4 minute read
Most developers today are familiar with feature flags and the benefits they bring to continuous delivery such as increased safety and agility. However, few have realized the value of feature flag data. On a feature delivery platform, such as Split, feature flags produce changelog data detailing flag updates and impression data detailing who saw which flag treatment and when.
Both data types are incredibly valuable when it comes to understanding and measuring the impact of individual features. They not only power Split’s monitoring and experimentation capabilities, but can also add value to other tools, such as application performance monitoring (APM) tools, customer data platforms, and product analytics tools. In fact, Split has several integrations with such tools enabling our customers to leverage these data in various ways. Today, we are excited to announce our two latest integrations with Dynatrace and Mixpanel.
Release Safer and Recover Faster with Split and Dynatrace
With our Dynatrace integration, Split changelog data is sent to Dynatrace and appears as custom annotation events for tagged entities such as hosts, services, and applications. As flags are configured and updated during the release process in Split, each flag change can be tracked alongside unwanted spikes and degradations in Dynatrace, making it faster and easier to pin down the root cause. Through Split feature flagging and Dynatrace performance monitoring, engineering teams can reduce release risk and time to recovery.
- Save time investigating root cause. Correlate a code change or feature release with a shift in performance or other key metrics.
- Instantly stop bad features. Reduce time to recovery using Split’s killswitch to instantly disable the feature at fault without having to rollback the entire release or implement a hotfix.
- Speed up time to recovery. Minimize downtime and outages caused by buggy new features. Detect and resolve issues quickly to protect revenue and the customer experience.
Experiment, Analyze, and Build Better Products with Split and Mixpanel
With our Mixpanel integration, Split impressions are sent to Mixpanel and appear as
split_impression events each with a
split property indicating the name of the feature and a
treatment property indicating the flag treatment a user was exposed to. Bringing this feature-level context into Mixpanel allows product teams to better understand the customer journey and conversion funnel by flag treatment. For example, product managers can experiment on a new checkout feature and gather key metrics in Split, and then analyze how the feature affected user behavior in each stage of the conversion funnel within Mixpanel. Using Split and Mixpanel, teams can make data-driven decisions on how to improve their features and deliver better digital experiences.
- Safely release and experiment with new features. Use Split targeting and experimentation to test out new features on specific audiences. Measure the statistically significant impact on key performance and business metrics.
- Analyze feature impact on user behavior. Break down Mixpanel reports by flag treatment and other attributes to compare impact on user behavior throughout the customer journey and conversion funnel.
- Make data-driven product decisions. Gather actionable insights on how to iterate on new features and improve the user experience.
Learn More About How You Can Pair Feature Flags with Data to Drive Real Business Impact
If you’re looking to get started on Dynatrace and/or Mixpanel today, we’ve got you covered. Sign up for a free trial of Split and check out our documentation on Dynatrace and Mixpanel. See what other tools Split connects with today, check out our integrations page and additional resources below.
- Out-of-the-Box Experimentation on High-Volume Data with Amazon S3 and Split
- Manage Feature Delivery with Jira and Split in Your Open DevOps Toolchain
- Deliver World-Class Experiences with FullStory and Split