Notification

This Help Center is moving to Cloud. Learn more about the migration.

Query your data in natural language with Gemini assistance (Preview)

Conversational Analytics can help you ask questions about about your data using natural language. 
Preview
 
This product or feature is subject to the "Pre-GA Offerings Terms" in the General Service Terms section of the Service Specific Terms. Pre-GA products and features are available "as is" and might have limited support. For more information, see the launch stage descriptions.

Conversational Analytics is an AI-powered data querying tool that lets people write questions in natural language, empowering users with no expertise in business intelligence to gain value from your data. To produce the most reliable answers possible, Conversational Analytics uses your LookML models to understand how it should query your data.

Gemini in Looker doesn't use your prompts nor its responses as data to train its model. For more information, see How Gemini in Google Cloud uses your data.
As an early-stage technology, Gemini in Looker can generate output that seems plausible but is factually incorrect. We recommend that you validate all output from Gemini before you use it. For more information, see Gemini in Google Cloud and responsible AI.

Requirements to use Conversational Analytics

To use Conversational Analytics, you must meet the following requirements.

  1. You must be a user under a Looker Studio Pro subscription. Looker Studio Pro licenses are available at no cost to Looker users.
  2. An administrator has enabled Gemini in Looker using the instructions in the Enable and disable Gemini in Looker for Looker Studio article.
  3. You want to analyze a dataset that is in Google BigQuery, Google Sheets, a CSV file, a data extract, or Looker (which is a different product than Looker Studio).

Data source setup

Before you can ask questions of your data, you must have data sources connected to Looker Studio. Conversational Analytics works with the Google BigQuery, Google Sheets, CSV, data extracts, and Looker data connectors.

Instructions for adding these data types can be found in the following articles:

You can view the data sources that have already been added to your Looker Studio account at https://lookerstudio.google.com/c/navigation/datasources.

Be aware of these data source limitations:

  • Conversational Analytics does not support BigQuery’s Flexible Column Names feature.
  • Conversational Analytics does not work well with data sources that have field editing in reports disabled because this setting prevents Conversational Analytics from creating calculated fields.
  • When the data source is Looker, Conversational Analytics cannot override the default value of an always_filter or a conditionally_filter parameter.
  • When the data source is Looker, Conversational Analytics cannot set the value of a parameter.

Navigate to Conversational Analytics

Conversational Analytics can be accessed in three ways:

  1. Navigate directly to https://lookerstudio.google.com/c/u/0/notebook.
  2. Choose Conversational Analytics from the left navigation panel of Looker Studio.
  3. Choose Conversation from the Create menu of Looker Studio (if you have selected the “Sandbox” project).

Manage Conversations

Sets of questions that you ask about a data set are organized by Conversation. Splitting work into multiple Conversations can be useful for organizing lines of inquiry. Conversations are listed in the left side panel of Conversational Analytics. Clicking on any existing Conversation lets you continue your investigation and ask additional questions.

Create a new Conversation

To create a new Conversation, click the New button in the upper left of Conversational Analytics. Questions and results are saved for future viewing in the “Sandbox” location of the project that you choose.

The first required step for a Conversation is to choose which data source you want to investigate. To choose a data source, click an existing data source, or create a new one by choosing Connect to data.

If you’re having trouble making a data source appear, validate that it’s one of the supported data connector types.

Name a Conversation

Conversational Analytics automatically generates a Conversation’s title based on your first question and response, but you can provide a custom name for a Conversation by clicking on the default and typing a new name:

Delete a Conversation

To delete a conversation, click the trash icon in the top right corner of the Conversation. You can find deleted Conversations in the Trash section in the left pane of Conversational Analytics, where they can be restored or permanently deleted. Click on a Conversation in the Trash to restore or permanently delete it.

Conversations that remain in the Trash for more than 30 days are automatically deleted.

Search Conversations

You can search a Conversation by using the Search Conversational Analytics search bar at the top of the page.

