<br />

> [!WARNING]
>
> **Preview**
>
>
> This product or feature is
>
> subject to the "Pre-GA Offerings Terms" in the General Service Terms section
> of the [Service Specific
> Terms](https://docs.cloud.google.com/terms/service-terms#1).
>
> Pre-GA products and features are available "as is" and might have limited support.
>
> For more information, see the
> [launch stage descriptions](https://cloud.google.com/products/#product-launch-stages).

Conversational Analytics, powered by [Gemini for Google Cloud](https://docs.cloud.google.com/gemini/docs/overview.md.txt), lets you investigate your data by asking questions about it in conversational language through an intuitive chat interface. When you use Conversational Analytics with a [user-defined dashboard](https://docs.cloud.google.com/looker/docs/2612/editing-user-defined-dashboards.md.txt) or a [LookML dashboard](https://docs.cloud.google.com/looker/docs/2612/reference/lookml-dashboard-overview), Conversational Analytics creates a data agent for you that is connected to the dashboard. As you converse with the dashboard agent, the agent will query your dashboard and any of its [query-linked tiles](https://docs.cloud.google.com/looker/docs/2612/viewing-dashboards.md.txt#dashboard_tiles). You can customize user-defined dashboard agents with context and instructions that are specific to your dashboard data.

> [!NOTE]
> **Note:** The page refers to dashboard agents --- data agents that query dashboards. For information about Explore data agents --- data agents that query Explores, see the [Query an Explore in natural language](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data.md.txt) documentation page. For information about creating Explore data agents, see the [Create and manage Explore data agents](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents.md.txt) documentation page.

This page discusses how to use Conversational Analytics to engage with dashboard data. It covers the following topics:

- [Starting a conversation from within a dashboard](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#start-conversation-dashboard)
- [Adding agent instructions and context](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#modify-dashboard-data-agent)
- [Asking questions about your dashboard data](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#ask-questions)
- [Managing conversations](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#manage-conversations)
- [Known limitations with conversations](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#known-limitations)

## Before you begin

Before you can use Conversational Analytics to engage with your dashboard data, make sure that the [setup and requirements for your Looker instance](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-setup.md.txt#setup-and-requirements) have been satisfied and that you have been granted the [appropriate permissions](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-setup.md.txt#required-permissions) to perform the tasks that are described on this page.

## Start a conversation from within a Looker dashboard

To start a conversation with a dashboard agent from a [user-defined dashboard](https://docs.cloud.google.com/looker/docs/2612/editing-user-defined-dashboards.md.txt) or a [LookML dashboard](https://docs.cloud.google.com/looker/docs/2612/reference/lookml-dashboard-overview), select spark **Chat with this dashboard**.

If you don't have sufficient permissions to chat with the dashboard, Looker displays an error.

Once you have created a conversation, you can [ask questions](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#ask-questions) about the data in the **Ask a question** field within the conversation. To access your recent conversations with the dashboard agent, select more_vert **Menu** \> **Recent conversations**.

In addition to [asking your dashboard agent questions](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#ask-questions) about the data on the dashboard or its query-linked tiles, for user-defined dashboard agents, you can [customize the agent's configuration](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#modify-dashboard-data-agent) with additional context and instructions.

### Enable or disable dashboard agents for a dashboard

If dashboard agents are enabled for your instance, they're enabled for all
user-defined dashboards by default. However, dashboard creators can control
whether users can chat with a specific user-defined dashboard by enabling or
disabling the dashboard agent setting at the individual dashboard level.

To turn a dashboard agent on or off for a user-defined dashboard, follow
these steps:

1. Select the more_vert **Dashboard actions** menu icon and then select edit **Edit dashboard**.
2. Click settings **Settings** in the dashboard toolbar.
3. In the **Settings** dialog, select the **General** tab.
4. Turn on or turn off the **Enable chat with dashboard** control.
5. Click **Save** to save your changes to the dashboard.


