Here’s a quick exercise for you to try at home. Invite 100 people with Web 2.0 startup ideas to come over to your house. Ask everyone who is building a social network to walk out your front door. Then ask everyone who is building a widget for a social network to walk out the back door. Of the 10 people who are remaining, ask them what type of startup they are building. I’ll bet $20 that it has something to do with music.
Of course I’m exaggerating, but there are a ton of music-centric sites out there today – which means there is a big opportunity to provide value-add services to these good. A good example of this is a service that use with Windows Media Player that allows me to rip a CD to my computer, and then matches these files against an online database that automatically tags the ripped music with the name of the album, the artist, genre, etc…
My idea is to create a service that takes this concept one step further – the ability to break a song down into its core components such as verse, chorus, bridge, etc.. The components could also be tagged to identify the “hook”, or the 30 second clip that is the most popular section of the song.
Scenario #1 – Imagine a company that was building an app that pulled together pictures and music into a 30 second slideshow. If a user uploaded their favorite song to use as the background, instead of pulling the first 30 seconds of the song, the app could automagically pick out the 30 second snippet that is the most interesting part of the song. Or better yet, for a 2 minute slideshow, the user could choose to have 4 30-second clips of the most recognizable parts of the songs, as opposed to having one song used as the background for the full 2 minutes.
Another use case might be for a service that allows users to send mixes of their favorite music for their friends. Let’s say I create a mix with 10 songs – instead of sending my friend a 30-minute music file (OMG! No way, I’m like totally ADD!), I can send them a 5 minute mix consisting of their favorite parts of the songs. Users could also use this service as an add-on to iTunes (or iZunes) to create a fun dance mix for their next party. That way, you can play songs like ‘American Pie’, get everyone to sing through the chorus, and then move on to the next song instead of making everyone stand around for the next 7 minutes trying to decide what to do during that not-fast-enough-to-really-dance-but-too-fast-for-slow-dancing section towards the end.
Last point – I’d argue to create a back-end service instead of a single application because A) there are so many players in this space – better to sell the pick-axes than go after the gold and B) lots of the big guys (especially the major record labels) are looking to buy their way into this space, and this service could make for a very appealing acquisition target.