Thanks to constantly growing data volumes and the increasing complexity of data streams and business processes, the definition of data models is becoming ever more important. The data model lays the foundation for the quality in which the data can later be used and provided in the respective use cases. In the context of product data, for example, the data model in product information management (PIM) determines what information and data is required to populate the product detail pages in the online shop and how it must be structured in order to fulfil the channel-specific requirements.
The diversity of data models and their consequences
However, many more areas are relevant in a company than just the PIM. Customer data, digital content such as images, videos and graphics, supplier and employee data as well as location-specific information or even competitor data are managed, maintained and utilised by the various business units. It is not uncommon for the same data to be stored by the individual business units in their own systems or Excel spreadsheets – this is particularly true in larger companies or with more complex organisational structures.
As a result, there can be a large number of different data models in a company. These data models are usually customised to the use cases that are relevant for the respective specialist departments. For example, the sales team in region A maintains the same product data as product management in region B – but as the users in the two regions typically have different requirements for the product data, they also use different attributes and value ranges to describe the products according to their use cases.
As a result, most companies have developed a system landscape that lacks a common language. While at first glance this does not seem to play a major role for the individual business units and their daily work, problems inevitably arise as soon as they try to communicate with each other or even develop common workflows.
The diversity of data models in the company therefore makes it very difficult for systems and data processes to interoperate. This in turn can prevent synergies as well as knowledge and efficiency gains in company-wide information management. In such situations, there is also usually a lack of basic transparency about what information is available where in the company and in what quality, which becomes problematic at the latest when more extensive transformation projects are planned that are intended to bring companies one step closer to the target image of a data-driven company.
Data Class Foundation – company-wide reference model
It should be emphasised at this point that the solution to this challenge cannot be to harmonise the existing data models. The specific design of the respective data structures ultimately ensures that the requirements of the specialist departments are covered in the best possible way. However, what is needed to optimally support cross-company data processes is clarity about which data exists where and how exactly it is maintained. If department A describes the colour of a T-shirt as ‘indigo blue’ and department B simply calls the colour ‘blue’, then this information must be recorded centrally and it must be made clear that this is one and the same colour.
This is the task of a Data Class Foundation. As a reference model, it collects all data models used in the company and records both their commonalities and their special features, thus serving as a central information base for all departments and for all future digitalisation projects.
The role of the Data Class Foundation in digitisation projects
The framework conditions under which companies have to operate today are becoming increasingly complex and at the same time more dynamic. The expectations of customers, suppliers, business partners and employees are constantly changing and, at the same time, legislators are demanding ever more detailed information about products, services and data storage, particularly with regard to security and sustainability-related aspects. A high degree of flexibility and agility is needed to respond to these requirements – prerequisites that only a data-driven company can fulfil.
In such a company, not only do optimal data structures and processes prevail in the individual specialist departments themselves, so that they are supported in their work in the best possible way – these companies are also characterised by a highly efficient digital value chain throughout the company. The reason for this is the perfect interoperability of the individual system and business processes, which leaves the individual requirements of the departments untouched.
Modern digitalisation projects have firmly anchored this target image and yet its implementation is usually very difficult in practice. Very few companies are really clear about which data models actually exist across the organisation, which framework conditions have prevailed for their design and which data overlaps at which point in the various specialist areas and use cases.
This often leads to new data models being created in the course of a digitalisation project that are not in line with the overall perspective and thus leave synergy potential unused and, in the worst case, lead to inefficiencies and even greater complexity that make costly workarounds necessary.
As a reference model, the Data Class Foundation acts as a mediator and leads to an extraordinarily high level of transparency and interoperability, from which the company benefits in the long term and which forms the basis for the development of a data-driven company – with every new transformation project.
If you would like to find out more, please contact our experts without obligation!
