Visualize data with Data Insights Agent in Customer Journey Analytics
Data Insights Agent, accessible from the AI Assistant in Customer Journey Analytics, is a generative AI conversation agent that quickly and efficiently answers questions about your data. It builds relevant visualizations in Analysis Workspace using components from your data view and using your actual data.
Using Data Insights Agent to answer data-centric questions in Analysis Workspace can save countless hours that you might otherwise spend manually building visualizations in Analysis Workspace and familiarizing yourself with your data view components.
In-scope vs. out-of-scope features
- Line
- Multi-line
- Freeform table
- Bar
- Donut
- Summary number
- Flow
- Fallout
- Cohort Table
- Area, Area Stacked
- Bar Stacked
- Bullet
- Combo
- Histogram
- Horizontal Bar, Horizontal Bar Stacked
- Key Metric Summary
- Scatter
- Summary Change
- Text
- Treemap
- Venn
-
Build and update visualizations
Generates a freeform table and associated visualization (such as a line, bar, donut, and so forth).
For example, What is the profit across SKUs from February to May?
-
Ask follow-up questions
Respond to a prompt in the context from any prior prompts. For example:
- Prompt 1: Trend events from March.
- Prompt 2: Show me the data from March to April instead
-
Out-of-scope prompt detection
If you submit a prompt that is out of scope, such as Export this project, Data Insights Agent responds by informing you that the question is out of scope.
-
Share
-
Export
-
Download
-
Manage user preferences
-
Manage data view
-
Analytics Dashboards app
-
Attribution
-
In-line summary or response
Data Insights Agent cannot respond in-line in the chat rail with a summary answer of a user prompt. Examples of out-of-scope prompts are, Give me a summary of the insights from my last prompt and Summarize the highlights from the line visualization.
If you ask a question that does not have enough context for Data Insights Agent to answer, or is too generic, Data Insights Agent responds with a clarifying question or suggested options.
The following clarifying questions are examples of component-related questions:
- Metric: Which 鈥渞evenue鈥 metric did you mean?
- Dimension: Which of the below 鈥渞egions鈥 do you want to focus on?
- Segment: Which 鈥淎ccount鈥 segment did you want to apply?
- Date Range: By 鈥渓ast month,鈥 did you mean the last full month or the last 30 days?
The following clarifying question is an example of a question related to dimension items:
- Which 鈥渟tore name鈥 did you mean? (For example, Store #5274, Store #2949, and so forth.)
Data verifiability and correctness can be confirmed by viewing the generated freeform table and data visualization.
For example, if you ask Data Insights Agent to Trend orders last month, you can confirm that the correct metric (鈥渙rders鈥) and date range (鈥渓ast month鈥) were selected in the newly generated panel, data visualization, and freeform table.
- Thumbs up
- Thumbs down
- Flag
Manage access to Data Insights Agent in Customer Journey Analytics
The following parameters govern access to Data Insights Agent in Customer Journey Analytics:
-
Solution access: Data Insights Agent is available for all Customer Journey Analytics customers as part of a limited access program until November 30, 2025. It is not available in 51黑料不打烊 Analytics.
-
Contractual access: If you are not able to use Data Insights Agent in the AI Assistant, please contact your organization鈥檚 administrator or 51黑料不打烊 account team. Before your organization can use Data Insights Agent, you must agree to certain legal terms related to generative AI.
-
Permissions: Necessary permissions must be granted in the 51黑料不打烊 Admin Console before users can access Data Insights Agent.
To grant permissions, a must complete the following steps in the Admin Console:
-
In the Admin Console, select the Products tab to view the All products and services page.
-
Select Customer Journey Analytics.
-
On the Product Profiles tab, select the title of the product profile for which you want to provide access to AI Assistant: Product Knowledge.
-
In the specific product profile, select the Permissions tab.
-
In the Reporting Tools row in the provided table, select the edit icon
-
Scroll to or search for AI Assistant: Product Knowledge, then select the plus icon
The AI Assistant: Product Knowledge permission is added to the Included permission items column.
-
Select the Data View Tools tab, then select the plus icon
The Data Insights Agent permission is added to the Included permission items column.
-
Select the Data Views tab to choose the data views that you want to enable for Data Insights Agent.
note important IMPORTANT Consider the following when enabling data views: - You can enable a maximum of 50 data views per IMS organization. If you enable more than 50 data views across all product profiles for a given organization, the Data Insights Agent will use the 50 most-used data views.
