Toggle is a simple but powerful user interface component. It allows you to change a program’s behavior in the moment based on a simple decision. But if used poorly, it can be confusing for your users and a nightmare to debug weeks or months down the road when changes start to impact other teams’ work.

Toggles can be implemented in many different ways, but they should always provide clear and direct labels, standard visual design, and deliver immediate results. It is tempting to create toggle switches that can change more than one feature at a time, but doing so will confuse your users and make it harder for other teams to understand what you’re trying to do.

When you’re ready to take a toggle into production it’s important to test it against the real-world use cases that will be affected. Many teams do this by adding a test to their backlog for each toggle they add to the system with the expectation that it will be flipped off once released. Some teams even put an expiration date on their toggles so that they’ll automatically fail a test at the end of the set time period.

It’s also best practice to name your toggles so that you can be specific about what the toggle does in case someone on another team comes across it in an error message. This not only makes it easier to identify where the error is coming from, but also provides some useful context when that other team is investigating why a particular toggle has been triggered.