### Conversation metadata

The **Chat with this dashboard** pane supports the following tasks for each dashboard type:

### User-defined dashboard

- To [ask a question](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#ask-questions) about the dashboard data, enter a question in the **Ask a question** field.
- To edit information about the dashboard agent, select tune **Manage agent** . In the **Editor** tab, you can enter [agent instructions](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents.md.txt#write-agent-instructions) and expand **Advanced options** to configure settings like **Advanced Analytics** . Select **Update** to save your changes to the dashboard agent. Use the **Preview** tab to preview your changes to the dashboard agent.
- To expand the dashboard agent conversation pane, select open_in_full **View full screen**.
- To access your recent conversations with the dashboard agent, select more_vert **Menu** \> **Recent conversations**.
- By default, conversations will be named based on your initial question. To rename a conversation, select a conversation from the **Recent conversations** menu option, select more_vert, and then select edit **Rename**.
- To delete a conversation with a data agent, select more_vert **Menu** \> delete **Trash**.

### LookML dashboard

- To [ask a question](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#ask-questions) about the dashboard data, enter a question in the **Ask a question** field.
- To expand the dashboard agent conversation pane, select open_in_full **View full screen**.
- To access your recent conversations with the dashboard agent, select more_vert **Menu** \> **Recent conversations**.
- By default, a conversation is named based on your initial question. To rename a conversation, select a conversation from the **Recent conversations** menu option, select more_vert, and then select edit **Rename**.
- To delete a conversation with a data agent, select more_vert **Menu** \> delete **Trash**.

## Modify your dashboard agent

> [!NOTE]
> **Note:** You cannot edit the agent instructions for a LookML dashboard agent.

To edit information about the dashboard agent, select
tune
**Manage agent**.

Optionally, you can enable [**Advanced Analytics**](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-advanced-analytics.md.txt) for your agent by turning on the **Enable Advanced Analytics** setting under**Advanced Analytics** . Turning on this setting allows your agent to use **Advanced Analytics** , which translates your natural language questions into Python code and executes that code to provide advanced analysis and visualizations. This setting is visible only when the **Advanced Analytics** admin setting is enabled on the **Gemini in Looker** admin page for your instance.

Select **Update** to save your changes to the dashboard agent. Use the
**Preview** tab to preview your changes to the dashboard agent.

### Write agent instructions

You can add free-form instructions that define your data agent's core behavior and provide it with foundational context to consider before processing a user's prompt.

> [!IMPORTANT]
> **Important:** Although the **Instructions** field is optional and the structure is up to your discretion, we recommend that you provide well-structured instructions to enable the agent to give more accurate and relevant responses.

Here are some examples of the types of context that you can provide in the **Instructions** field:

- Key fields: The most important fields for analysis
- Excluded fields: Fields that the data agent should avoid
- Filtering and grouping: Fields that the agent should use to filter and group data
- Persona: A role or expertise, character, or tone that you assign to the agent

> [!NOTE]
> **Note:** The types of authored context that you can provide a data agent in Looker differ slightly from those that you can provide an agent that you're creating by using the Conversational Analytics API. Some of what you provide in the **Instructions** field in Looker can be given to an agent as [structured context](https://docs.cloud.google.com/gemini/docs/conversational-analytics-api/data-agent-authored-context-looker#structured-context-looker) by using the Conversational Analytics API.

For tips and best practices for writing agent instructions, see the [Best practices for configuring Conversational Analytics in Looker](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-best-practices.md.txt) documentation page.

## Ask questions about Looker data

When you're beginning a new conversation, Conversational Analytics suggests some starting questions to ask. The questions don't need to be in a specific format or use a specific syntax. However, they do need to relate to the Explore that you've selected.

Type your question in natural language into the **Ask a question** field. Select a [question mode](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#ca-question-mode) and click send. After you submit your query, you can cancel Conversational Analytics' response by clicking
**Stop response** . Conversational Analytics stops running the query and displays the following message: `The query was cancelled.`

For more guidance on the types of questions that you can ask, see [Limitations on questions](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#limitations-on-questions).

