Enterprise Graph initial setup in production instance

---
title: Enterprise Graph initial setup in production instance
description: Setup and use Enterprise Graph Schema, a unified Knowledge Graph schema, that captures all the ServiceNow and third-party tables and their connections.
locale: en-US
canonical_url: https://www.servicenow.com/docs/r/intelligent-experiences/knowledge-graph/initial-setup-for-enterprise-graph-schema.html
release: brazil
product: Knowledge Graph
classification: knowledge-graph
topic_type: task
last_updated: "2026-09-10"
reading_time_minutes: 1
breadcrumb: [Using Enterprise graph schema, Knowledge Graph, Managing data for AI, Enable AI Experiences]
---

# Enterprise Graph initial setup in production instance

Setup and use Enterprise Graph Schema, a unified Knowledge Graph schema, that captures all the ServiceNow and third-party tables and their connections.

## Before you begin

When you install Enterprise Graph, it automatically begins configuring instance tables. This may take several days and until the setup is complete, the schema won’t respond effectively to queries.

If you have already completed the setup in sub-production instance, you can import the generated descriptions to production instance and fasten the process.

Assuming that you have completed the setup in sub-production instance, do the following to complete the initial setup in production instance:

-   Create an integration user, in the sub-production instance. For more information, see [Create a user](https://raw.githubusercontent.com/ServiceNow/ServiceNowDocs/brazil/markdown/platform-administration/t_CreateAUser.md)
-   Verify if the descriptions are generated in sub-production instance.
-   Use the following procedure to:
    -   Import the data into production.
    -   Run the associated Transform Maps.
-   Remove tracker records from the production instance.

Role required: admin

## Procedure

1.  Navigate to **All** > **Connections & Credentials** > **Connection & Credential Aliases**.

    \[Omitted image "image.connection-and-credntials-page"\] Alt text: Connection and credentials navigation

2.  Open the record with name **Description\_Connector**.

3.  Select **Create New Connection & Credential** from the related links section.

    \[Omitted image "image.connection-and-credentials-aliases"\] Alt text: Create connection and credentials aliases

4.  Enter the required details in the **Create Connection & Credential** form.

    \[Omitted image "image.create-connection-and-credntials"\] Alt text: Create connection and credentials form

5.  Add the user name and password that was used, while creating the new user.

    See [Create a user](https://raw.githubusercontent.com/ServiceNow/ServiceNowDocs/brazil/markdown/platform-administration/t_CreateAUser.md) for more information.

6.  Navigate to **All** > **System Import Sets** > **Administration** > **Data Sources**.

7.  In the name field search bar, enter **KG** to view the related tables.

    The following tables are displayed:

    -   KG Choice Picker Description.
    -   KG Column Picker Description.
    -   KG Table Picker Description.
    -   KG Triplet Picker Description.
    \[Omitted image "choice-tables-kg.png"\] Alt text: Picker description options

8.  Open one of the displayed table and select **Load All Records** from the Related links section.

    Ensure the data is loaded.\[Omitted image "image.table-picker-kg"\] Alt text: KG Table picker

9.  Navigate to **System Import Sets** > **Advanced** > **Import Sets**, once the data is loaded.

    For detailed information, see import set documentation [Import sets](https://raw.githubusercontent.com/ServiceNow/ServiceNowDocs/brazil/markdown/integrate-applications/import-sets-landing-page.md)\[Omitted image "image.import-set-kg"\] Alt text: Import set

10. Open the import set and select **Transform** from the Related links section.

    For detailed information, see Transform map documentation [Transform maps](https://raw.githubusercontent.com/ServiceNow/ServiceNowDocs/brazil/markdown/integrate-applications/c_CreatingNewTransformMaps.md)

11. Select **Transform** to complete the transformation.

    \[Omitted image "image.transform-page-kg"\] Alt text: Transform option

12. Navigate to the selected table and verify if the data is loaded.

13. Repeat the import set and transformation steps for other three tables.