Source Data Manipulation

All organizational performance measurement systems rely foundationally on their source data. And like a house built on sand, these performance measurement systems fail if their source data becomes corrupt.[wcm_restrict plans=”41526, 25542, 25653″]

Source data manipulations occur for many reasons; some innocent and others malintent. Regardless of the reason, altering metric data necessarily generates a measurement output different than that expected; driving decision-making in a potentially dangerous direction. Therefore, unintended or purposeful performance measure data manipulation should be avoided.

Performance measure data manipulation occurs because of either a lack of preventive measures and/or errant behaviors. While not all inclusive, the four lists below, Process-Based Warning Flags, Process Execution Warning Flags – Behaviors, Potential, Observable Results, and Potential Causes, are designed to help organization leaders recognize whether performance measure source data is or is at risk of being altered. Only after a problem is recognized and its causes identified can the needed action be taken to move the organization toward improved performance.

Process-Based Warning Flags

• Non-existent or loosely defined data standards
• Lack of data change controls
• Broadly assigned data system access authority beyond read-only access
• Absence of system reinforced data entry quality standards
• Lack of data transparency

Process Execution Warning Flags – Behaviors

• Routine correction of system data rather than input improvement and standards reinforcement
• Executive and manager withholding of typically unrestricted data

Potential, Observable Results

• Step changes in metric output
• Incongruousness between related metrics

– in time series

– different inter-dependent measures

• Unexplained delays in releasing metrics and reports
• Protectiveness or secrecy of a metric, especially those that were previously more widely distributed

Potential Causes

• Lack of system use training, especially process-based training highlighting the interrelatedness of processes, importance of data accuracy, and impacts of data changes
• Misdirected effort to correct errant data
• Accidental alteration
• Malintent

[/wcm_restrict][wcm_nonmember plans=”41526, 25542, 25653″]

Hi there! Gain access to this article with a StrategyDriven Insights Library – Total Access subscription or buy access to the article itself.

[/wcm_nonmember]