---
sourceDocument: Brazil Platform Analytics
sourceDocumentLink: https://www.servicenow.com/docs/r/now-intelligence

 Release :

    - brazil

ft:locale :

    - en-US

ft:publication_title :

    - Brazil Platform Analytics

ft:clusterId :

    - par

bundleId :

    - par

workflow :

    - Platform


---

# Questions and responses in an exploration

# Questions and responses in an exploration {#ariaid-title1}

Release version: Brazil  
Updated September 10, 2026  
![](https://www.servicenow.com/docs/portal-asset/ico-clock) 5 minutes to read  
Ask the AI specific questions in AI Data Explorer, to which it responds with data visualizations, a summary, and suggested follow-up questions.

To ask a question in an exploration, launch AI Data Explorer from a data visualization or list or open an existing exploration. You will see a field with the placeholder "Ask a question about data." For more information, see [Launch AI Data Explorer](https://www.servicenow.com/docs/WiIVShgJ9yI2c9SdsH_9AQ "You can launch AI Data Explorer from either unified navigation or a data visualization or list . If you have the right role, you can create an exploration.").  
Note:  
* The question you ask has to be about data in either indicators or one of the tables listed in the Query Generation Semantic Table Configuration table. These tables can include database views or Workflow Data Fabric tables. For more information, see [Add a table to the semantic data layer](https://www.servicenow.com/docs/E3BMoMkuGdhpOLEm_VXDig#add-table-semantic-layer "Add more tables to the Query Generation semantic layer so that users can use ServiceNow Otto for Platform Analytics applications to ask questions about the data in those tables.").
* The system first looks for a relevant indicator to be the data source. If it does not find one, it falls back on table data sources.
* If the data is from a protected application scope, access to that scope must be configured for AI Data Explorer. For more information, see [Enabling access to protected scope applications for AI Data Explorer and Query Generation](https://www.servicenow.com/docs/OSIvck0g675MxcYZ9GHeUQ "AI Data Explorer and Query Generation require additional security configuration to access tables in protected scopes such as Human Resources or Employee Profile scopes.").
* When you have submitted a question, you cannot submit another question or do other work in the exploration until your question is processed. You can cancel the processing of your question.
{#ask-expl-questions__ul_p5q_zsl_tgc}

When you write a question in an exploration, the AI converts the question to a database query and returns a response. The response includes the following sections:  

<br />

* ![Area 1]() An expandable set of actions to take on the response. For more information, see [Duplicate, delete, copy, or move an answer in an exploration](https://www.servicenow.com/docs/J_0PtPfWcARuDPzdZQGW6w "Duplicate, delete or reorder an individual question and response from inside an exploration. Yiou can also copy a response to another exploration.").
* ![Area 2]() Your original question. You can edit this question to generate new output.
* ![Area 3]() The title of the response and a summarization of the AI findings.
* ![Area 4]() If extended analysis is enabled, you get additional insights after the title and summary. For more information, see [Extended analysis](https://www.servicenow.com/docs/hkJC7hoQOoXsMc9je1q4Hw "Generate a deeper level of analysis that can reveal new insights, enabling you to make more informed decisions.").
* ![Area 5]() A list or data visualization. This response can be an existing visualization instead of a generated one. For more information, see [Launch AI Data Explorer](https://www.servicenow.com/docs/WiIVShgJ9yI2c9SdsH_9AQ "You can launch AI Data Explorer from either unified navigation or a data visualization or list . If you have the right role, you can create an exploration.").  
  You can add the list or visualization to a dashboard or change its height by interacting with controls in its corner. Point at the corner to make the controls appear. For more information, see [Add a data visualization from an exploration to a dashboard](https://www.servicenow.com/docs/WQQ2X18nfTGk53uBT_JQhA "Put the visualization contained in a response from AI Data Explorer on a new or existing dashboard. Do so without interrupting your workflow in your exploration.").  
  ![Controls in the corner of a data visualization, with height adjustment control selected.]()  

  If the data visualization is on an indicator, information about the indicator score is displayed. If the data visualization is a single score, sparkline and score change information is included.
{#ask-expl-questions__ul_bdh_tqg_xfc}

## Viewing the response source {#ask-expl-questions__section_lvx_3jc_rgc}

After you receive a response from the ServiceNow AI Platform, point at the response to see the technical details of the response. The source details for a table source include the following information:

* The source table
* The filter conditions
* The metric
* Any grouping criteria
{#ask-expl-questions__ul_fm5_4tk_tgc}

For an indicator source, the details include the time series aggregation and the collection date.  
If the exploration is too narrow on the screen, select View source instead of pointing at the response.  

