Since becoming addicted to data analytics, especially data from YouTube, I have been fascinated how predictable data patterns can be and what patterns are revealed over time. Below is a screen shot of views data from my YouTube Channel from 1st January 2010 to 31st December 2013:
The arrows point out the dramatic fall in numbers of views over the Christmas and New Year holiday period every year. While the overall trend is upwards, you can also see a slight fall off in views each summer followed by sharp increases in the autumn. Another noticeable trend is that the height of the wave-like trend is getting bigger as weekend views are not increasing at the same rate as mid-week views.
Taking a closer look at the Christmas and New Year holiday period is also interesting - below are the views data from 1st December 2013 to 31st January 2014:
Early in this period on 3rd December there were 11,761 views (a daily record for the channel), but you can see a steady fall in views towards Christmas. I was fascinated to see that 2,189 people viewed my videos on Christmas Day. It is not until the beginning of January 2014 that the views figure start to climb again. If previous years' data are to be repeated, the daily views will not recover to the early December 2013 numbers until March this year, but then continue to grow until the annual summer decline.