Usage statistics for the last.fm normaliser

We've just knocked together a stats page for our last.fm normaliser application, that rejigs your charts based on an estimate of how long you have actually spent listening.

On top of the usual right-brained joy of graphs and tables of data, there are a few points that I thought were particularly interesting:

  • We've served up over 133,000 normalised charts over the first six months. We're all pretty pleased with that. Just goes to show, you get out of life what you put in, etc.
  • Nearly half of those charts have been served up directly via the "API" in XML format. That's a lot more than I expected. I think this is the final proof, if it were needed, that exposing this kind of data in machine-readable format is something that people are interested in. However, it also means that nearly 60,000 hits aren't being recorded by the lovely Google Analytics application we normally use for this type of thing. It also means that nearly half the traffic doesn't go anywhere near the Google Ads we've added to the site to try and make a bit of beer money. Ad income doesn't even come close to covering our current addiction to seeing Dan & Scroobius perform. Not a problem, because we're not relying on a revenue stream from this app, but noteworthy.
  • People overwhelmingly use the default settings. The application lets the user choose between artist or album charts for different time periods, but the default options of "top artists" data (94%) and "overall" chart (78%) are by far the most popular.

We'd love to know how these XML feeds are being used, if anyone feels like dropping us a note.

© 2018 Associative Trails Ltd. 4m@