But about a month ago, I decided to actually look at the data about it and test my assumption. Over the course of the next few weeks I gathered 200,000, random, bit.ly-link-containing Tweets. I used the bit.ly API to calculate a click through rate clicks on a link divided by number of followers of tweeter. And then I analyzed the relationship of the link’s position inside the Tweet and it’s CTR. I figured the best way to visualize this would be through a heat map.

via New Twitter Data: Optimal Link Placement for Clicks | Dan Zarrella.