LaunchDarkly (Actions) Destination
LaunchDarkly is a feature management platform that empowers development teams to safely deliver, control, and measure their software through feature flags.
With LaunchDarkly, you can run experiments on any feature flag. This destination allows you to connect existing Segment events to LaunchDarkly custom metrics for use in LaunchDarkly experiments.
Success!
Good to know: This page is about the Actions-framework LaunchDarkly Segment destination. There's also a page about the non-Actions LaunchDarkly destination. Both of these destinations receives data from Segment.
LaunchDarkly (Actions) provides the following benefits over the classic LaunchDarkly destination:
- Improved customization. You determine the mapping between the data Segment receives from your source and the data Segment sends to LaunchDarkly. For example, you can map an arbitrary event property to the LaunchDarkly metric key.
- Increased transparency. You can see the data that is sent to LaunchDarkly and when Segment sends it. Additionally, you can subscribe to alerts when the delivery rate to LaunchDarkly dips below a configurable threshold.
To get started with LaunchDarkly (Actions):
- In LaunchDarkly, navigate to Account settings and copy the client-side ID for the project and environment that you would like to connect to Segment.
- From the Segment web app, click Catalog, then click Destinations.
- Search for LaunchDarkly (Actions) and select it.
- Click Configure LaunchDarkly.
- Select the Source you want to connect to LaunchDarkly (Actions).
- Paste the LaunchDarkly client-side ID you copied in step 1 into the LaunchDarkly client-side ID field on the destination settings page.
Find and copy the client-side ID in the LaunchDarkly account settings page.
Optional
Your LaunchDarkly events host name. If not specified, the default value of events.launchdarkly.com will be used. Most customers will not need to change this setting.
LaunchDarkly (Actions) has the following presets
| Preset Name | Trigger | Default Action |
|---|---|---|
| Alias User | Event type = "identify", Event type = "alias" | Alias User |
| Track Event | Event type = "track" | Track Event |
Build your own Mappings. Combine supported triggers with the following LaunchDarkly-supported actions:
Mapping limits per destination
Alias an anonymous user with an identified user key.
Alias User is a Cloud action. The default Trigger is type = "identify" or type = "alias"
The user's unique key.
The user's unauthenticated identifier.
Optional
The time when the event happened. Defaults to the current time.
Track custom events for use in A/B tests and experimentation.
Track Event is a Cloud action. The default Trigger is type = "track"
Optional
The event's context kind. If not specified, the context kind will default to user. To learn more about context kinds and where you can find a list of context kinds LaunchDarkly has observed, read Context kinds.
The unique LaunchDarkly context key. In most cases the Segment userId should be used.
Optional
A mapping of additional context kinds to context keys. To learn more, read Contexts and segments.
The name of the event to track. This name typically corresponds to a LaunchDarkly metric with the same key.
Optional
The numeric value associated with the event. This value is used by the LaunchDarkly experimentation feature in numeric custom metrics, and will also be returned as part of the custom event for Data Export.
Optional
Optional object containing the properties for the event being tracked. These properties assist with observational analytics for LaunchDarkly Data Export destinations. These properties are not saved to the LaunchDarkly user.
Optional
The time when the event happened. Defaults to the current time.
A note about Identify calls
LaunchDarkly doesn't store user attributes for use with flag evaluations. As a result, Segment doesn't support mapping Identify calls to LaunchDarkly actions.
In order to take full-advantage of the LaunchDarkly (Actions) Destination, you need to create metrics in LaunchDarkly that correspond to Segment track events. Read Creating metrics to learn how to create metrics in LaunchDarkly.
Be sure to disconnect the classic LaunchDarkly destination before enabling the LaunchDarkly (Actions) destination to avoid duplicate experimentation events in LaunchDarkly.