Optimizely is one of the leading A/B Testing tools on the market. With Optimizely, you can test anything from headline ideas, CTAs, colors, pricing algorithms, and more. Optimizely lets you control how many people see your experiments and then measures their responses by segment.
 
Similar to Tealium, Optimizely is a commercial-grade solution with multiple pricing plans. Users can choose between Optimizely Web and Optimizely Full Stack, and each plan splits into 3 plans: Essential, Business, and Enterprise. 
 
Optimizely is analysis tool-agnostic. Whether you’re using Google Analytics or Adobe Analytics, Optimizely is impartial to either one. You can use your reporting tools to dive deeper into the statistics generated by Optimizely. 
 

Key Features of Optimizely

 
Optimizely supports multiple platforms, ranging from simple websites or single-page web apps (SPAs) to complex applications. One of the reasons it works well with any type of environment is that Optimizely has server-side optimization. The full-stack option allows you to test many areas you can’t test on the front end. 
 
Optimizely is able to test even in conditions difficult for other solutions, such as applications where the page does not reload but instead instantly displays the new view by using JavaScript (SPAs). 
 
Let’s say for example that you have an e-commerce website, and you want to test a product listing page. You want to test if changing the category order will have any impact, and if so, which one should be placed first. 
 
A test that is conducted solely on the front end can be extremely difficult because of many elements on the page. And, if you’re manipulating website elements on the front-end, it’s easy for the user to see the content or other elements of the page flicker. 
 
With full-stack, you’re able to do the test on the backend, which is a completely different story. The backend developers can call Optimizely, and the tool can choose if users should see version A or version B, and then report back to the developers. By the time the user gets the test, it’s smoothed out and the user won’t know the difference. 
 
Another interesting feature is Feature Flags. Let’s say you want to try out a new feature on the website but you’re not sure if that feature is going to work for your audience. With Feature Flags, you can turn certain functionality on and off without deploying new code. This allows for better control and more experimentation during the full lifecycle of features.
 
The idea behind feature flags is to build conditional feature branches into code to make logic available only to certain groups of users at a time. If the flag is on, new code is executed. If the flag is off, the code is skipped.
 
If you rolled out a feature to 100% of your audience and it fails, it would have a huge impact and you’d have to roll back out of it. Optimizely allows you to ramp up that feature. 
 
If you have a new feature, you can start deploying it at 10% of your audience, analyze it, see how it impacts your metrics, and then increase the percentage over time. If the feature is a complete failure or has a big error, you can simply turn it off altogether directly from Optimizely. 
 
For example, let’s say you’re using the feature of a Chatbot. All the features and the code of the Chatbot are in your app. You’re making a call to Optimizely and asking if the tool should display the Chatbot. If yes, Optimizely will determine in the backend the percentage of your users who are going to see the Chatbot. 
 

What Does an Optimizely Implementation Require?

 
Optimizely is an open platform in terms of measuring your tests. Essentially, it’s main specialty is A/B testing, so it doesn’t have a complete analytics tool. Optimizely puts a lot of emphasis on integration, so you can integrate it with Google Analytics, Adobe Analytics, or your reporting platform of choice.  
 
If you have just a website, not a mobile app, the implementation is a tag, and you can do a lot of the changes directly within the Optimizely UI. There’s no development effort necessary for a traditional website. When you want to do a full-stack implementation with feature rollouts, you’ll need help from your developers. 
 
Another great thing about the tool is that it can help scale your testing program. If you start with the basic version and run simple tests, but in time you need more, you can use Optimizely for more advanced implementations. 
 

When You Should Implement Optimizely

 
In order for a company to use Optimizely, they first need to have enough traffic. A small business with low traffic might have difficulties gathering enough statistical significance to make the most out of Optimizely. 
 
Decent analysis capabilities are also a must-have before implementing Optimizely. You generally need to understand your digital property, audience, and behavior of your users.  use data to guide any changes to your digital property in order to optimize the effectiveness of your business, such as revenue, user experience, or simply improve your digital property.
 
If you need help with better understanding your digital property and how to optimize it, we can assist you in every step of the implementation. Take the first step toward becoming data-driven, and transform the way you do business by leveraging data for every decision you make.

About the author

Jose Bergiste

Jose Bergiste
Jose loves all things analytics, software development, and product management. His wide range of experience from small startups to large Fortune 1000 companies has helped him gain a broad and deep knowledge of the industry. Jose’s passion is to help companies realize the full power of data and analytics to maintain and develop products that customers find delightful.

New eBook Release: A Guide to User Journey Analysis

Related Articles