One of my photos, the Empire State Pigeon (also shown to the right), was recently featured in the official Flickr Blog. The Flickr Blog has the property that it is featured on the main page of Flickr, at least for the logged in users. Now, as Flickr is a high traffic site, anything on the front page is likely to get high traffic too, and so did my photo.
I posted the photo to Flickr about 19 months ago. In that time it had gathered some 5000 views, which I found an impressive number. However, when the blog post was published, the view count surged all the way up to 30000 in just a few days! Below is the graph that shows the daily traffic statistics before, during and after the event. Click to zoom to the full size view:
The graph shows that the photo was getting a steady amount of some thirty views a day and then, on April 14, it suddenly got 11133 views, 106 comments and 468 favorites. The next day it got even more views, 12635, but way less comments (8) and favorites (40). Why?
Well, that can be explained by drilling into the referrer statistics: on April 14, only 4784 of the 11000 visitors came through Flickr.com. Almost as many, 4594, came from another high traffic site called Reddit.com to where someone had posted a link to the photo. On April 15, 9798 visitors came via Reddit and only 517 via Flickr, as by then the Flickr Blog already had new posts. Reddit users don't necessarily have Flickr accounts and thus cannot leave comments or favorite the photos.
Another difference I observed in the behavior of Flickr visitors and Reddit visitors was that Flickr visitors looked at my other photos too. Thus the aggregate views over all of my photos was a lot higher on April 14 than it was on April 15, even though that one photo got more visitors on April 15:
(And now for something completely different. If those graphs above felt dry, try the GraphJam site instead!)