How Feature Flagging Platforms are Progressing DevOps
DevOps fever is spreading fast as more companies begin to adopt the set of practices in order to boost efficiency while continuing to move at high velocity. From startups to enterprise organizations, companies around the world have become focused on enabling technical/non-technical collaboration and implementing tools that simplify the software delivery process. While this has proven to be highly effective for boosting productivity, much of this progress would not be possible without reliable feature flagging systems. In fact, feature flags have been enabling advancement in DevOps collaboration practices, as well as continuous delivery methods for years now. This has manifested as a change in company culture, but is ultimately derived from the implementation of feature flag driven development into the software delivery process.
“If you want to do DevOps, you have to constantly deliver value to your customers and gather evidence of the success in production. This means constant, zero downtime deployments and testing new features with smaller sets of customers (A/B testing, canary releasing).”
– Mike Kaufmann, There is No DevOps Without Feature Flags
Feature Flags as a Catalyst
From their onset, feature flagging platforms have accelerated the progression of DevOps practices. Feature flags have always fit into DevOps practices due to the increased control over the delivery cycle that they offer, but the mitigation of risk and prevention of associated technical debt has brought flagging platforms to the spotlight as quintessentially “DevOps”. These changes alongside the newly offered ability to monitor all changes to a flag via an audit log have widened the reach of the methodology. Simultaneously, these systems deeply supplement the methodology as it currently exists and provide additional use cases and benefits that push the limits of DevOps beyond what was possible before.
Benefits of Feature Flagging as a Service
Prior to a reliable feature flagging platform, the only solution was to develop your own home-grown system. These systems required many development hours, lacked risk mitigation features (audit logs, percentage rollouts, or custom roles to name a few), and often were not built to scale properly. This is in addition to these flaws, many of the home-grown systems would end up without a user interface of any kind, leaving it completely inaccessible to non-technical staff.
Thankfully, there are companies like LaunchDarkly that offer SaaS feature flagging platforms to businesses in need. With this solution, the risk of releasing unfinished code, or toggling the incorrect flag is largely circumvented. These systems are simple for an engineering team to set up, and are usable by non-technical staff. Continuous delivery is now more accessible than ever, and this has enabled companies of all sizes and experience levels to embrace modern DevOps practices such as trunk-based development and other current continuous delivery techniques.
Check out the following case studies for further insights into how “feature flags as a service” can be utilized in real-world DevOps systems.
As SME and large enterprise companies begin to embrace DevOps further, feature flagging platforms will be riding shotgun as one of the premier ways to transition a company from old-school waterfall releases into a culture of modern continuous delivery practices.
In summary, the benefits of a feature flag management platform are:
- Reducing the engineering bottleneck by empowering non-technical users
- Allows companies to provide better software, faster, and with less risk
- Enables DevOps practices to be scaled to the Enterprise level
- Simplified A/B testing + Feature rollout control