At Mobile1st, we understand the importance of building and analyzing experiments to help our clients achieve their goals and metrics. That’s why we created FacilXP, a process that enables us to run A/B and multivariate experiments using our clients’ existing technologies and analyze the results in their analytics platform.

Why We Built FacilXP

We analyze experiment results in their Google analytics platform for all our clients since it’s predominantly their source of truth. This allows us to analyze how the experiment impacts all the goals and metrics the client has set up in their analytics platform.

We build and run experiments in the experimentation platform the client has licensed. Many of our clients chose Google Optimize as their experimentation platform because it is free, the client doesn’t need personalization from an experimentation platform, and can run up to five experiments simultaneously.

When Google announced their plans to sunset Google Optimize in September, we wanted to deploy a solution that allowed us to continue experimentation, avoid any traffic limits or caps, and continue to be free for our clients.

With that in mind, we created FacilXP. 

How FacilXP Works

Any experimentation platform needs to accomplish the following steps:

  1. Decide if a user is in the experimentation audience
  2. If the user is in the audience, flip a coin and assign a user to a variation or control.
  3. Present an alternative user experience if the user is one assigned to view the variations.
  4. Analyze which variation performed best across all of the goals and metrics.

Our process documented in FacilXP uses Google Tag Manager to complete steps 1-3. We use our clients’ analytics source of truth (often Google Analytics) to analyze the impact of an experiment (step 4). 

Who Can Use FacilXP

FacilXP was designed specifically for Mobile1st to manage for our clients. It can execute and analyze any experiment that can be targeted with Google Tag Manager, but it does not perform personalizations, which are better suited for Optimizely and Monetate.

Looking for an easy, effective, reliable experimentation solution with no licensing cost? Let’s talk.