I've used data from the Netflix platform and the Power BI tool to put together a dashboard of my audiovisual consumption over the past 5 years.
From government data to how many movies you've watched on a streaming platform, pretty much everything is available (or it can become available if you know how to extract) in database form to be transformed, analyzed, and viewed.
Motivated by the ease of downloading the data from everything I've watched to this day on Netflix (if you want to do the same, just click here and, in the footer, click "Download everything") – and why not say, understood too – I decided to mount an interactive data view par discovering… well, to reaffirm that I am permanently and consistently addicted to Rupaul's Drag Race. 😛
I created my Netflix account in 2015, like a lot of people, excited about the idea of diminishing the exhausting work of searching for and syncing subtitles for movies I downloaded by torrents (and that was often the wrong movie – or a horrible version with subtitles in russian stuck — or, at worst, a virus).
The truth is that Netflix buried my willingness to watch movies, replacing it with tireless marathonseries that I've often watched and re-watched.
Downloading all this was as easy as playing in the 14th season of Grey's Anatomy and although I was slightly frustrated with the superficiality of the data made available (a .csv file with 2 unique fields: the title and date of what you saw), with a little mode data lagem it was possible to extract some interesting insights.
There were 1,147 titles seen and of these, only 67 were films. And I found this by analyzing how the streaming giant makes the view data available – the : it is used to separate the name from the series of the season, and the same : separates the season from the name of the episode. Using a delimiter separator it was easy to find that where the "Season" field was empty, it meant that only the title was filled and, consequently, this entry was a film and not a series.
Without any pride, I admit I've spent 11 hours watching Dark or 8 hours stuck in La Casa de Papel. I found that between 2015 and 2016 my TV consumption was not regular but when I stopped to turn on the TV, the thing got serious: in June 2015 I spent 59 hours watching Rupaul's Drag Race and Mad Men – in July 2016 were 57 hours dedicated to virtually only Orange Is The New Black.
My last 2 years on Netflix have been less full of passion: you don't see so many access spikes, but the volume of usage has increased.
It's quite true that for me (and for many other people, I think), Netflix is the bedtime companion, so it stays on for less hours, it's present in pretty much every day – proof of that, taking weekends where the number of hours watching something is naturally much higher, there are no big differences in consumption between days of the week… between Tuesday and Friday there is only 2% difference in episodes/movies seen.
However, thinking about extracting this data by looking at a spreadsheet or cutting all this in Excel would be too work for a mere curiosity. For just over a year I have been working with the powerful Microsoft Power BI dataviz tool, where it is possible in a very intuitive way to model data and present it graphically and interactively quite easily. For study and improvement purposes, I decided to compile this information and make it available in the dashboard you see below: