Best practices for Data Quality Rule IDs: Proposed Taxonomy design & Automating generation at scale
Hi Collibra Community,
We are currently re-architecting our naming conventions and structural standards for Data Quality Rule assets in Collibra, integrated with an external Data Quality execution engine (Ataccama ONE).
Our Current Setup & The Legacy State
Technical Integration: We leverage Collibra's native 128-bit
UUIDas the immutable physical key, mapped directly to the DQ engine'sexternalIdvia REST APIs for metric write-backs.Current Legacy IDs: Today, our logical rule IDs rely on a simple legacy pattern like
V01,V02, etc. Assigned manually, this approach lacks business context, carries a high risk of duplicate sequences, and clearly does not scale as rule volume grows into the hundreds or thousands.
The Proposed Idea & Problem Statement
We want to replace V01 with a structured, human-readable taxonomy. We are considering the following pattern:
$$\text{[DOMAIN]\_[SYSTEM]\_[TABLE/ENTITY]\_[DQ\_DIMENSION]\_[SEQUENCE]}$$
(Example: FIN_SAP_CUST_CMP_001)
However, our leadership correctly pointed out that we should not reinvent the wheel. We want to validate both the design of this taxonomy and the method of generating it to ensure long-term scalability without human friction.
Questions for the Community
Taxonomy Best Practice: Is
[DOMAIN]_[SYSTEM]_[TABLE/ENTITY]_[DQ_DIMENSION]_[SEQUENCE]aligned with industry standards for Collibra, or does it become too rigid in practice? What naming pattern has worked best in your mature deployments?Automated Generation: How do you automate business ID generation to eliminate manual typing by Data Stewards? Do you enforce this exclusively via Collibra BPMN Workflows (Activiti / Groovy scripts) during the rule intake form?
Sequence Counters: What is the recommended approach to maintain dynamic, auto-incrementing sequence counters (e.g., finding the next available
_001,_002per domain/system) within Collibra?Out-of-the-Box / Marketplace Solutions: Are there OOTB Collibra features, Workflow building blocks, or Marketplace accelerators specifically designed for auto-incrementing asset codes?
Reliable Resources & Reference Materials: If possible, could you share links to official Collibra documentation, whitepapers, community articles, or reliable study resources that cover DQ rule asset modeling and automated ID generation best practices?
We would love to hear your insights on both the naming design, automation mechanics, and any reference documentation you can recommend so we can build a 100% hands-off, scalable solution!
Thanks in advance for sharing your experience!
