Usability Enhancements

Keyboard Accessibility

Users can now use tab-based navigation to access Split’s login page, change summary modal (including approval flow from email), and definitions tab. This is supported on the Edge, Firefox, Safari, and Chrome browsers.

Usability Enhancements

Switch 

Switch is an in-app AI assistant designed to streamline the use of the Split product. It offers multilingual support, rapid responses, and knowledge-based assistance by utilizing our public documentation and blogs. Switch makes it easy for all developers to get the help they need, without ever leaving the Split interface.

Feature Experimentation

Sequential Testing Update

Sequential Testing will now be the default statistical method for all net-new organizations using monitoring and experimentation. 

The minimum sample size for Sequential Testing has been reduced from 200 to 100. This allows users to get statistically significant results, faster.

Dimensional Analysis Update

Users can now create up to 20 dimensions with 20 values per dimension. This gives users even more flexibility when doing a deeper analysis of their feature experiments or releases.

API Enhancements

Different Access Levels for APIs: Roles and Scopes for API Keys 

Define specific roles and scopes for Admin API keys. Restrict access to resources at the organizational, workspace, or environment levels. This gives admins more flexibility when granting access to Split’s Public API.

SDK Enhancements

PHP in-memory (PHP Thin Client SDK) updates 

Split’s PHP in-memory SDK now supports event tracking via the Track () call, getting treatments with Dynamic Configurations, and retrieving information on cached flags via SDK Manager.

Monitoring

Feature Flag Significance Alerting

With Feature Flag Significance Alerting, users can now receive notifications when a statistically significant difference has been observed between two treatments on their flag’s key metrics. Feature Flag alerting is enabled automatically for releases with a percentage allocation. This enables users to make fast, accurate release decisions in a timely manner.

Out-of-the-Box Metrics, Browser RUM Agent

Split now automatically creates metrics for any events being auto-captured by the Split SDK (Browser RUM Agent). This reduces the manual effort of creating metrics and allows users to easily calculate their engineering and performance metrics.

Integrations

Split-Segment Integration Update

The orginalTimestamp precision has been updated to go from seconds to milliseconds when sending impressions to Segment. This update makes our timestamp field more consistent with the precision Segment uses.

Feature Management Console 

Feature Flag Editor Enhancements 

The feature flag definitions tab has added two minor UX updates to the editor flow. The treatment section will now automatically collapse once a flag definition has been created/updated. Also, users will be able to view treatment information via the environment cards on the left. These enhancements help the user understand which treatments are available in their environment and guide them to the targeting section.

SDK Enhancements

RUM Agents – Web & Android

• Users can now automatically capture event data in Split for their Web and Android SDKs. This eliminates the manual process of sending events to Split, allowing users to quickly set up metrics and alert policies.

Monitoring

Custom Analysis Time Frame

• With Custom Analysis Time Frame, users can now analyze across date ranges regardless of any changes made to the feature flag definition. This enables better flexibility when analyzing results.

SDK Enhancements 

PHP in-memory (PHP Thin Client SDK)

• Split now supports running PHP locally using the Split Daemon (splitd) process to store and maintain feature flag data. This eliminates the need to use Redis & the Split Synchronizer while using our PHP SDK and provides a simpler set-up process.

SDK Enhancements

Instant Feature Flags

• To reduce the latency of updates and increase the reliability of SDKs, feature flag updates will not require any network request.

Integrations

SDK@Edge, Split-Vercel Integration 

• Split’s integration with Vercel’s Edge platform provides teams with the ability to incorporate feature flags and experiments into their Edge application and workstreams. Streamline Split data into the Edge without the extra network requests to retrieve config data.

Feature Management Console

Viewer Role

• Admins can now assign the role, Viewer, to users. With the Viewer role, users can now be assigned a role that allows them to view data and objects within the Split application without the ability to make modifications. This will give admins more flexibility and control when assigning roles.

Experimentation

Sequential Testing

• Sequential Testing is a statistical testing method that allows users to obtain statistical results without the constraint of an experiment review period. This allows users to receive faster experimentation results, so that they make informed decisions about releases, quickly.

Monitoring

Metric Filtering Multiple Comparison Correction (MCC) Update

When filtering your organizational metrics on the Metrics impact page, MCC will only be applied once for all organizational metrics. This will prevent users from seeing different p-values when metric results are filtered.