### Select a question mode

When you ask a question, you can select the question mode from a drop-down menu that contains the options [**Fast**](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#fast-mode) and [**Thinking**](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#thinking-mode). The UI describes **Fast** as being intended for answering quickly and **Thinking** as being intended for solving complex problems. Conversational Analytics defaults to **Thinking** question mode. Conversational Analytics will maintain the same question mode throughout any [multi-turn conversations](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#multiturn), unless you manually change it.

#### Fast mode

When you ask a question in **Fast** mode, Conversational Analytics attempts to directly map your natural language query to the LookML parameters that are defined in the LookML models that underlie your conversation's dashboard or Explores. Conversational Analytics is able to respond quickly because it relies on the governed definitions of your LookML, and it doesn't use or [display](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#show-thinking) any kind of reasoning.

For example, a query like "What was our total revenue last month?" can be quickly translated into a query that selects the `total_revenue` measure and filters for the previous month.

Select **Fast** mode if your query is asking for specific facts or predefined metrics from your data.

#### Thinking mode

**Thinking** mode is intended for more complex analytical requests that require analysis beyond direct LookML lookups. In this mode, the agent "plans" its approach, deciding which tools to use and how to combine results. This mode allows for solving multi-step problems and performing advanced data science tasks that may not be possible with a single SQL-based query.

Select **Thinking** mode when asking *why* about your data, when you're comparing trends, or when you're making more complex analytical requests that may require multiple steps. This mode is also especially helpful when testing an agent to understand how it's using the underlying LookML of its data sources.

<br />

### Multi-turn conversations

Conversational Analytics will take previous questions and answers into account as you continue the conversation. You can take previous answers and build on them by further refining results or changing the visualization type.

> [!IMPORTANT]
> **Important:** You can only reference previous questions in answers from the active conversation. You cannot reference previous questions and answers from other conversations.

For more guidance on creating questions, see [Limitations on questions](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#limitations-on-questions).

<br />

## Manage queries within a conversation

When you converse with data, you can manage the conversation by [stopping an active query response while it is running](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#stop-response) or by [deleting the most recent question and its response](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#delete-question).

### Delete the most recent question

To delete the most recent question and its response, follow these steps:

1. Hold your cursor over the most recent question, and then click **Delete message**.
2. In the **Permanently delete message?** dialog, click **Delete** to permanently delete the question and its response.

## Understand query results and calculations

Conversational Analytics provides the following parameters for understanding query
results and calculations:

- Details about [how your query was interpreted](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#show-thinking)
- Details about [how the answer was calculated](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#how-was-this-calculated)
- Additional details, such as [analytical details and debug info](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#analytical-details-debug-info), if enabled for the agent

### Determine how your query was interpreted

If you use [**Thinking** mode](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#thinking-mode) to ask your question, you can see how Conversational Analytics reasoned through your query. To see its reasoning, expand the **Show thinking** option. To hide its reasoning, click **Hide thinking**.

Conversational Analytics analyzes each query and thinks about how to respond, using the keywords from your query to infer the relevant dimensions, measures, and other parameters from the semantic layer of the conversation's associated datasets and interpreting from your query what aggregations may need to be performed. When you expand **Show reasoning**, Conversational Analytics displays a plain text explanation of the steps that it took to interpret your query. The explanation also includes the duration that Conversational Analytics thought about the query.

### Determine how an answer was calculated

To see how Conversational Analytics arrived at an answer or created a visualization, click **How was this calculated?** within the query results.

When you click **How was this calculated?** , Conversational Analytics displays a **Text** section. The **Text** section provides a plain text explanation of the steps that were taken by Conversational Analytics to arrive at the given answer. This explanation includes the raw field names that were used, the calculations that were done, the filters that were applied, the sort order, and other details.

### Analytical details and debug info

If the data agent that's used for the conversation has specific advanced options
enabled, responses can contain the following details:

- **Show thinking**: If enabled, the agent's response displays its reasoning steps, generated Looker queries, and underlying data tables.
- **Show debug info**: If enabled, the agent's response displays the steps that were executed by the agent for that response, which can help with troubleshooting.

