The rest

Festival memories, or: quantified fun!

Topics
Analytics, Festival, mobile, Open source, Ruisrock, Sharing

You still remember what you've done in Ruisrock festival? No worries, we do! This year, we wanted to get analytics data from our beloved Ruisrock application. Basically, we collected data about when and how people use the app: what features were they using, how much, and at what time of the day. We didn't go all NSA on you though, it's all anonymous. In an usual app development scenario this kind of data is used for continuous improvement of the application: if we find out that people tend to use a hard-to-find feature a lot, it gets promoted to a more accessible location; seldom used features on the other hand get demoted out of the way or even removed. It can also be used to measure general interest, engagement, and other more or less interesting indicators. In an yearly event companion app such as this, continuous development is a bit harder. I'm sure you'll all agree that statistics as such are always interesting though, so let's take a look at them! First of all, some basic stats: 6 227 unique users, 107 255 total sessions, and 3 960 total hours spent using the app. On a per-day basis, approximately 20% of Ruisrock visitors used the iOS or Android app, so there's still room for improvement! Of course, the rest were probably using the Windows Phone app, which was unfortunately left out of the analytics.

Basic statistics on application usage.
The app usage peaks during afternoon and evening, which suggests that the app was used on the way to the area and before the big gigs. After midnight there doesn't seem to be a lot of interest. Not surprisingly, the usage peak on Sunday is hours earlier, as was also the general festival schedule.
App usage by time.
Most platform specific artists on each platform.
There seemed to be some interesting disparities between the app usage on the two platforms: as you can see from the graph above, there were clear differences on when the app was used. On Friday and Saturday the Android crowd seemed to use the app more evenly during the day, whereas the iOS usage had a clearer peak; and on Sunday it was the other way around. By day, iOS folks were relatively more active on Sunday and Android users on Friday. Strange! Could the people using iPhone be interested in a different kind of music than the people using Android? As it happens, we can kind of answer that! Among other usage data, we monitored which bands people “starred”, or marked as favourite. As you can see from the table below, people across platforms agreed on the top 4 bands, but then the tastes diverged. By counting the difference of the platform specific top list positions, we found out that, for example, the most iOS specific artist was Erykah Badu, who was 15 positions higher on the artist list based on iOS users. The Android end, on the other hand, had many artists from the heavier side, such as Stam1na and Amorphis.
Top 10 favourited artists on each platform.
I'd be very interested to hear where these differences come from! My personal guess is that Android users are younger and male…er, and heavier music is more popular among them. Go figure! Given a wide enough app adoption rate, this kind of information could perhaps be used to tweak venue choices, in case an artist turns out to be surprisingly popular. The great thing is that you'll be able to try this out at your own festival next summer! We'll soon release the source for our Ruisrock application, enabling any festival to build their own companion app with minimal effort. Watch this space! I still have a couple of graphs left, reinforcing some stereotypes about festival behaviour. ruisrock-lostandfoundruisrock-firstaid