Honestly, my impression of this site is that the creators don't listen to much music. I know it’s probably not true, and that it sounds pretty snobby, but for instance, segmenting by a genre as broad as Rock is too reductive. There are possibly hundreds of rock subgenres, many of which have very different audiences, with some overlap. Take Pop Rock vs punk vs grindcore vs metal vs folk rock vs whatever. And then all of the subgenres within those. Subgenreception.
IMO, community curation isn't something that will work very well with music, though I'd love to be proved wrong. Somebody said something about subreddits in another comment, and that analogy seems good. There is also another startup that collaborated with metal musicians and had them build out radio shows, which also seems promising - especially because it's addressing a very specific market.
Also, 20 seconds is way too long to listen to a song that you know you're not going to like from the first couple of seconds.
I do listen to a lot of music and I definitely agree with you about the music genres. The issue is just ensuring there are enough listeners and musicians per genre. The site is very new and we are slowly increasing the number genres as we expand. We started out with only 4 genres and are now up to 8. We look to support many more in the future.
As for the 20 second time for more casual listening I recommend listen or skip mode where you build up quick skips and can skip to the next song 3 seconds in. For the ranking mode it is very important to artists that the people ranking their music and determining its exposure give it a chance.
Allow for user groups where people can actively share songs with one group or another. From there the ranking begins. Upvotes/downvotes work, but the Facebook style of choosing a "type of like" is much more useful in the long run.
Trying to rank songs without any context is almost impossible. You'll just end up with one particular style that is popular for the entire site.
Thanks for the suggestion. Songs are ranked and listened to only within a genre so I would expect each genre to end up with its own popular set of songs.
I am definitely open to further experimentation but so far users have enjoyed the ranking process and it is easier to say I like this song more than this song than it is to say this song is good vs bad.
Also the upvote downvote system is ripe for abuse comparatively. Because LaunchSong doesn't let you link to a song to rank it gives artists without an existing audience a level playing field (because people cannot say go upvote my songs). Furthermore the elo rating system lets newer songs climb more quickly
Tried it without registering, the idea is great but the UX could be made better.
For the 5 songs to rank, for example, I only enjoyed 1 while the rest were unlistenable. There should me be a way for me to delete the ones I don't like and get new ones in until there is a set of 5 good ones worth ranking.
For the yay or nay rank I was getting good songs, but there was no way to mark them as such. To make this site go viral and enable the exposure you strive for, publish rankings and enable visitors to randomly hear new music and rank it.
The domain and the idea is great - I am sharing it with friends already.
1. Your rankings are actually what helps prevent other users from stumbling upon these same songs. The lower the ranking the less likely to appear for other users.
2. For the yay or nay, the lack of clear explanation is definitely my fault as it is a newer feature. The idea is that how much you listen to the song vs skip determines the ranking of the song. This mode intends to be lower touch for more relaxed listening
3. Would you mind elaborating on the ranking publish idea. I am always looking for new ideas to promote it, but am not sure I understand this idea.
Hey, I'm here to say simply that this is a great idea! Way to put it together. Very few posts seem to be about new digital products to help the music industry, the last few years seem to have mostly been about consolidation. Keep up the good work!
Exactly! Thanks for answering the question! We exist to help with music promotion for artists and music discovery for listeners but claim no rights to any of the songs!
Thanks for the feedback. I definitely agree with you on the design and information fronts and will be working a lot on this piece.
For functionality piece I would love more feedback. Do you have any suggestions of features you would love to see? I have received very positive feedback from musicians, but am only just starting to get listeners using the site!
I am also curious if you used the site on mobile or web. For now I'd say the experience is far superior on web!
People keep getting this problem wrong. The solution is to have bubbles/groups. You join a group of like minded people and then rank/moderate your own content in ways that are less magical.
Subreddits are a great example, or highly specific forums like HN.
Interesting suggestion! I'm working on expanding smgmusicdisplay.com. SMG's a tool used by Twitch streamers to display what music they're currently playing.
I'm designing a song played history page for streamers where their viewers can browse through their most played songs, their history, favorites etc... But I really like this idea of adding community moderation into the mix as well!
Streamers are known for having tightly knit communities. So far I've been excited about building this idea, but I ran into a roadblock where smaller artists would still not get much more exposure... But in this case, if an individual streamer likes a particular artist, that can actually result in a lot of exposure, even for small artists!
