Thanks, Mauricio.

Expand full comment
Aug 7Liked by Felix Lee

Simple design and less cognitive load wins over crowded design/elements.

Expand full comment
author

These always work at the moments it matter most.

Expand full comment

I agree with the two hypothesis put forward but I believe there is a third that has been overlooked.

I think the third is, who the artists are in the background image impacts who may sign up. Modern music has thousand of genres and sub-genres leading to more niche music tastes.For example, I am a big fan of the rock / metal genres (death metal, progressive metal / Djent etc) but looking at the image, all I see are artists I do not like or have no interest in like Drake, Ariana Grande etc. This immediately isolates anyone who does like traditional popular music which is a large demographic of potential customers. Now whilst seeing bands like Sleep Token, Periphery, Tesseract etc. would definitely entice me to join Spotify if this was the case, it would also alienate a lot of other users who do not like that music.

Going with the plain background allows for a broader stroke across the entire demographic without alienation and as a side benefit gives a less distracting, cleaner, minimalist design which is more attractive.

Keen to hear what others think.

Expand full comment
Aug 10·edited Aug 10

That was my thought exactly so happy to see your comment! So much more eloquent than I would have said it too!

Expand full comment

I like this 3rd consideration posited by CHY -- and also bristle at Lee's use of the term "boring" vs simple, minimal, streamlined or "less distracting" -- I am guessing that it is used in the title to facilitate engagement on a more emotional level / click-bait -- but I feel the thinking / rationale here by Spotify is likely an emphasis on the needs articulated by @felixlee's article - none of which fall under the desire to "be boring" and your idea / theory here as well.

...It would be great to get some commentary from Spotify as a follow-up to the article to clarify!

Expand full comment

about experiment that Spotify ran:

- by win what exactly it means, do you have any numbers? Like, Variant B leads to more signups by 25%, something like that. Or it performs better by 0.00001%?

Expand full comment

It might work for Spotify because their brand is so well-established.

How about the apps just get started that nobody knows, elements such as images, illustrations, and videos is important especially for convincing users to make a conversion (in this case sign up)

Expand full comment

Take every source of inspo from big brands with a grain of salt because most of them won't work in a lot of contexts.

But with that being said, packing too much content can also hurt how users interact with your product. Minimize the pizzazz to the barest minimum, especially if it's getting in the user's way

Expand full comment

Agreed. One of those most well known and frequently downloaded apps in the world is not a realistic source of inspo for how the rest of us should or shouldn't add visual flair or moments of brand awareness.

Expand full comment

We shouldn’t accept boring design if it just lifts some number. Where’s the soul? Designer’s take?

Maybe there’s a 3rd design option that could perform even better. Simple and boring aren’t the same thing too. If you’re a young designer reading this, please try to bring forward a personality in design. It shouldn’t always be just made to increase numbers, also make sure that you never compromise with UX.

Expand full comment

Do you have any insights on whether there has been more success in converting new users or recurring users or both?

If the data shows that new user conversion was more successful, I'd propose an additional hypothesis:

"Users who didn’t identify with the artists displayed on the screen are less likely to convert."

Expand full comment

As a user, with version A I see "songs", and with version B I see images of artists I don't listen to or align my music preference/identity with - that's not me = I don't want that. I wonder if another background would have worked better - either a lot more artists so it's almost a pattern or other images related to music - album covers, notes, etc.

Expand full comment

How do you convince clients to go with plain,boring design? While most of the clients would definitely go with version B, as it might seem more exciting and visually enriched to them. I usually go with so called boring and simple designs to focus on functionality and usability but the feedback from client is usually based on the UI part making it more attractive to them and to the users, so it seems quite challenging.

Expand full comment

in terms of version B :) it's nice, but I would not want to solve which specific artists should I choose for this background, should be Drake the first-one? or Kanye West... and why not another genres, Metalica or Tiesto :D ... I can imagine that some users might be discouraged by this... acch, why Drake again? :D will this platform favour rap artists? ...

so it is not only distraction, but also misleading too

Expand full comment

I used to work in marketing at an SVOD channel and created a similar graphic for our sign up page. We never split test the two against each other properly. But now I’m wondering what the results would have been if we did 🤔

Expand full comment

I did choose version A. Because I think it was much less distracting for me. My choices were clear.

However, I was wondering, in version A, would it come across like the entire page hadn't been loaded yet ? since we're also quite used to apps that are driven by imagery.

Expand full comment

Any data to back? That boring is better?

Expand full comment