Facebook Ab Testing Open Source

We have open sourced a full-stack JavaScript application that automates the A/B testing of advertising creatives on Facebook. Whether you’re a startup or an established brand, our open source code will enable you to control advertising spend and boost results by testing multiple variations of your creatives with different messages, images and videos. Set up within minutes, with no IT required.

Facebook is one of the best places for testing targeted audience for your business. If you are running a small business and does not have enough money to test different audience, this article will explain how you can use Facebook to create simple A/B tests.


Conductrics is my favorite experimentation platform. I like the CEO, Matt Gershoff, and he’s been an excellent industry mentor and friend to me. He is also one of the most thoughtful leaders in the analytics and optimization space, so you can trust the direction of the platform.

He has also built a really great piece of software that, in many ways, was far ahead of its time and definitely ahead of most of the others on the list.

Power/feature-wise, it’s the best you can get. You can run A/B tests (client or server side), multivariate experiments, and bandit tests, all from a WYSIWYG visual editor (if you want). You don’t need to install a script until you set things live, as you can use the browser extension to edit and preview variants.

Additionally, the tool lets you run predictive targeting to exploit different segments of your audience. Basically their AI/machine learning will discover which combinations of visitor traits respond best to each of your content variants, automatically.

The future of adaptive experimentation with Ax and BoTorch

Over time, we will refine the beta releases of our software, expand the set of available algorithms, and provide out-of-box integration with popular scheduling software. We look forward to working with the community to add user-contributed modules to Ax and BoTorch to further improve the platform.

Research areas we are particularly excited about include high-dimensional Bayesian optimization and multi-fidelity optimization. We also believe there are significant opportunities for improving the performance of numerical optimization of acquisition functions using novel parallelism-aware solvers. We plan to further explore these as well as other new features.

Used in tandem, Ax and BoTorch significantly accelerate the process of going from research to production, and we hope it will inspire new use cases for adaptive experimentation within the broader community.

Both Ax and BoTorch are available now, so engineers and researchers can start using them today.


If you want almost all of the features of Optimizely (with the addition of truly wonderful customer support and a free CXL Institute membership), but at a fraction of the price, Convert is the way to go.

For starters, the interface will feel familiar and intuitive to experienced experimenters. For newbies, too, it’s not a cumbersome tool to learn. It’s got a great WYSIWYG editor, and of course all of the typical HTML, CSS, & Javascript editing tools that serious optimizers use in the testing process.

Convert has all the typical functionality you’d expect from a client-side A/B testing tool: A/B tests, split page tests, multivariate, and personalization.

What I like most about Convert is the founder and the team clearly truly care about the industry and their customers. They’re much more transparent about their features and the market, and I find the lack of marketing spin quite refreshing (see: their feature comparisons).

Plus they offer a free 15 day trial, so you’ve got nothing to lose from checking it out and seeing if you like it. I’d recommend those currently only looking at Optimizely or Adobe to take a look and see if Convert.com might meet your needs as well.


Kameleoon is a great experimentation platform hailing out of Paris, France but serving customers everywhere. They are known for their powerful personalization capabilities, but you can get all the basics done as well (AB, MVT, etc.).

As part of their personalization offerings, they have predictive targeting (let the machine find proper segments to exploit), business logic targeting, email personalization, content personalization, and segment analysis. They also have a feature called KCS (Kameleoon Conversion Score) that weighs the relative probability that someone will convert (likely a productized version of propensity modeling).

It’s one of the more robust personalization packages, and top it off with a trustworthy A/B test delivery system, and you’ve got a great tool.

Seeing as personalization is one of the easier things for marketing leaders to talk about and imagine, and one of the harder things to actually implement in a meaningful way, this platform could help you walk the talk.

Ax: an extensible platform for adaptive experimentation

Ax provides easy-to-use APIs to interface with BoTorch, along with the management necessary for production-ready services and reproducible research. This allows developers to focus on the applied problems, such as exploring configurations and understanding trade-offs between objectives. Similarly, it allows researchers to spend more time focusing on the building blocks of Bayesian optimization. At Facebook, Ax has been broadly applied by engineers who do not have extensive experience with machine learning, as well as by AI researchers.

The figure below illustrates how Ax and BoTorch are used within the optimization ecosystem. At Facebook, Ax interfaces with our major A/B testing and machine learning platforms, as well as simulators and other types of backend systems, requiring minimal user involvement for deploying configurations and gathering results.

Ax enables developers to create custom optimization applications, or optimize on an ad hoc basis from a Jupyter notebook. New algorithms can be implemented using the BoTorch library or other applications. Ax provides a framework for dispatching configurations to, and querying data from, external systems used in evaluation of configurations.

AB Tasty

AB Tasty is another France-based experimentation platform, this one very powerful and feature rich.

For marketing teams, they’ve got your basic client-side A/B testing and multivariate experiments, as well as personalization and social proof notifications for online retailers.

For product teams, you’ve got robust server-side experiment delivery, feature flags, and progressive rollout. They recently launched a whole new “suite” for product teams called Flagship, marking their seriousness in providing full stack experience optimization capabilities and supporting product teams and engineers.


The best free, open-source Facebook ad testing tool that helps you to plan, manage and optimize your Facebook ad campaigns easy as never before.


No Comment.