After speaking with artists I think there are several issues with this approach (though also advantages). These groups either have a few tastemakers controlling the opinions of the masses which musicians already consider a major issue. On the other side of the spectrum are groups that operate as a free for all where musicians are posting so many links that each song is not given a real chance or where musicians set up like for like rings to go scale the reddit/group vote ladder.
I also think elo rating is a very transparent system.
Does anyone have an example of an automated feed proving more useful to them? The closest I can think of is play next while listening to music on YouTube.
Great question. The site may not be clear but the reality is anything but handwavy. The higher rated the song, the higher the probability it will show up in a set of songs to rank. This means more exposure since ranking sets of songs is the primary mechanism with which users find new songs on LaunchSong!
IMO, community curation isn't something that will work very well with music, though I'd love to be proved wrong. Somebody said something about subreddits in another comment, and that analogy seems good. There is also another startup that collaborated with metal musicians and had them build out radio shows, which also seems promising - especially because it's addressing a very specific market.
Also, 20 seconds is way too long to listen to a song that you know you're not going to like from the first couple of seconds.
I do listen to a lot of music and I definitely agree with you about the music genres. The issue is just ensuring there are enough listeners and musicians per genre. The site is very new and we are slowly increasing the number genres as we expand. We started out with only 4 genres and are now up to 8. We look to support many more in the future.
As for the 20 second time for more casual listening I recommend listen or skip mode where you build up quick skips and can skip to the next song 3 seconds in. For the ranking mode it is very important to artists that the people ranking their music and determining its exposure give it a chance.
I second this, I had actually managed to create a ranking but wasn't able to submit due to the 20s minimum.
Allow for user groups where people can actively share songs with one group or another. From there the ranking begins. Upvotes/downvotes work, but the Facebook style of choosing a "type of like" is much more useful in the long run.
Trying to rank songs without any context is almost impossible. You'll just end up with one particular style that is popular for the entire site.
I am definitely open to further experimentation but so far users have enjoyed the ranking process and it is easier to say I like this song more than this song than it is to say this song is good vs bad.
Also the upvote downvote system is ripe for abuse comparatively. Because LaunchSong doesn't let you link to a song to rank it gives artists without an existing audience a level playing field (because people cannot say go upvote my songs). Furthermore the elo rating system lets newer songs climb more quickly
For the 5 songs to rank, for example, I only enjoyed 1 while the rest were unlistenable. There should me be a way for me to delete the ones I don't like and get new ones in until there is a set of 5 good ones worth ranking.
For the yay or nay rank I was getting good songs, but there was no way to mark them as such. To make this site go viral and enable the exposure you strive for, publish rankings and enable visitors to randomly hear new music and rank it.
The domain and the idea is great - I am sharing it with friends already.
1. Your rankings are actually what helps prevent other users from stumbling upon these same songs. The lower the ranking the less likely to appear for other users.
2. For the yay or nay, the lack of clear explanation is definitely my fault as it is a newer feature. The idea is that how much you listen to the song vs skip determines the ranking of the song. This mode intends to be lower touch for more relaxed listening
3. Would you mind elaborating on the ranking publish idea. I am always looking for new ideas to promote it, but am not sure I understand this idea.
For functionality piece I would love more feedback. Do you have any suggestions of features you would love to see? I have received very positive feedback from musicians, but am only just starting to get listeners using the site!
I am also curious if you used the site on mobile or web. For now I'd say the experience is far superior on web!
> 3: Highly Rated Songs Get More Exposure
..how? This is the fundamental problem.
Subreddits are a great example, or highly specific forums like HN.
I'm designing a song played history page for streamers where their viewers can browse through their most played songs, their history, favorites etc... But I really like this idea of adding community moderation into the mix as well!
Streamers are known for having tightly knit communities. So far I've been excited about building this idea, but I ran into a roadblock where smaller artists would still not get much more exposure... But in this case, if an individual streamer likes a particular artist, that can actually result in a lot of exposure, even for small artists!
Thanks for this suggestion!
I also think elo rating is a very transparent system.
Does anyone have an example of an automated feed proving more useful to them? The closest I can think of is play next while listening to music on YouTube.