I love clickstream data – the data we collect as people use the site.  We know – anonymously – the search terms used and articles viewed.  We have just developed the ability to ‘mine’ this data; nearly 100 million items it!

Below are two images, both taken from searches for hypertension.  The first is for all the data we have from 2009 (tens of thousands of searches for hypertension) and the second is from 2015 only (still thousands).  What we’ve done is taken these searches, removed the term hypertension and created a wordcloud of the remaining words.  These words being the additional terms users used when searching for hypertension.  So, if a user searched for hypertension pregnancy, we simply chart the pregnancy term.

Searches since 2009
Searches in 2015

They look fairly similar but from a cursory analysis five terms are more prominent in 2015, those being:

  • placebo
  • aspirin
  • patient
  • enalapril
  • obesity

Interesting for sure, but more than that?

Advertisements