Set up Conversational analytics in Looker

This page guides you through setting up conversational analytics in Looker, including setup requirements, required permissions to use conversational analytics, and supported data sources.

Learn how and when Gemini for Google Cloud uses your data.

Before you begin

To use conversational analytics within a Looker instance, you and your Looker instance must meet the following requirements:

  1. Gemini in Looker must be enabled for the Looker instance.
    • To access these features in a Looker (original) instance, a Looker admin must enable the Gemini in Looker setting and the Conversational Analytics setting in the Looker (original) instance settings. To use conversational analytics dashboard agents or agentic workflows, update your instance to Looker 26.8 or later.
    • To access these features in a Looker (Google Cloud core) instance, a user with the Looker Admin (roles/looker.admin) IAM role must enable Gemini in Looker in the Looker (Google Cloud core) instance settings in the Google Cloud console. A Looker admin must enable the Conversational Analytics setting on the Gemini in Looker page in the Admin panel of the Looker (Google Cloud core) instance.
  2. If you have configured an IP allowlist for your instance, you must configure it to allow connections from Google Cloud services:
  3. Optionally, an admin can enable Advanced Analytics to use Python to answer advanced questions. Under Conversational Analytics, turn on the Advanced Analytics setting.
  4. Optionally, an admin can enable dashboard data agents (Preview) to query Looker dashboards. On the Gemini in Looker admin page, under the Enable Trusted Tester Features preview settings, turn on the Enable Chat with Dashboard setting. The Enable Trusted Tester Features setting must also be enabled.
  5. Optionally, an admin can enable agentic workflow (Preview) to create dynamic workflows that monitor your data and alert you of changes. On the Gemini in Looker admin page, under the Enable Trusted Tester Features preview settings, turn on the Agentic Workflows setting. The Enable Trusted Tester Features setting must also be enabled.

Required Looker permissions

The use of conversational analytics is managed through a combination of content access, data access, and feature access. To perform the tasks that are described in the following table, you must be assigned a Looker role that has the required permissions for the models that underlie the individual Explore or that the data agent will query and, in some cases, access to the agent itself.

A Looker admin can grant Looker roles and permissions on the Roles page in the Admin section of the Looker instance. For more information about Looker roles, see the Admin settings – Roles documentation page.

The creator of the data agent can manage individual users' access to the agent by sharing the agent.

Task Required Looker permissions on the model that underlies the data Required level of content access
Chat with a data agent from the Agents tab in the conversational analytics interface access_data

gemini_in_looker

chat_with_agent		 View access on the data agent
Chat with a Looker Explore from the Explores tab in the conversational analytics interface access_data

gemini_in_looker

chat_with_agent

chat_with_explore
Create, edit, share, and delete conversational analytics data agents access_data

gemini_in_looker

chat_with_agent

chat_with_explore

save_agents

Users can create agents that use only the Explores for which they have been granted the save_agents permission on the underlying model. To edit, delete, or share a data agent that was created by another user, users must be granted a role that contains the save_agents permission on every model that is used by the agent.		 Manage access; Edit on the data agent (this access is granted automatically if the user creates the agent; otherwise, Manage access; Edit access must be granted by the agent's creator by sharing the agent)

Manage access; Edit on the dashboard
Chat with a dashboard agent from a Looker dashboard (Preview) access_data (on at least one model that underlies the dashboard)

see_user_dashboards (to interact with user-defined dashboards) or see_lookml_dashboards (to interact with LookML dashboards)

gemini_in_looker

chat_with_agent (on at least one model that underlies the dashboard)		 Manage access; Edit on the data agent (this access is granted automatically if the user creates the agent; otherwise, Manage access; Edit access must be granted by the agent's creator by sharing the agent)

Manage access; Edit on the dashboard
Publish agents to other Google Cloud applications (Preview) access_data

gemini_in_looker

chat_with_agent

chat_with_explore

save_agents

Added 26.6 publish_agent_externally

Users publish existing agents to other Google Cloud applications, such as Gemini Enterprise.		 Manage access; Edit (this access is granted automatically if the user creates the agent; otherwise, Manage access; Edit access must be granted by the agent's creator by sharing the agent)
Create an agentic workflow with a dashboard agent (Preview) access_data (on at least one model that underlies the dashboard)

see_user_dashboards (to interact with user-defined dashboards) or see_lookml_dashboards (to interact with LookML dashboards)

gemini_in_looker

chat_with_agent (on at least one model that underlies the dashboard)

see_looks

create_alerts

see_alerts (to view the workflows that you have created on the Manage Workflows user page)		 Manage access; Edit on the data agent (this access is granted automatically if the user creates the agent; otherwise, Manage access; Edit access must be granted by the agent's creator by sharing the agent)

Manage access; Edit on the dashboard
Create an agentic workflow with an Explore data agent (Preview) access_data

gemini_in_looker

chat_with_agent

see_looks

create_alerts

see_alerts (to view the workflows that you have created on the Manage Workflows user page)		 Manage access; Edit on the data agent (this access is granted automatically if the user creates the agent; otherwise, Manage access; Edit access must be granted by the agent's creator by sharing the agent)

Manage access; Edit on the dashboard
Create an agentic workflow within an Explore conversation (Preview) access_data

gemini_in_looker

chat_with_agent

chat_with_explore

see_looks

create_alerts

see_alerts (to view the workflows that you have created on the Manage Workflows user page)		 Manage access; Edit on the data agent (this access is granted automatically if the user creates the agent; otherwise, Manage access; Edit access must be granted by the agent's creator by sharing the agent)

Manage access; Edit on the dashboard
Create, edit, share, and delete conversational analytics data agents access_data

gemini_in_looker

admin_agents		 Content access doesn't need to be granted

Looker also has the following default roles that contain subsets of these permissions for all models on the instance:

  • Conversational Analytics Agent Manager: With this role, a user can create, edit, share, delete, and chat with agents that they have Manage access; Edit access for and chat with Explores.
  • Conversational Analytics User: With this role, a user can chat with an agent that they have View access for.
  • Admin: By default, this role (Looker Admin) has all permissions and content access across the instance.

Supported data sources

You can converse with the following data sources through the conversational analytics interface:

When you "chat" with an Explore or a data agent, you're essentially having a conversation with a specific, predefined dataset. This is a direct way to ask questions about the data within that Explore. You can have a conversation with up to five Explores.

Explore data agents are AI-powered assistants that you can create and customize. Dashboard agents are created automatically when you "chat" with a dashboard. You can give data agents instructions and context that are specific to your data.

For strategies and best practices to help Looker admins and LookML developers successfully configure, deploy, and optimize conversational analytics, including LookML best practices for conversational analytics, tips on when to add context to LookML versus conversational analytics, and best practices for setting up an Explore for use with conversational analytics, see the Best practices for configuring conversational analytics in Looker documentation page.

What's next