Re:How Coderwall grew over 50% with an A/B test
by ssharp on Tuesday September 27, 2016 @10:48AM
(#12589967)
I don't really understand the methodology on this test. How are you getting a fair control/variant split in Google SERPs?
[ Parent
| on HN ]
Re:How Coderwall grew over 50% with an A/B test
by zck on Tuesday September 27, 2016 @10:53AM
(#12590012)
They took 20,000 pages, split them into test and control groups, and changed the title of questions in the test group from "$QUESTION - Coderwall" to "$QUESTION (Example) - Coderwall". They then measured how many more clicks the test group got compared to the control group.