Alert policies With alert policies, you can respond quickly to performance degradations for each environment you operate in. You can configure a specific degradation threshold and customize who should be notified if your metric degrades. Split will detect a degradation in metrics in the first 30 minutes of a new version of a split.
Sentry integration When configured, Split will ingest error and exception metrics from Sentry and ties these to the feature a user is exposed to. Split will display Sentry exception data in the Split platform for analysis. You can control what environments and traffic types you capture exceptions for in Split’s dashboard without code.
Real User Monitoring (RUM) Agent Track data such as Time to First Byte, Time to DOM Interactive, Page Load Time, Time to Interactive, Errors – key indicators of user experience. In addition, the agent collects key properties like connection type, url and userAgent to give you a full picture of the user using your feature.
Feature Experimentation Multiple Treatments and Metric Values in Metric Details You can now visualize metric values, in addition to metric impact, over time in the metric details and trends view. You can now also compare multiple treatments against your baseline treatment within this view.
Feature Monitoring Configurable monitor windows Configurable monitor windows allow you to select the length of time that Split will check your alert policies. You can select a timeframe between 30 minutes to 90 days after a split version change. The default monitor window will be set to 24 hours. SDK…
Management Console New Usage Tab Within the admin panel you can view under the Usage tab your Split usage in terms of impressions, events and monthly tracked keys. Monthly tracked keys are defined as the monthly unique IDs across all environments and across all workspaces for impression data.