We’re bringing integrations with feature management services LaunchDarkly and Rollout into Jira Software. Before releasing this integration, we custom-built it into Jira Software and it’s been instrumental to the way we build products. Read on to learn more, sign up for the webinar, and get the integrations.


Join our webinar with Taylor Pechacek, Sr. Product Manager for Jira Software Cloud, to learn how feature flagging was an essential part of our agile process as we built and rolled out the new Jira experience.

Sign up for the webinar

Want to see how to plug in the integrations? Watch the video below!


At Atlassian our product teams are launching new features every day, but that probably doesn’t mean exactly what you’re imagining. We’re not doing big-bang launches with lots of fanfare on a daily basis. Rather, we’re releasing new features incrementally, little by little every day, with a careful eye towards delivering the right customer experience.

Most product teams have at some point in their journey delivered an experience that hasn’t been the right one, whether that’s shipping software with a bug in it or something new that doesn’t work quite right for customers. At Atlassian, we use feature flags as one of our key strategies to ensure we deliver the best possible experience as we roll out new features, helping us move fast with less risk, and giving us greater control over the value we deliver to customers.

Feature flags enable us to roll out product updates to a highly segmented group of customers, for example to customers in a specific user tier, region, or with a specific product configuration. Every time we release, we look for validation that the feature is having a positive outcome (for example, it’s moving a metric we’ve set or if people give us positive feedback). If we’re seeing the right signals, we continue to roll out the feature to more and more customers. If something breaks or if we get negative signals after a release, we can easily roll back in a matter of seconds without updating any code. Each time we release is an opportunity to learn something new. This relatively simple ‘build-measure-learn’ loop is enhanced by feature flags and has become a cornerstone of the way Atlassian build products.

While this approach to product development isn’t new, the emergence of feature management services like LaunchDarkly and Rollout have made it easier for all software development teams to take a test-and-learn approach to development. Before the dawn of feature management services, product teams who wanted to do iterative rollouts would have to employ a manual process to implement feature flags. This still affords the short term benefits for which feature flags are known (enabling developers to move faster with less risk, as well as the ability to A/B test product rollouts), but it also comes with the additional overhead of manually managing, tracking, and eventually removing flags. If not managed well, this can result in technical debt that accrues over time. The consequences can be products that are harder to support and debug, or even accidental releases to customers.

Our new integrations with LaunchDarkly and Rollout

That’s why we’re so excited about our new integrations with LaunchDarkly and Rollout. We believe bringing additional context about flags into Jira will help improve team coordination and collaboration around product releases and generally make the practice of feature flagging more effective for both developers and the rest of the team.

Feature flags integrated into Jira Software, showing the percentage of customers a feature is reaching

For example, if you’re rolling out a new product release in phases, it can be tough to align all relevant stakeholders around the most up-to-date status of a product, which can result in problematic scenarios. Like if customers reach out to Support and Support doesn’t have visibility into what version of a product customers are seeing. Or if Marketing is planning a launch and it’s not clear that a feature has been rolled out to 100% of customers. By bringing this information into Jira, you can easily keep all team members in the loop about the latest status of the product.

This extra visibility and context helps bring peace of mind to product managers and developers. If incomplete code has been pushed to production behind a flag, stakeholders may want quick assurance that it hasn’t reached customers, without jumping between multiple applications. With the LaunchDarkly and Rollout integrations it’s easy to confirm, within Jira Software, which features have and haven’t been released to customers.

Finally, there’s nothing to slow down the pace of development like context switching across multiple applications. By bringing more relevant context into your Jira issues – especially live data from other applications, like LaunchDarkly and Rollout – it helps to ensure greater visibility and less friction across the team, keeping everyone in sync at critical times.

How-to video

Getting started with our feature flagging integrations is easy. Check out our how-to video below on setting up the integration.

Work smarter, better, and faster with weekly tips and how-tos.

Subscribe now