Most major web services today have specialist data teams whose purpose is to make user behaviour data accessible by analysing it in useful ways. Here are our favourite data visualisations!

1/ At IMDB they have collected together tens of thousands of user generated ratings for all episodes of each season of particular TV series such as Breaking Bad, The Simpsons, Friends and Dexter.
It is extremely easy to see, at a glance, how the standard of the episodes varies over the course of a given season, and even from season to season.


More: http://uproxx.com/tv/2014/03/website-tracks-rise-fall-every-tv-show-graph-form/

2/ At Facebook, user data is a sensitive topic and analyses are rare. Nevertheless, they recently provided us with a study that makes it possible to see, via user check-ins in the 14 days following marriage, the average distance travelled on honeymoon by the inhabitants of various countries in the world. Who is way out ahead? The South Koreans!


More: http://blogs.leaderpost.com/2014/09/25/facebook-analyzed-top-honeymoon-destinations-based-on-facebook-check-ins/

3/ Twitter offers us a minimalist, geographically-based view of all the replies tweeted in the hour following the 2011 earthquake in Japan.


Further data visualisations from the Twitter data team are available on their Flickr account: https://www.flickr.com/photos/twitteroffice/sets/72157633647745984/

4/ Uber clearly has a very active data science/data analysis team. You can find many studies on their blog. This one, for example, shows us the volume of trips made by users between pairs of cities (those cities where Uber is present of course). This makes it easy to see that Zurich users mainly travel to Paris, NYC or London, or that Palm Springs is a favourite destination of users from NYC or the big cities of California.


Link : http://blog.uber.com/uber-jetsetters

5/ Even more impressively, and inhabiting the same world as IMDB, Netflix provides us with a ranking for the US’s largest cities of the number of views, per neighbourhood, of the different programs provided through its VOD service. For example, it can be seen here that there appears to be a correlation between the socio-professional category of a district’s population and the position of “The Dark Knight” in the Netflix chart.


Link : http://www.nytimes.com/interactive/2010/01/10/nyregion/20100110-netflix-map.html?_r=0

In the future, we also intend to provide you with user-data based studies that could be useful for developing a better understanding of social media. We have already made a start here and here, and that’s just the beginning!
If you have any ideas you’d like to suggest to us, don’t hesitate to give Professor OG a shout on Twitter @Pr_OG!

Article published by Xavier BK in SocialMedia

the 21 October 2014