Metric Insights is compatible with a wide range of external tools, including data governance software such as Collibra, Atlan, and Alation. The connection is established via the corresponding plugins and allows you to transfer not only basic data but also metadata, such as Glossary terms and Custom Fields. Read this article to learn the recommended best practices for governing metadata, which specialists will be involved in establishing the connection, and what choices need to be made before setting it up.
Connecting to External Tools
Data flows in a single direction: from your source tool to the Metric Insights plugin, and from there into Metric Insights. Depending on your requirements, a plugin can either import data directly into Metric Insights (as Datasets and Metrics) or catalog and embed reports that continue to reside in your source tool (as External Reports).
Metric Insights integrates with a broad range of third-party tools. For a complete overview, refer to the full list of supported Metric Insights' plugins.
Decision 1: Auth Type
To establish a connection, the plugin goes through an authentication process, so you need to decide which Auth Type to use:
- Password: Authentication uses the credentials of a user account (Username and Password).
- Identity Profile: Identity Profiles allow you to store user credentials outside the MI application using various security access and management tools. Each time credential retrieval is triggered, the MI application requests the credentials from the security access and management tool.
- Service Principal: If the tool supports Service Principal creation, you can select this Auth Type and provide the Client ID and Client Secret to authenticate.
- OAuth: If your tool supports OAuth, authentication can be performed using the Client ID and Client Secret generated in your tool account.
Not every tool supports all of the listed authentication options; the list for each tool includes only the options that apply to it.
As a best practice, use token-based authentication (like Service Principal or User OAuth) wherever the tool supports it, as it is the most secure and convenient option.
Decision 2: Remote Data Agent
A Remote Data Agent is a Windows-based service that allows Metric Insights to connect to Data Sources that require Windows integration or are located behind a firewall separating them from the platform.
You only need a Remote Data Agent when Metric Insights cannot reach the governance tool directly. Whether that applies usually depends on how your catalog is hosted:
- Self-hosted or on-premises catalogs, such as Collibra or Alation running inside your network behind a firewall or VPN, typically require a Remote Data Agent.
- Cloud (SaaS) catalogs, such as Atlan, Purview, Dataplex, or the cloud versions of Collibra and Alation, usually don't need one. A direct connection is simpler: you only need to allow outbound access from Metric Insights to the vendor’s API.
Metadata and Glossary Synchronization Process
Glossary Terms and Metadata fields can be created and populated manually in Metric Insights, but it is far more efficient to do so automatically. However, to tag Metric Insights Elements (such as Reports) automatically and accurately with Glossary Terms and Metadata fields from your governance tool, the source Dataset must contain a common identifier that exists in both systems. For example, Metric Insights needs to know that a Term named “Revenue” in the data governance software applies to the “Q4 Sales Dashboard” in Metric Insights.
This is achieved by using a dedicated column in the Dataset that links the two systems.
Decision 3: Key Column
You must configure the Dataset to include one of the following keys:
- External BI Tool Report ID (recommended): The unique GUID from the tool database. This is the most robust method, as the ID persists even if the Report name changes.
- Element Name: The exact name of the Report in Metric Insights. This option is less reliable; if a Report is renamed, the association may break.
- External GUID: A global identifier, if one is maintained consistently across both systems.
The full process of connecting the tool to Metric Insights, including the steps to synchronize Glossary Terms and Metadata, is described in Metadata Sync & Enrichment Workflow Playbook.
Required Administrator Roles
| Role | Responsibilities |
|---|---|
| Data Governance Tool Administrator |
|
| Identity (IAM) Lead | Ensure "Owner" names in the tool exactly match User IDs or Identity Profiles in MI to prevent assignment failures. |
| Metric Insights Administrator |
|
Pitfalls to Avoid
- Ignoring the prerequisites listed for each Data Source in its corresponding article.
- Forgetting to check metadata for duplicates. During the automated sync, the system skips duplicate entries to prevent conflicts.
- Skipping the check that User IDs or Usernames match between the data governance software and the MI authentication section. If they differ, the connection will fail.
Recommendations Before Start
Before you begin the technical implementation, we recommend that you:
- Ensure that each combination of Glossary Term and Section is unique within the source data. If duplicates exist, the system skips those records to prevent conflicts.
- Identify which Metadata fields are Standard (such as Description and Owners).
- Identify which Metadata fields are Custom (such as Formula and URL Reference). You will need to create a Custom Field for each of them during setup.
- Strictly define the input type for each Custom Field (for example, “Text” for formulas, and “URL” or “Text” for links) to ensure the Custom Field Editor accepts the incoming data format.
- Determine whether the synchronization should be a Snapshot (where only new terms are loaded during each update) or a full reload. To maintain a synchronized mirror of Collibra, a full reload is generally recommended, as it captures all updates to existing definitions.