Lars Iffert, Analyst – Data & Analytics, BARC Timm Grosser, Senior Analyst – Data & Analytics, BARC
Companies depend on reliable data. Not only to control their operational processes and optimize products and services, but also to make the right decisions or to offer new services and products to their business partners.
The need “to be able to use data correctly” has recently been recognized. Data landscapes are developing rapidly, in some areas even chaotically. The days when companies only had to look after a manageable amount of uncomplicated data are long gone.
Instead, new data collections are constantly emerging in the most diverse environments with their own challenges and questions.
Departments across the business now face challenges with regard to poor data quality, poor data availability, lack of data access and insufficient knowledge of existing data. In addition, those involved are dependent on data from other areas.
In addition to political and organizational hurdles, technical challenges lie primarily in the lack of interfaces, the lack of uniform master data and the integration of necessary data.
The key to addressing these problems and being successful in BI and analytics today often lies in data management. But how can companies establish sustainable data management across all areas?
Never before have IT departments and business units had so many technical possibilities and solutions at their disposal, which they can use to address exciting and promising tasks. However, this development carries the risk that much potential is lost by creating isolated data silos. Only the ability to link the knowledge contained in these silos opens up new possibilities to position the company competitively and sustainably.
Unfortunately, a basic concept to control, develop, maintain and protect the production factor (i.e., data) in a comprehensive and sustainable manner is often lacking.
To ensure that data is of sufficient quality, two decisive factors must be taken into account aside from the technology used. Reliable master data is the linchpin, but the establishment of responsibility for data (e.g., via data owners) is essential too. Without these, data quality projects will usually fail.
Business success through the linking of knowledge
Master data comprises basic information about objects such as products, business partners and organizations, which is required for ongoing processing and which itself is rarely (or never) changed.
Together with metadata, master data is the central link that enables the meaningful linking of knowledge from day-to-day business, BI or data science labs. It forms the basis of uniform semantics through the clear definition of business objects in the company and thus contributes to a uniform understanding and the correct use of data.
Data responsibility is the key to escaping the data dilemma
Without clear responsibility for data, speed, quality and compliance (as required, for example, with regard to GDPR) cannot be ensured across the board. Also, some use cases can only be implemented through central data control, which creates framework conditions and can bring together partial knowledge from the environments. The delegation of responsible “data owners” for essential master data such as “product”, “customer” and “material” is often the first and most important starting point towards using data more economically and becoming more successful in general