Strategic planning for structuring master data

Strategy and planning are the cornerstones of every well-organized database. The planning process is known as data modeling and the subsequent product, which provides the basis for your master database, as a data model. Planning should be thorough in order to keep track of things at all times.

Your database and its datasets are based on real-world entities, for example your customers and suppliers or your products. The purpose of data modeling is to clearly specify these objects, as well as their properties and how they relate to one another.

Your database must remain in place even if there are changes in work processes, company functions or the software you use. Your data model should, therefore, be independent of these factors in order to ensure a long service life.

Aktenschrank

Creating a data model

In the vast majority of cases, data models are created in three consecutive steps: first a conceptual, then a logical, and finally a physical data model.

The conceptual data model

The conceptual data model is, as the name implies, the model that embraces your concept. It is independent of the software you use. It is also referred to as a semantic data model and represents a section of the “perceived world” in a concept, i.e. all relevant entities, their properties and how they relate to one another. Examples of these entities are people, companies, places or orders. An example of how entities might relate to one another is the assignment of orders to individuals, so that one person or company is always assigned to one order, but one person is assigned to several orders. These facts and relationships are represented as a model in a diagram.

The logical data model

The logical data model is based on the conceptual model and is also referred to as a database model, because it should be the foundation for the digital database. A tabular relation model is typically used for this. Using software, the schema of the conceptual model can be taken over automatically and then optimized. Redundancies must be avoided, as these can cause anomalies in the database later on. For example, a database that stores not only the order number but also the address of the customer for each new order will be storing duplicate address data if more than one order is placed. If the customer’s address then changes, all records have to be adjusted. To avoid this, it is necessary to split the data structure. The customer’s address is then saved only once and is conveyed via a reference. This process is called normalization.

The physical data model

As the logical data model is not tied to a particular database management program, the next step is to transfer the database model to the desired software. A data type is defined for each attribute, for example integers, floats, string or booleans (true / false values). The attributes and tables are given as abbreviations, usually completely in capital letters, so that the database software can process the data.

Why planning is so important

As you can see, when planning a database for your master data management, there are many steps, all of which must be taken carefully. If your company does not start with an in-depth analysis of factors, actors, and entities in the real world, mistakes can be made right through to the finished database, dramatically impairing your data structure. It is also important to take steps like normalization, or the prevention of redundancy, so that the resulting database is sustainable and can form the basis of your company’s digital transformation.

By using this site you accept the usage of cookies. More information

The cookie settings on this website are set to "Cookies enabled" to give you the best experience. If you use this website without changing the cookie settings or click "Accept", you agree to this.

Close