I love this graph and frequently use it in lectures. It shows the growth in traffic at a university web server, from 1994 to the present. It represents a classic growth curve, and traffic was unaffected by the stock market crash of 2000, as you would expect from an academic institution. In fact, the real take-off in web use began in 2001.
I ask for an explanation of the pronounced third quarter spike in traffic followed by a fourth quarter dip. (I do this to test that people can relate web server requests with patterns of human behaviour). The answer is that the university year traditionally begins in October, rapidly followed by the quiet period in December.
Most organisations would be very secretive about such data, so it’s a credit to the University of Cambridge that it’s publicly available. (I was a student there, but long before the web era. I even paid someone to type out my dissertation; that’s how long ago.)