Variant Randomized Controlled Experiment (RCE) Container, Version 0.5

Share Article

Variant Announces Immediate Availability of Its Randomized Controlled Experiment (RCE) Container, Version 0.5 (Beta)

Variant Logo

Variant Logo

Control-Test Any Change To Your Web Application

Variant is pleased to announce Beta availability of its Randomized Controlled Experiment (RCE) Container software. Variant RCE Container is a radically new approach to what is traditionally referred to as “A/B testing,” doing away with the currently prevalent technique of client-side variance injection via a vendor’s service API in favor of a novel full-stack instrumentation technology. This approach is a fundamental rethinking of the problem space which enables instrumentation of arbitrarily complex changes and for the first time makes it possible for enterprises to control-test every material change to their interactive applications.

The need to control-test any new application feature has long been recognized, yet none of the existing A/B testing tools come close to supporting it. Traditional A/B testing tools are strictly optimizational — they use the browser-side variance injection technique and can only address top 10% of the technology stack of a typical Web application. While this may be good enough to address some optimizational tweaks, it does nothing to help control-test broader, evolutionary changes.

Variant’s principal features:

  • Novel Experiment Definition Model (XDM) — a self-contained abstraction, comprising a complete set of entities, relationships between them and operations on them that are required to fully describe a set of RCEs operating over a computer application.
  • Full stack instrumentation — Variant is capable of instrumenting any change, no matter how complex.
  • Platform neutrality — no assumptions are made about the technology stack on which the target application is built. This makes Variant ideal for experiments that run on heterogeneous technology stacks, e.g. an experiment may span a Web application and a native mobile app.

Separation of experiment instrumentation and experience implementation — the application code need not be concerned with the details of the experiment instrumentation, while Variant makes no assumptions about the application under test.

Full support for concurrent experiments — Variant is able to completely define the entire variance space across all concurrent experiments, including hybrid variants.

Share article on social media or email:

View article via:

Pdf Print

Contact Author

Igor Urisman
+1 (510) 205-3525
Email >
Visit website