> "90% of tests fail" This is a valuable point for those looking to get into split testing, growth testing, growth hacking in general. Unless there is this kin

> "90% of tests fail" This is a valuable point for those looking to get into split testing, growth testing, growth hacking in general. Unless there is this kind of understanding going in, a 90% 'failure' rate (failure to find something that works) can be quite demoralizing. Additionally, startups that hire or look for a growth hacker should be obsessed with the number and quality of tests, not just the graph magically beginning to go up and to the right. It's easy to think that anyone that gets things 10% right isn't really good at their job - don't fall into this trap with growth hackers. The good ones are testing 10 things a week, so good results will come.