The journal Angewandte Chemie recently published an article which, among other things, showed the download statistics for a chemical research article published in their journal over time:
A Communication that was published online in EarlyView on Wiley InterScience (now called Wiley Online Library) on a Friday had the highest number of “full-text downloads” that very same day. The weekend was quieter, and the few days immediately following again showed a high number of readers. After that, the interest waned, and it was briefly renewed when the issue containing the article was published online.
I’m amazed that the visibility window is narrower than I had imagined before it goes into oblivion. This is chemistry, and most biomedicine will probably follow the same pattern. Humanities articles are most probably much more long-lived.
(thanks to Derek for the tip)