Issues Caused by Data Entry Processes
- Data Entry Interface Issues: Poorly designed data entry interfaces can contribute to data quality issues. If a data entry interface does not have edits or controls to prevent incorrect data from being put in the system data processors are likely to take shortcuts, such as skipping non-mandatory fields and failing to update defaulted fields.
- List Entry Placement: Even simple features of data entry interfaces, such as the order of values within a drop-down list, can contribute to data entry errors. Inter-Fields validation is recommended. For instance Delhi city must be validated with Country India. Geographic information can be enhanced through address standardization and Geo-coding, which includes regional coding, municipality, neighborhood mapping, latitude / longitude pairs, or other kinds of location-based data.
- Field Overloading: Some organizations re-use fields over time for different business purposes rather than making changes to the data model and user interface. This practice results in inconsistent and confusing population of the fields.
- Training Issues: Lack of process knowledge can lead to incorrect data entry, even if controls and edits are in place. If data processors are not aware of the impact of incorrect data or if they are incented for speed, rather than accuracy, they are likely to make choices based on drivers other than the quality of the data.
- Changes to Business Processes: Business processes change over time, and with these changes new business rules and data quality requirements are introduced. However, business rule changes are not always incorporated into systems in a timely manner or comprehensively. Data errors will result if an interface is not upgraded to accommodate new or changed requirements. In addition, data is likely to be impacted unless changes to business rules are propagated throughout the entire system.
- Inconsistent Business Process Execution: Data created through processes that are executed inconsistently is likely to be inconsistent. Inconsistent execution may be due to training or documentation issues as well as to changing requirements.
Issues Caused by Data Processing Functions
- Incorrect Assumptions about Data Sources: Production issues can occur due to errors or changes, inadequate or obsolete system documentation, or inadequate knowledge transfer (for example, when SMEs leave without documenting their knowledge). System consolidation activities, such as those associated with mergers and acquisitions, are often based on limited knowledge about the relationship between systems. When multiple source systems and data feeds need to be integrated there is always a risk that details will be missed, especially with varying levels of source knowledge available and tight timelines.
- Stale Business Rules: Over time, business rules change. They should be periodically reviewed and updated. If there is automated measurement of rules, the technical process for measuring rules should also be updated. If it is not updated, issues may not be identified or false positives will be produced (or both).
- Changed Data Structures: Source systems may change structures without informing downstream consumers (both human and system) or without providing sufficient time to account for the changes. This can result in invalid values or other conditions that prevent data movement and loading, or in more subtle changes that may not be detected immediately.
