Enterprise grade A/B testing features at self-service prices.

Convert Signals™ is a system powered by Convert Intelligence that tracks micro-frustrations on websites and combines them into patterns pointing to the presence of potential bugs, UX/UI glitches, and other technical issues. 

Convert Signals™ offers a privacy-friendly selection of the most impactful issues in the form of session recordings at Locations and in Reports (by variation), helping you find key (but anonymous) user interface and usability problems per device, page zone and by other dimensions.

Quick refresher: Multi-arm bandit isn’t a conventional A/B test. It is a reinforcement learning framework where more traffic is dynamically or automatically allocated to better performing variants, in real time, to maximize immediate gains. 

Ideal for promotions and offers, where learning what worked may not be worth the trade-offs of hitting statistical significance, and quick, fast wins are preferred.

We’re giving you the option to choose from three MAB algorithms: 

• Epsilon Greedy ⓘ
• EThompson Sampling ⓘ
• Upper Confidence Bound (UCB) ⓘ

Use data locked in your customer data platform (CDP) of choice to segment audiences and trigger tests.

Use 9 goal templates to help gauge the success of your variants. Set up advanced goals with 40+ filter conditions. Or import goals from Google Analytics. Set scroll goals and click goals.

Build tests directly in our JS editor—which gives you full acess to your site’s JS, jQuery and CSS. Plus, create tests based on JS event pushing, so you can track and account for user behavior.

Sources, channels, and geo-locations that underperform, don't need to stay that way. Segment off users by historical behavior, third party input using Data Sources, cookies, or JS events. And, give underperformers the boost they need.

From CTR, AOV, RPV, to ROI—compare any of your KPIs with high confidence, and low costs. We stop tests at the right time: once they reach the capped visitors, statistical significance, or your fixed time window. So you save money, and can test something new.

Tackle anomalies head-on with both Winsorizing (adjusting outliers based on percentiles) and Capping (setting fixed limits). Whether you’re analyzing unpredictable traffic or refining high-stakes tests, this ensures reliable data.

And now? It works seamlessly across Shopify, APIs, and Webhooks—making outlier detection available everywhere.

Even big tech firms see Sample Ratio Mismatch creep into 6%-10% of their experiments. SRM (or the instance where the control & the variant in your experiment do not receive equal split of traffic) is such an important validity threat that most experts consider peeking for the purpose of detecting SRM an acceptable risk.

In Convert reports, you can now choose to be notified of SRM when it occurs.

When you've run an experiment but due to external factors, one or multiple days of data have been rendered unusable, this feature lets you manually remove a specific dataset from the experiment, keeping the rest intact.

Observations and the Knowledge Base work in tandem for test lifecycle management.

Use Observations to capture what inspires you to run tests. You can append meta data to your Observations, covering site area where potential tests will be run, metrics impacted, and even screenshots.

Once an Observation has informed a hypothesis and has been run as a test, it can be sent to the Knowledge Base in one click where it is stored to be filtered by the experiment result, learning, who proposed the idea and more.

Transparency, security and compliance come to the forefront with the Convert Experiences Change History. This centralized stream of activity spans all projects in your account, controls and monitors access of information in the company and serves as an audit trail during incident reviews.

With BYOID you have the ability to randomize visitors/site traffic based on IDs from your own tech stack (and data platforms). You no longer use Convert’s IDs for bucketing users into experiences. This means you have more control over cookie persistence (it is tied to the persistence mandates that govern where/how the IDs are generated at your end). And you can also tie your own insights to test results more easily.

**Only available in Enterprise plans.

Download complete projects, with all their settings. Or choose to export only certain elements like complex goals & advanced audience targeting segments. These items are exported as JSON files which can then be integrated into new Convert accounts. This is a tool to standardize experimentation workflows and for agencies, to bring their unique process to blank accounts in mere seconds.

Complete control over who accesses your Convert Experiences account. Only allow members of your organization to log in with Single Sign-On that integrates with your existing authentication layers. Ditch messy passwords. Using your A/B testing tool got easier, yet more secure.

Something you know + something you have = unparalleled data security. That’s the premise for the Convert Experiences two-factor authentication as default on every plan. With this feature you add much needed layers of protection to your testing system. For further ease of use we also offer Single Sign-On (SSO) for specific plan levels to integrate with your existing authentication layers.