IT Operating Environments Best Practices

Overview

Dirty data - data that is incomplete, inaccurate, improperly formatted, or otherwise failing the quality standards required for the governance context of a higher environment - is a persistent risk in environment pipelines where data movement between environments is not governed with explicit quality controls. When lower environment data moves to higher environments without quality validation, the quality problems it carries contaminate the higher environment’s data landscape and undermine the reliability of every validation activity conducted there. A UAT environment populated with dirty data from SIT produces acceptance test results that reflect data quality failures rather than solution behavior, making it impossible to distinguish between a solution defect and a data defect.

Best Practice

Establish explicit controls governing the movement of data from lower environments to higher environments and require that any such movement be authorized, validated, and documented. Data that is being promoted alongside a solution from one environment to the next should be validated against the data quality standards of the target environment before the promotion is executed. Validation should confirm that the data meets the completeness, accuracy, formatting, and classification requirements of the higher environment. Data that fails validation should be remediated before promotion or excluded from the promotion with a documented justification. Automated data quality checks should be integrated into the promotion pipeline wherever feasible, ensuring that data quality validation is consistently applied rather than dependent on manual review that may be skipped under delivery pressure.

Benefit(s)

Governing the movement of data between environments prevents the data quality contamination that undermines environment fidelity and produces misleading validation results. Validation activities in UAT, PEN, and PSTG produce reliable signals about solution quality because the data in those environments is known to meet the quality standards of the governance context. Remediation of data quality failures occurs at the point of detection - in the lower environment where they are inexpensive to address - rather than after contamination has propagated to higher environments where remediation is more complex, more disruptive, and more expensive.