What do you like best about Optimizely Web Experimentation?
Optimizely is being used by our dev team as well as our client's marketing team to conduct A/B tests and Personalization test on client e-commerce website to help increase conversion, average order value and other key metrics, as well as to quickly and easily make cosmetic changes until the development team can catch up and implement them on the back-end. Prior to using Optimizely, we had other tool (should I mentioned it or not; not sure) but it had so many issues working with new technologies, very hard to use by marketing team and super costly. This tools solve all of these issue. A few key features that really attracts us to work with this tool.
1. Optimizely's user interface and WYSIWYG editor are easy to use and don't require coding experience, but allows you to get dirty in JavaScript and CSS if necessary. This makes it easy for anyone to pick up and start working with it. One of the great feature is extensions where in a Dev can create sort-of prototype template for certain feature(example Popup/Lightbox) and any non-dev person or marketing team can quickly utilize that by just feeding required information (Popup/Lightbox text content). So easy and quick. Reduce the pain for dev and no support required.
2. They provide excellent and attentive support. My team have called in to their support line a few times to raise issue or ask questions on understating of tool in-depth and they were always very easy to work with and helped my team solve their problems.
3. Test results are clear and easy to read. They have great graphs and tables to help you see how different variations performed, how likely a variation is to beat the baseline, and how well that stacks up against other variations. Even with multivariate tests, the results are clear and concise.
4. The ability to add custom JavaScript and CSS per test is very convenient and makes it easy to make more detailed changes right from within Optimizely. The concept of "Shared Code" specific to test or globally is an awesome feature. Dev can split their code to keep at right place and hence load efficiently, reduces code size, execution time and device load.
5. Traffic allocation options are convenient and allow you to easily split traffic to different variations.
6. Prioritization between two experiences is also a nice feature and helps when two experiences are trying to manipulate same element to decide the order of changes taking place
7. Optimizely integration is super easy. With one line of JavaScript, you are able to install Optimizely on your website and start testing. Minor configurations may be required within the tool based on product specific requirements but never takes much time.
8. Doumentation is excellent and very helpful. Review collected by and hosted on G2.com.
What do you dislike about Optimizely Web Experimentation?
Well, this tool cover mostly all the feature you require to run a test quickly. But with new technologies and product architecture, there is always a scope to make little-little improvements. Here are some of my observations so far.
1. Depending on the page scenario, it sometimes loads pages very slowly.
2. Linting feature in this tool for JavaScript and CSS is excellent. But sometimes, it takes little time (~15-30 Sec) to load large code and switch between CSS to JavaScript editor and vice-versa. Since developers tend to have their boilerplate setup with all linting and validation, there should be a toggle to ON/OFF linting feature to increase load time.
3. Concept of profile scripting wherein we can divide an audience into two buckets randomly or based on some JavaScript logic. A feature like this will help businesses to run two A/B test parallel due to their reasons. This kind of requirement is not rare and keeps coming often. Ideally, running two A/B tests in parallel is not recommended. Review collected by and hosted on G2.com.
