Design guidelines

Users typically have multiple devices that store and display information for specific uses. Don't try to replace smartphone, tablet, or laptop features with Glass. It works best with information that's simple, relevant, and current. Instead, use Glass to supplement these devices with hands-free technology that can increase the availability of their information.

To deliver the best experience possible for your users, we suggest that your applications prioritize the following qualities:


Glass is designed to be there when you need it and out of the way when you don't. Your applications must function in the same way.

Keep interfaces uncluttered and easy to read. Organize information within a clear hierarchy. Ensure users can immediately identify each screen’s message and the app’s intended action.

Avoid the need for users to squint to read the UI. Also, keep your content short, so users don’t need to scan through extensive loads of information to get what they need.


Deliver information at the right place and time for your users. The most relevant experiences are also the most impactful and lead to increased engagement and satisfaction.

Design around the core actions of suggest and demand:

  • Suggest involves making useful, contextual suggestions to users.
  • Demand involves taking commands from users.

Easy to tap

Historically, Glass's best interactions allow for only one tap action per card with additional advanced actions available with a double tap. If you leave this model, and allow for more than one tap target per screen, you might need to keep the pointer visible.

Make sure your tap targets are well-spaced and easy to tap. Your app shouldn't rely on heavy user input. If that can't be avoided, explore creative techniques to facilitate that input.

Time saving

Create efficient flows that help users complete tasks quickly. Don't create complex experiences that rely on multiple steps to get a task done. Whenever possible, reduce the number of steps needed to complete a task. Anticipate which actions a user would take at each step.