These settings are configured in the advanced options when
[modifying the dashboard agent](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents-dashboards.md.txt#modify-dashboard-data-agent).

## Manage conversations

Each conversation remains in the **Recent conversations** section the dashboard conversation more_vert **Menu**. You can change the names of conversations, delete conversations, or restore them from the trash folder.

- To access your recent conversations with the dashboard agent, select more_vert **Menu** \> **Recent conversations**.
- By default, conversations will be named based on your initial question. To rename a conversation, select a conversation from the **Recent conversations** menu option, select more_vert, and then select edit **Rename**.
- To delete a conversation with a data agent, select more_vert **Menu** \> delete **Trash**.

### Delete a conversation

To delete a conversation with a data agent, select the conversation's more_vert three-dot menu, and then select delete **Delete**.

> [!NOTE]
> **Note:** Conversations that remain in the trash for more than 30 days are automatically deleted.

### Restore or permanently delete a conversation

To restore or permanently delete a conversation from the trash, follow these steps:

1. Select more_vert **Menu** \> delete **Trash**.
2. In the **Trash** pane, find the conversation that you want to restore or permanently delete. Select the conversation's more_vert three-dot menu, and then select one of the following options:

   - **Restore** : Restores the conversation. The conversation can be accessed from the **Recent conversations** menu option.
   - **Delete Permanently**: Permanently deletes the conversation.

## Known limitations

Conversational Analytics dashboard agents have the following known limitations:

- [**Advanced Analytics**](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-advanced-analytics.md.txt#suggested-questions) isn't supported for dashboard agents.
- Dashboard agents query the Production Mode of dashboard data.
- Dashboard agents inherit the dashboard's sharing model; access to the agent is governed by access to the underlying dashboard.
- You cannot modify the context or instructions for LookML dashboard agents.

### Limitations on visualizations

Conversational Analytics leverages [Vega-lite](https://vega.github.io/vega-lite) for conversation chart generation. The following Vega chart types are fully supported:

- Line chart (one or more series)
- Area chart
- Bar chart (horizontal, vertical, stacked)
- Scatter plot (one or more groups)
- Pie chart

The following Vega chart types are supported, but you may encounter unexpected behavior when rendering them:

- Maps
- Heatmaps
- Charts with tooltips

Chart types that exist outside the Vega catalog are not supported. Any charts that are not specified in this section are considered unsupported.


### Limitations on data sources

Conversational Analytics has the following data source limitations:

- Conversational Analytics can return a maximum of 50,000 rows per query.
- Conversational Analytics cannot set the value of a filter-only field that is defined using the LookML [`parameter`](https://docs.cloud.google.com/looker/docs/2612/reference/param-field-parameter.md.txt) or [`filter`](https://docs.cloud.google.com/looker/docs/2612/reference/param-field-filter.md.txt) parameters.

### Limitations on questions

Conversational Analytics supports questions that can be answered by a single visualization, for example:

- Metric trends over time
- Breakdown or distribution of a metric by dimension
- Unique values for one or more dimensions
- Single metric values
- The top dimension values by metric

Conversational Analytics doesn't yet support questions that can only be answered with the following types of complicated visualizations:

- Prediction and forecasting
- Advanced statistical analysis, including correlation and anomaly detection

## Related resources

- **[Conversational Analytics in Looker overview](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-overview.md.txt)**: The landing page for Conversational Analytics with a list of key features links to all Conversational Analytics documentation.

- **[Query an Explore in natural language](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data.md.txt)**: Start a conversation with a Looker Explore to ask about Explore data using natural language.

- **[Create and manage Explore data agents](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-data-agents.md.txt)**: With Explore data agents, you can customize the AI-powered data querying agent by providing context and instructions that are specific to your Explore data, which helps Conversational Analytics generate more accurate and contextually relevant responses.

- **[Best practices for configuring Conversational Analytics in Looker](https://docs.cloud.google.com/looker/docs/2612/conversational-analytics-looker-best-practices.md.txt)**: Strategies and best practices to help Looker administrators and LookML developers successfully configure and optimize Conversational Analytics.