One of the downsides of the under-representation of classical music in existing streaming services, is that we have very limited exposure to datasets consisting purely of classical music usage. We’re only able to access data on an aggregate level, and since classical music streaming represents about 1% of the streaming market, any meaningful insights we could’ve gained from the peculiarities of classical music lovers, are quickly pushed to oblivion by the more popular genres.
Precisely for this reason, we understood the importance of incorporating data insights into our product development cycle from the very beginning. Although still at a very early stage, we’ve already gained some interesting knowledge about how users consume classical music withinPrimephonic, which we’d like to share with you, by looking at a snapshot from the entire month ofDecember 2018.
Streaming activity over days
The above graph shows streaming usage distributed across days of the week. Sunday seems to be a clear winner when it comes to the most popular day. But music is such an important part of our lives, that maybe a daily overview is a bit too high level. What if we were to split this up into hours?
Streaming activity per hour
Some interesting patterns start to emerge. A quite obvious one is that from midnight to around 6am, there’s very little relative activity during the weekdays, with a bit more activity after hours (between 12–1am) on Saturdays and Sundays. This is reassuring, as it shows that classical music lovers are, after all, normal human beings which cherish their sleep 🙂
We can also see usage picking up after the early morning hours and throughout the day during weekdays in a pretty consistent manner. As confirmed already by our previous analysis, Sunday pops up as having higher user engagement, but what’s interesting is that Sunday mornings seem to be a particularly important moment for our users. We may be tempted to jump into conclusions — maybe classical music lovers prefer to set some time aside for active listening vs. the more passive listening behaviour shown in popular genres (e.g. background music during work, party music, etc.)? We’re not quite ready to make that conclusion though.
Take the following listening behaviours from 2 different, anonymised users for example:
Comparison of 2 different listening behaviours
The figure on the left represents a user with an average listening time per day of around 8 hours, with quite a bit of musical variety throughout the day (each color represents a unique track streamed), while the user on the right shows a completely different behaviour — indicating someone with a very clear idea of the exact tracks they want to listen to, and when. What can we learn from this? That even within this very small subset of the streaming market, there is no 1 size fit all solution and we must still provide our audience with truly unique content and experiences tailored towards their own individual preferences.
There’s still a lot to be done of course, but this sort of analysis gives us unprecedented access into a world very little explored until now.
If we can understand how classical music streaming differs from other genres, we’re in a much better position to build a truly unique product, and hopefully contribute by doing our part in keeping classical music alive as music consumption worldwide transitions into this new medium.
Want to help us build the tech that will bring classical music to the digital age?Join us!