An experiment in A/B Testing my Résumé

Paul Butler – July 1, 2010

Résumé variations with different grades

Finally, I wanted to test whether it pays to include social media links on the résumé. I chose GitHub, LinkedIn, and twitter as the links to test. GitHub was an obvious choice because it emphasizes my free-time projects. LinkedIn seemed like a good one to test, given that it is for professional networking. I chose twitter as another variation because I was curious to see what the reaction to a more personal social networking site would be. All résumés linked back to my blog as well. Finally, I had another resume which linked only to my blog, as a control group.

Résumé variation with a link to GitHub

In all, these three variations resulted in 16 unique résumés. Fortunately I didn’t have to create them all by hand. I was already using LaTeX for my résumé, using one of the elegantly typeset templates from The CV Inn as a base. I simply threw my latest résumé into a Mako Template and wrote some python code to spit out the 16 possible variations of the LaTeX code. Then I used pdflatex to create pdf files. Since I was putting the résumés online, I made a landing page. To keep things simple, the landing page was just an image version of the résumé with a link to download the pdf, and just enough CSS to look presentable.

One of the Google ads I ran

Scroll Rate by Length

The short résumé also resulted in more downloads and blog views, but not enough to be statistically significant with the amount of data I collected.

Blog view rate by link

I created a heatmap-like visualization from the relative significances of each link to each other. For example, the upper leftmost cell means that it is 97.2% likely that if a sufficiently large group of people were exposed to each of the LinkedIn and blog-link-only versions of my résumé, the group that saw the blog-link-only version would visit my blog more. Jesse E. Farmer has written more about the details of how this is calculated.

Blog View Rate by Link Heatmap

Oddly, the effect was reversed when you consider downloads rather than blog views. The résumés without any social media links were far less likely to be downloaded than those with. Even a résumé with a twitter profile did better than one without, though not by enough to be statistically significant.

Download Rate by Link

Download Rate by Link Heatmap

The additional links also reduced the frequency of readers scrolling to the bottom of the page.

Scroll Rate by Link

Scroll Rate by Link Heatmap


To be notified of new posts, follow me on Twitter.