June 2024 Summaries

With feature flags, developers can implement staged rollouts, focused releases, remote configuration, A/B testing, and quickly turn off features as needed, regardless of where they are running the feature flags. However, when using feature flags on the client side of a website or application, there are technical aspects to consider, particularly with Javascript SDKs. Local evaluation involves sending the feature flagging payload directly to the browser, while remote evaluation requires a network call to a service to determine which features should be enabled for a user. Both methods have pros and cons, including speed and security considerations. To address these concerns, GrowthBook offers features such as payload encryption, secure attribute hashing, and hiding experiment names, which can help protect sensitive data and ensure the integrity of feature flagging. Remote evaluation is slower than local evaluation but provides complete control over feature flag states and enables more dynamic lookups based on user identifiers. Ultimately, using feature flags on the client side adds valuable functionality to a site or application.

Client-side experimentation with visual editors has its advantages, such as making optimizations easy to launch and requiring no code changes. However, it also has drawbacks like page flickering or flashing, slowing down page load speeds, and ad-blocked experiments. To address these issues, Edge worker SDKs have been developed by enabling CDN edge workers, which render experiment variants directly to the HTML served from the edge to the client, eliminating flickering, delays, and ad-blocking. This solution also enables URL redirect experiments and feature flags on the edge, making it a powerful tool for experimentation and optimization. By leveraging Edge worker SDKs, developers can unlock visual experiments without compromise and gain more control over their experimentation workflow.