A toggle is a hardware or software switch that turns something on and off. For example, the keyboard’s Caps Lock and Num Lock toggle on and off specific functions. This is the same principle that applies to features that can be enabled and disabled in options menus across most applications.

There are many ways to manage feature toggling ranging from simple commenting through to using a preprocessor’s #ifdef to hardcode the toggle configuration. However, these static methods do not allow for a toggle to be re-configured dynamically at runtime. To address this savvy teams will move to some form of centralized toggle storage which is usually a form of application DB and often accompanied by a build-out of a management interface that allows system operators, testers and product managers to view and modify the toggle configuration.

In addition to their flexibility, toggles are ideal for use with experimentation and data-driven decision making. For example, an ecommerce company can create a toggle that enables a new algorithm B and then deploy it to specific cohorts of users. This will allow the ecommerce team to see whether or not the new algorithm has an impact on their conversion rate.

While toggles can be a useful tool, it’s important to remember that they can add cognitive complexity to the user experience. It’s also important to keep the inventory of toggles low. Leaving old toggles lying around in code comes with a significant maintenance cost for the teams that must regularly weed them out. This can be mitigated by adding a toggle removal task to the team’s backlog or incorporating the process into your management platform.