## Tips for asking questions {#ask-expl-questions__section_kbr_yx1_cgc}

The goal of AI Data Explorer is to understand your prompts in your own words, delivering the analytics insights you want. However, if you do not know where to begin to formulate questions, or you're unsatisfied with the
results, here are some tips:  

Name your table{#ask-expl-questions__dlentry_name-your-table}

:   If you know the name of the table that contains the data you are interested in, add it to your prompt. Partial names or similar names are fine too.Example: Instead of "How many P1s were opened this week," write "How many
    P1 requests were opened this week," which references the request tables. Better yet, write "How many P1 catalog requests were opened this week," which references the specific Catalog Requests table.

Explain what you mean{#ask-expl-questions__dlentry_explain-what-you-mean}
:   Query Generation tries to understand your terms, but you can add details to help guide it. If you get unexpected results, try being more specific about what you're looking for.

    Example: Instead of "Show me all stale
    incidents," write "Show me all incidents not updated in 5+ days."

Be specific with names{#ask-expl-questions__dlentry_be-specific-with-names}

:   When filtering by referenced records like users, groups, or services, try to use their full display names for best results. The AI model may learn from previous queries in the same document, but using full names ensures accuracy.Example: Instead of "Cases with Workplace Ops," write "Cases with Workplace Operations."

Edit and refine queries{#ask-expl-questions__dlentry_edit-refine-queries}
:   If the generated query isn't quite right, you can manually edit the filter conditions. The AI model will learn from your edits and apply them to future questions in the same document. For more information, see [Regenerate a response in an AI Data Explorer exploration](https://www.servicenow.com/docs/Y4HMWmbzWf_sXgT9ZIKwMQ "Change the filter conditions for a table source or data visualization parameters for an indicator source. Then regenerate a response with updated visualizations.")

    Example: You ask "Show me critical incidents from the network team" but are not satisfied with the response. Instead of asking repeated variations of the same question,
    hoping for a better result, edit the filter to find records where Assignment Group is 'Network Operations' and Priority is '1 - Critical'. Then ask "Show me the inflow trend for these incidents over time".

Don't leave bad queries in your exploration{#ask-expl-questions__dlentry_delete-bad-queries}
:   The AI model uses the previous document context to write the next query. Therefore, if you cannot refine a query to get a useful response, delete it. Otherwise bad queries can accumulate in your exploration, leading to
    ever-worsening responses.

Import complex filters{#ask-expl-questions__dlentry_import-complex-filters}

:   For complex data that's hard to describe, import data visualizations or lists into your exploration. If the visualization or list is on a dashboard, you can apply any filters on the dashboard before importing. The AI model will use imported queries to understand related questions in the same document.Example: Don't ask "Show me servers about to retire by location." Such a prompt is vague and complex. Instead, import a visualization from a
    dashboard titled "PostgreSQL servers nearing retirement," with the desired values for the dashboard filters Lifecycle State and Days Until Retirement pre-applied. Then ask "Show me the same servers but grouped by
    location".

Once you have a productive exploration going, with a lot of context, you may find that you can ask more abstract questions and get useful answers. However, these tips might help you get started.
* **[Indicator vs Table data source selection](https://www.servicenow.com/docs/MrUsI45a7Y8JvmJ9vrcBZQ)**   
  After you submit a query to AI Data Explorer, the system checks the query for specific keywords that indicate whether to use table or indicator data. If there are no such keywords, it falls back on a default set in a system property.
* **[Extended analysis](https://www.servicenow.com/docs/hkJC7hoQOoXsMc9je1q4Hw)**   
  Generate a deeper level of analysis that can reveal new insights, enabling you to make more informed decisions.
* **[Add a data visualization from an exploration to a dashboard](https://www.servicenow.com/docs/WQQ2X18nfTGk53uBT_JQhA)**   
  Put the visualization contained in a response from AI Data Explorer on a new or existing dashboard. Do so without interrupting your workflow in your exploration.
* **[Refresh a response with new data](https://www.servicenow.com/docs/jgAoATkqQaAl~xsCA4GkJw)**   
  Look at the age of the response to a question in an AI Data Explorer exploration. Then regenerate the response with fresh data.
* **[Change the question in an exploration](https://www.servicenow.com/docs/N4UcQsH9dWoMWbn3cHRcIw)**   
  Edit a question in an AI Data Explorer exploration and submit it to overwrite the original response.
* **[Regenerate a response in an AI Data Explorer exploration](https://www.servicenow.com/docs/Y4HMWmbzWf_sXgT9ZIKwMQ)**   
  Change the filter conditions for a table source or data visualization parameters for an indicator source. Then regenerate a response with updated visualizations.
* **[Duplicate, delete, copy, or move an answer in an exploration](https://www.servicenow.com/docs/J_0PtPfWcARuDPzdZQGW6w)**   
  Duplicate, delete or reorder an individual question and response from inside an exploration. Yiou can also copy a response to another exploration.