Optimizely will be sunsetting Full Stack Experimentation on July 29, 2024. See the recommended Feature Experimentation migration timeline and documentation.

Dev GuideAPI Reference
Dev GuideAPI ReferenceUser GuideGitHubNuGetDev CommunitySumbit a ticketLog In
GitHubNuGetDev CommunitySumbit a ticket

Get started experimenting at the edge

The Optimizely Edge SDK for Full Stack Experimentation is a set of templates and starter kits to implement feature flagging and experimentation across major serverless edge compute platforms.

Modern content delivery networks (CDNs) and cloud platforms offer flexible and robust serverless edge computing capabilities that provide a natural environment for executing performant and influential experiments.

To optimize decision performance and bundle size with these CDNs, Optimizely has extended and embedded our Javascript SDK into several edge computing providers.

Use cases

The Optimizely Edge SDKs let you implement feature flagging and experimentation within the most popular serverless edge services, creating new use cases and improved performance. Edge functions are best suited for small but impactful user-specific calculations where speed is essential. For example,

  • Localization
  • Dynamic advertisement targeting
  • Content stitching
  • Authentication

📘

Note

The Optimizely Edge SDK for Full Stack Experimentation is a set of SDKs for running experiments on popular serverless edge computing platforms. This documentation provides instructions and examples of these SDKs.

This differs from Optimizely Performance Edge. Performance Edge is a faster way to deliver client-side Optimizely Web experiments that utilizes the structure of CDNs to deliver experiments to visitors faster by downloading a microsnippet. For more information on Performance Edge, select it from the Products dropdowns above, or click on the link.

CDN providers

Select a CDN provider to get started using templates and starter kits to build feature flags and experiments across the major serverless edge compute platforms.