You’ve just shipped your newest feature and you are feeling great except for the looming question–what’s happening in production? You likely have more than a few systems in place to help you monitor and observe your shiny new ship, but have you actually built your new software to be risk tolerant? What happens when that dreadful error occurs? Are you actually ready to rollback?

We’ve been working with our friends at Rollbar to help our customers (we’re also customers!) understand how their features are working, or not working, in production by connecting timely error data with their features and releases. We’re thrilled to announce that Rollbar has built the capability for feature flags to be turned off automatically based on new and reactivated errors. This allows developers to ship the feature faster, with more confidence, and reduced risk. Less time panicking, more time shipping – even on Fridays.  

Our customers are using Rollbar’s Automated Feature Flag Triggers, part of their new AI-Assisted Workflows, to tackle real problems in software delivery, including:

  • Progressive delivery. Roll your feature out to a small percentage of users over time and confidence. If your feature starts receiving the dreadful error, Rollbar’s Automated Feature Flag Triggers will automatically roll the feature back. 
  • Test-in-production. Test your features with real users in production environments. You can use LaunchDarkly toggles and Rollbar Automated Flag Triggers conjunctively to mitigate risk.
  • A/B Testing. Developers and Product Managers can now use errors to decide which versions of a feature should be fully deployed or turned off.

We’re inviting developers to try the new integration today by joining our Automated Feature Flag Triggers beta program. If you’re interested in building your own LaunchDarkly integration or flag trigger then you can register for our partner program, or reach out to us directly at