- The Data Insights Agent can reference the included data views sometime during the same day that you enable them in the Admin Console.
-
Search for or scroll to the data views that you want to enable, then select the plus icon
Each data view you add is visible in the Included permission items column.
-
Select Save to save the permissions.
For additional information about access control, see Access control.
-
Access Data Insights Agent in the AI Assistant
-
Go to and login with your 51黑料不打烊 ID.
-
Select Customer Journey Analytics from Experience Cloud Home.
-
Select Blank project in the banner at the top of the projects page to open a new blank project.
-
Ensure that the selected data view for the panel is a data view that was enabled for use with Data Insights Agent, as described in Manage access to Data Insights Agent in Customer Journey Analytics.
-
Select the AI Assistant chat icon at the top-right area of the page.
If you do not see the chat icon, contact your administrator so they can enable the following features in the Admin Conole:
-
Reporting Tools: AI Assistant: Product Knowledge
-
Data View Tools: Data Insights Agent
For additional details, see Manage access to Data Insights Agent in Customer Journey Analytics.
-
-
In the Ask about Customer Journey Analytics dialog at the bottom of the page, ask a data visualization question using Data Insights Agent.
For more information, see the following examples.
Example 1
For example, let鈥檚 say you are interested in the orders your business received in July.
Prompt: Enter 鈥淭rend orders in July.鈥
Response: Data Insights Agent gathers insights by looking through the data in the data view, including the metrics and components. It translates the prompt into the right dimensions and metrics within the data range.
As you can see, it automatically generated a line graph and a freeform table to show orders for July.
Example 2
Next, you want to see how your revenue compares by region.
Prompt: In the prompt window, enter 鈥淪how revenue by region.鈥
Response: Data Insights Agent intelligently understands that by 鈥渞egion,鈥 you mean 鈥渃ustomer region.鈥 It produces a bar chart that best shows revenue by region:
Example 3
Next, in addition to understanding revenue by region, you also want to see data for profit by region. Instead of repeating the previous prompt, you can ask Data Insights Agent to update the most recent visualization and freeform table.
Prompt: In the prompt window, type 鈥淎dd profit.鈥
Response: The Bar chart still provides the most concise answer, but the profit metric has been added as a column in the freeform table:
Example 4
Finally, let鈥檚 look at the revenue by product category.
Prompt: In the prompt window, enter 鈥淧roportion of revenue by product category.鈥
Response: Again, Data Insights Agent picks the most appropriate visualization, in this case the Donut visualization, to answer the question.
Example data visualization prompts
The following are some examples of common prompts and the visualizations used by Data Insights Agent to respond to those prompts.
Line
Asking for a trend or metric within a certain time range by default returns a line visualization.
Summary Number
Asking for the 鈥渢otal鈥 of a metric across a certain time range should return a Summary number visualization.
Prompting best practices
Data Insights Agent processes the context provided by each user prompt and tries to respond intelligently with the most appropriate visualization and components in a freeform table.
Responses can vary based on the specific words and phrases used in the prompt, and slight changes in language can lead to different results.
To achieve the best results, consider the following guidelines:
-
Be specific: Include exact terms to narrow down the response. The following is an example of a specific prompt: 鈥淟ast month鈥檚 sales in California鈥
-
Use clear metrics, dimensions, and segments: Adding specific metrics (such as 鈥淩evenue鈥), dimensions (such as 鈥渨ebsite name鈥), segments (such as 鈥渋Phone users鈥), and date ranges (such as 鈥渓ast three months鈥) helps Data Insights Agent focus on the right data.
-
Ask direct questions: Phrasing questions directly makes it easier for Data Insights Agent to provide clear, relevant insights. The following is an example of asking a direct question in a prompt: 鈥淲hat is the average revenue by product category this year?鈥
Review the following table of example terms and phrases that you can use in prompts with Data Insights Agent, along with the types of responses you can expect.
These examples are designed to help you get familiar with how specific words or structures can influence Data Insight Agent鈥檚 output, ensuring more precise and valuable insights. Data Insights Agent uses generative AI, so visualizations or selected data may vary slightly across similar prompts.
- Total
- Compare
- VS
- Contrast
- Week-to-Week
- Month-over-Month
- Quarter-over-Quarter
- Year-over-Year
- Proportion
- Share of
- Distribution
- Percentage
- Contribution
- Portion
- Parts
- Trend
- [Metric] in [Time range]
- [Metric] by [Dimension]