Actions on Google is a developer platform that lets you create software to extend the functionality of the Google Assistant, Google’s virtual personal assistant, across more than 500 million devices, including smart speakers, phones, cars, TVs, headphones, watches, and more.

Users engage Google Assistant in conversation to get things done, like buying groceries or booking a ride (for a complete list of what’s possible now, see the Actions directory). As a developer, you can use Actions on Google to easily create and manage delightful and effective conversational experiences between users and your own 3rd-party fulfillment service.

Actions for the Google Assistant

Unlike with traditional mobile and desktop apps, which use computer-centric paradigms, users interact with Actions for the Assistant through natural-sounding, back and forth conversation.

Smart home Actions

You can build smart home Actions that let users control Internet of Things (IoT) devices through the Google Assistant. Building smart home Actions lets you connect, query, and control devices through your existing cloud infrastructure.

To learn how to build smart home Actions, please see our documentation.

Conversational Actions

To start a conversational Action, the user needs to invoke your Action through the Assistant. Users say or type a phrase like “Hey Google, talk to Google IO 18”. This tells the Assistant the name of the Action to talk to.

From this point onwards, the user is talking to your Action and giving it input. This conversation continues as a two-way dialog until the user’s request is fulfilled or the conversation is finished.

How conversational Actions work

Android example
Figure 1: Example of using an Action on a mobile device
  • When you build an Action for the Assistant, you design your conversations for a variety of surfaces, such as a voice-centric conversation for voice-activated speakers or a visual conversation on a surface that the Assistant supports. This lets users get things done quickly through either voice or visual affordances.
  • Users can invoke your Actions from any of these surfaces:
    • Wear OS devices
    • Assistant-enabled headphones
    • Chromebooks
    • Android TV
    • Android phones and tablets
    • Smart displays and speakers
    • iPhones
  • Your Actions run entirely in the cloud, even when users talk to them on their phone, smart home device, or watch.
  • The user’s device sends the user’s utterance to the Assistant, which routes it to your fulfillment service via HTTPS POST requests.
  • Your fulfillment figures out a relevant response and sends that back to the Assistant, which ultimately returns it to the user.

Example: Learn about numbers with Geeknum

Wear OS example
Figure 2: Example of using an Action on a Wear OS device

To understand how users interact with your Action, we'll use a hypothetical Action called 'Geeknum'. Users can invoke Geeknum through the Assistant on their phone or watch to learn fun facts about any number.

  1. A user talks to the Assistant and requests an Action: "I want to learn about numbers". The Assistant looks for the best Action to handle the user's intent and returns an Action named 'Geeknum'.
  2. The Assistant asks the user if she wants to invoke Geeknum and she says "Yes". Behind the scene, the Assistant sends a request to the Action's fulfillment by HTTPS POST and receives a response that the Assistant renders to users.
  3. The Assistant renders the response in its user interface and displays a welcome message to the user. The Assistant introduces Geeknum, then hands off the user to Geeknum. At this point, the conversation between the user and Geeknum begins.
  4. During the conversation, the Assistant sends subsequent user input directly to the fulfillment and the fulfillment responds directly to the Assistant. This conversation continues until the fulfillment gathers the user input it needs to fulfill the user's intent.
  5. The conversation ends when the user is done learning about numbers.