5 ms·
Not so be mean but Songkick is on a one year lowest on compete.com: http://siteanalytics.compete.com/songkick.com/ http://siteanalytics.compete.com/songkick.com
by iaskwhy 17y ago
Not so be mean but Songkick is on a one year lowest on compete.com: http://siteanalytics.compete.com/songkick.com/ http://siteanalytics.compete.com/songkick.com/
I believe the idea is nice but I'm not sure if anyone can survive referring ticket sales online.
- ig1 17y agoOnce you have a sizeable customer base, transitioning to being a direct seller and getting a bigger cut seems the obvious route to take.
- pclark 17y agothey're well funded. If they can accumulate data about consumer behaviours and music trends that makes for a really valuable data asset, no?
- iaskwhy 17y agoI'd say all data has its value. If I were to compare their data to, say, Last.fm, I believe Last.fm's data is far more important/valuable for music trends. From Songkick, I think it would be interesting to understand if people buy tickets online or not. I believe it's too expensive but as soon as tickets start to be digital only that will be a non issue.
- jedc 17y agoI'm a big fan of Songkick, myself. I used to subscribe to a number of newsletters to try and find the gigs I wanted to go to. Now I get an e-mail from Songkick whenever an artist that I listen to is in town, and I certainly do use their links to go and buy tickets. It's been really useful for me. I don't think the value is just in the listing of the gigs, but how they've combined it with tools to learn who you listen to (aka track), and the social media connections (capturing tweets while you're at gigs, etc.). So while Last.fm has a more valuable dataset, Songkick as a whole has just as compelling a value to me as a concert-goer.
- ian 17y agoLast.fm's data is amazing, and definitely more valuable than ours for predicting trends in recorded music. We focus 100% on live music though, so have some interesting information of our own on the concert industry - for example which tickets are selling fastest. One interesting stat I just discovered relating to your point is that 90% of tickets are now bought online, which is up from something like 40% a few years back.
- iaskwhy 17y agoYou mean 90% of all the tickets sold for a concert? That's pretty different from my reality (which is okay because I'm from a small european country where I pay, at least, 50% more just because I buy the ticket online which is the only data I've known about this until now). Just to get this clearer, I commented on this because I'm a fan of Songkick and have been working around the music industry since I was a teen as a journalist first, a founder of a record label and also a promoter so while I know my data is pretty accurate I also know how limited it is geographically. I know this business in the UK is much more similar to the US than to europe in general too. That's it, keep up the good work!
- axod 17y agoI think compete have become the least accurate of the stats websites. Alexa and Quantcast seem to give a far better idea of what is happening.
- ian 17y agoI'm not really sure how Compete could be getting that from our stats. We've actually seen some nice growth over the past year. Maybe I should just get our site quantified and put it all out there via Quantcast. Alexa is usually the most accurate I've found.
- axod 17y agoyeah Compete is just crazy+depressing. For Mibbit it shows a downward trend for the last 12 months: http://siteanalytics.compete.com/mibbit.com/ http://siteanalytics.compete.com/mibbit.com/ However, the actual data, and the quantcast version is an upward trend, with traffic doubling: http://www.quantcast.com/mibbit.com http://www.quantcast.com/mibbit.com Maybe compete have got some issue with all their estimations that means everyones site has dropped over the last year. I don't trust anything from compete though. Quantified is definitely the way to go as long as you don't mind having the exact stats out there for all to see.