Experiment end-to-end with Optimizely Web Experimentation and Optimizely Feature Experimentation

Optimizely Web ExperimentationOptimizely Web Experimentation and Optimizely Feature ExperimentationOptimizely Feature Experimentation share a common infrastructure and certain resources, such as Stats Engine and CollaborationCollaboration. However, the two products are often used independently by different teams in an organization.

Marketing teams typically use Optimizely Web Experimentation to drive higher conversion, increase lead flow, and boost engagement. Product teams usually use Optimizely Feature Experimentation to improve the user experience across all digital platforms.

Despite these differences, there are times when you want to run an experiment across both platforms. Doing so can keep messaging, offers, and experiences consistent for any users in a targeted audience after they are bucketed into a variation.

One example is displaying consistent pricing and packaging offers throughout the acquisition and trial process. Your marketing team might experiment with some packaged configurations for a particular vertical. When a user has seen a particular packaged offer, they must continue to see it throughout the rest of their interactions with the site, even when they move to the trial phase and the subsequent conversion process.

Another use case is mutually exclusive experiments between Optimizely Web ExperimentationOptimizely Web Experimentation and Optimizely Feature ExperimentationOptimizely Feature Experimentation. If a user is in Experiment A in Optimizely Web Experimentation, they cannot be in Experiment B in Optimizely Feature Experimentation, or vice versa.

Here is an example where you test two competing pricing and packaging offers against each other:

Your user sees Offer A and continues to see it on subsequent visits. On the third visit, they decide to start a free trial. After they sign up, Optimizely Experimentation tags them as members of the Offer A group. Any future interactions or experiments in Optimizely Feature Experimentation must continue to use the Offer A pricing and packaging.

While this user is on your website, Optimizely Web Experimentation uses a cookie to track which bucket they are in and which offer they see. Step 1 in the previous graphic represents the visit on which they decide to sign up for a free trial. Optimizely Web Experimentation creates a user object during the signup process, which includes the variation ID as an attribute (step 2). Any subsequent Optimizely Feature Experimentation experiments refer to this attribute through audience targeting (step 3) to ensure the user continues to see the same information.

If the checkout team later decides to experiment with two different checkout flows, this user continues to see Offer A, regardless of which checkout flow they see. When the user converts to a paid customer, the initial offer (Offer A) and Optimizely Feature Experimentation checkout variation (A.A) are recorded (step 4).