In the past, Data Integrity was defined and enforced with data edits but this method did not cope up with the growth of technology and data value quality greatly suffered at the cost of business operations. Organizations were starting to realize that the rules were no longer appropriate for the business.
The concept of business rules is already widely used nowadays and is subdivided into six categories which include data rules. Data is further subdivided Data Integrity Rules, data sourcing rules, data extraction rules, data transformation rules and data deployment rules.
Data Integrity is very important in database operations in particular and Data Warehousing and Business Intelligence in general. Because Data Integrity ensured that data is of high quality, correct, consistent and accessible, in is important to follow rules governing Data Integrity.
A Data Value Rule or Conditional Data Value Rule specifies data domains. The difference between the two is that the former specifies the domain of allowable values for a data attribute which applies to all situation while the latter does not apply to all situations but only when there exceptions or certain conditions that applies.
A Data Structure Rule defines that cardinality of data for a data relation in cases where there are no conditions of exceptions which apply. This rule makes data structure very easy to understand. A conditional data structure rule is slightly different in that is governs when conditions or exceptions apply on data cardinality for a data relation.
A Data Derivation Rule specifies the how a data value is derived based on algorithm, contributors and conditions. It also specifies the conditions on how the data value could be re-derived.
A Data Retention Rule specifies the length of time of data values which can be retained in a particular database. It is specifies what can be done with data values when its use for a database expires A data occurrence retention rule specifies the length of time the data occurrence is retained and what can be done with data when it is no longer useful. A data attribute retention rule is similar to a data retention rule but the data attribute retention rule only applies to specific data values rather than the entire data occurrence.
These Data Integrity Rules, like any other rules, are totally without meaning when they are not implemented and enforced.
In order to achieve Data Integrity, these rules should be consistently and routinely applied to all data which are entering the Data Warehouse or any Data Resource for that matter. There should be no waivers or exceptions for the enforcement of these rules because any slight relaxation of enforcement could mean a tremendous error result.
As much as possible, these Data Integrity Rules must be implemented in as close to the initial capture of data so that early detection and correction of potential breach of integrity can be taken action. This can greatly prevent errors and inconsistencies from entering the database.
With strict implementation and enforcement of these Data Integrity Rules, data error rates could be much lower so less time is spent on trying to troubleshoot and trace faulty computing results. This translates to savings from manpower expense.
Since there is low error rate, there can only be high quality data that can be had to provide better support in the statistical analysis, trend and pattern spotting, and decision making tasks of a company. In today’s digital age, information one major key to success and having the right information means having better edge over the competitors.