When we first implemented Optimizely, I was the only one using it to run tests. As I gained momentum and other members of the marketing team, management and customer service became familiar with the platform and possibilities, they began contributing to the testing backlog and even running tests themselves.
While it's great they offer a free version, technical support for more difficult experiment set up is unavailable unless you are an enterprise level member.
Optimizely has a synchronous tag, which means it cannot be added to a website using Google Tag Manager (>90% of all sites using a TMS use GTM so this affects most users.)
I'm sure most of us mathletes would appreciate a more in depth explanation of the statistics used in their new Stats Engine. I did a lot of digging when it was first released and wasn't able to get answers from their support team. It's great to be able to test more rapidly but understanding the maths behind the decision engine is paramount - or your essentially flying blind.
Optimizely is great for testing on sites with moderate to heavy traffic. It doesn't seem to be as effective with lower trafficked sites because the sample sizes just aren't there. I'd recommend going with a User Testing-type platform to perform CRO on lower trafficked sites.