The Foundation of Data Quality
Data Quality starts long before the first business transaction. It can also fail long before you build your first Customer or Supplier Management Systems.
Data Quality starts with asking, and answering, the question, “What data does the enterprise need in order to operate effectively and efficiently and deliver value to our customer, suppliers, employees and partners?”
Logical Data Model (LDM)
The answer to this comes not from data but from Function. It is by knowing WHAT it is that the enterprise OUGHT to be doing (Business Functions) that you can identify what data it needs. It is the corporate Logical Data Model (LDM) that enables you to define the individual elements and structure of this data.
IMM Data Structure Modelling techniques enable you to extract all of the data elements from the Business Functions and build them into a fully normalised LDM.
Benefits of the LDM
In essence, without the LDM, you will not be able to achieve Data Quality!
The LDM defines:
- All Data Entities required by the business and their attributes. In essence, if it is not in the LDM it is not required by the business.
- The relationships between data entities. Data Attributes are important, however, the most important thing to know before you build any database are the relationships between the individual Data Entities. You can always add attributes (columns) to a table at a later date. However, structure (foreign keys) is much harder to add and, if they are not there from the beginning, you might never be able to recapture the relationship.
- Unique Identifiers (UIDs). These are the key to preventing data duplication in databases. UIDs answer the question, “What is it, in this enterprise, that makes one occurrence of this Data Entity unique and distinct from every other occurrence of the Entity?”