Ask Questions

Ask questions about a data source

Once you have created a Conversation, you can type questions about the data into the Ask a question field at the bottom of the screen. The questions do not need to be in a specific format or use a specific syntax. However, they do need to relate to the data source that you’ve selected. Conversational Analytics will take previous questions and answers into account as you continue the Conversation.

Conversational Analytics supports questions that can be answered by a single Looker Studio 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 does not yet support questions such as the following:

  • Percent change of a metric over time, including period-over-period analysis
  • Prediction and forecasting
  • Advanced statistical analysis, including correlation and anomaly detection

For more guidance on crafting questions, see the Best Practices section of this article.

Suggested questions

If you’re not certain where to begin, Conversational Analytics suggests starting questions at the bottom of a Conversation under What questions can I ask?. Click a suggested question to generate an answer.

Edit questions

You can modify your most recent question in a Conversation by clicking the pencil icon next to the prompt. The question can then be edited and updated. If you choose to update the question, the previous prompt and response is removed and replaced with the updated prompt.

Get additional insights

When Conversational Analytics is able to provide additional data insights about a response, an Insights button will appear below the response. Click it to see additional information about your query. The Insights feature only analyzes the data that was returned by your prompt and will not run additional queries to fetch additional data.

Determine how an answer was calculated

To see how Conversational Analytics arrived at an answer or created a visualization, click on How was this calculated? below the response. This will provide an overview of the steps taken by Conversational Analytics, including the fields that were used, the calculations that  were done, the filters that were applied, and other details.

Edit an answer

After clicking the How was this calculated? option below a response, you may want to adjust the fields that were used. Click Edit, and then click any field name that you want to change and select your preferred field from the drop-down menu. Edits are indicated by a darker background color. When you are finished, click the Update button to apply your changes and generate a new response.

Note that, while Conversational Analytics will generate a new answer, the original prompt that you provided does not change. However, an indication appears next to the timestamp to show that the answer has been edited.

Edit the chart type of a response

To edit the chart type that is returned by Conversational Analytics, click the Edit chart type button below each response. For example, if the data is shown in a pie chart but you would prefer a bar chart, you can make that change.

Note that every chart type isn’t available for every response, because chart types are dependent on the types of fields that Conversational Analytics chooses to return.

Add results to a Looker Studio report

Results can be opened in a new Looker Studio report for further customization, sharing, or validation. Click the Open in report link at the bottom right of most results to open the results in a new report.

The Open in report feature won’t be available when a visualization contains a filter that is based on a calculated field. When the feature is unavailable, the Open in report option will be grayed out.

Best practices

Setting up a data source

Setting up a data source in the ideal manner can help Conversational Analytics provide the most helpful answers. Consider following these best practices when creating a data source:

  • Only include fields in the data source that should be used for analysis by end users. You can exclude other fields entirely, or hide them in the data source.
  • Give each field a clear and concise name. For Looker data sources, the field labels that are defined in Looker are automatically used by Looker Studio.
  • Give each field a clear description, including example values where relevant. These field descriptions are included in the prompt that is sent to Conversational Analytics, and they can be helpful for providing context. Example values are especially helpful for string fields.

Prompting

When writing questions for Conversational Analytics, consider following these best practices:

  • Use the exact field names that are included in the data source when possible. This will help Conversational Analytics disambiguate similarly named columns. You can view these field names by clicking View data source in the data sidebar.
  • To narrow down results by including or excluding specific data, state the field and the filter value directly when possible. For example, instead of asking for “German sales”, ask for “sales where the country is Germany”, or “sales where the region is DE”.

Provide feedback

You can provide feedback to Google about Conversational Analytics in either of the following ways:

  • Click the thumbs-up or thumbs-down icon below each response.
  • Click the Send Feedback button at the bottom of the left-hand pane.

If you’re sharing negative feedback, you’ll have the option to include additional details, including a copy of the Conversation that you were having. Please do not share sensitive information with Google such as passwords, payment information, or personal customer details.

Was this helpful?

How can we improve it?
Search
Clear search
Close search
Google apps
Main menu
15552326529023037843
true
Search Help Center
true
true
true
true
true
102097
false
false