Nils Klute from the eco association conducted the following interview with Jörg Bienert.
Companies often lack an overview of their data landscape. The main remedy is a clear data strategy, says Jörg Bienert, chairman of the KI Bundesverband e.V. and partner at the data science consultancy Alexander Thamm. We asked him how to create added value with data, and what role data plays in the Service-Meister project..
Mr. Bienert, why do entrepreneurs need a data strategy?
Data is a central asset of companies. Along the entire value chain, data is generated that can represent enormous value. However, this is not being sufficiently used at the moment. Data is either not being evaluated or is processed only in silos, which means that cross-company context is missing. In order to use data efficiently and holistically within the company, it is necessary to create a comprehensive, company-wide data strategy. Until now, the issue has been approached rather shabbily and often only in isolated solutions.
Why is that?
Individual company divisions work with different systems, store data in uncoordinated formats and do not have comprehensive approaches. Companies lack an overview of their own information landscape. Too often, the possession of data and information is still equated with knowledge and power, which means that the willingness to share data openly is low. We need a data-driven, entrepreneurial mindset at all levels.
And how can we tap into this way of thinking?
Part of a data strategy can, for example, be that data should always be freely accessible within the company. Exceptions must be appropriately justified by the data owners. At the same time, if the production of data in one department is complex and involves high costs that have to be borne by other departments using the data, billing systems should be established.
How should business operators proceed?
Companies need transparency first. What data is actually collected? Where is it stored? And what quality is it? Documenting data and databases in a catalog helps make them accessible. Then we come to questions of data governance, which means defining processes; for example, how data can be used and protected at the same time.
Data catalogs and data governance – how should users best implement this?
Step by step! In recent years, attempts have been made to collect data in elaborate data lakes projects. These projects became very complex, lengthy, and expensive, but in most cases, they didn’t provide any direct benefit as the completion of the lake was prioritized ahead of the applications based on the projects. It is therefore advisable to have an emergent data strategy that defines the framework, rules and structures. Catalogs, storage systems and technologies for the respective use cases can then be set up to match.
What do the use cases actually look like?
The number of use cases is unlimited and covers all elements of the value chain and all industries. Existing evaluations should be standardized as far as possible as part of the data strategy and supplemented by additional cross-company, horizontal analyses. Additionally, whole new possibilities for analyzing and processing data are now available based on artificial intelligence
What are the differences between digitalization and data strategy?
One thing arises from the other; first digitalization emerges from the corporate strategy and then the data strategy follows. Constructed like this, everything intentionally interlocks. While the digitalization strategy looks at all aspects of the use of information technology for process optimization and the design of new business models, the data strategy focuses on how existing or additionally collectible data can contribute to this.
How do you turn data into products?
You need experts for that. Using sensor values, measurement results and key figures, data scientists obtain information that is relevant to business. They also include a wide range of skills in their work. So it’s not only about computer science, software tools, and programming languages, but also about mathematics, statistics, business administration, and finally machine learning. All of this is necessary to extract information from data lakes with intelligent algorithms that allow digital products and business models to flourish..
And what role does that play in the Service-Meister project?
A very big one! After all, data from industrial services should be able to be used in intelligent applications for technicians. In order to make AI applications feasible, a data strategy is required that is oriented towards the goals of the company and use cases. This is the only way to turn product, wear, machine, and customer data into digital tools for technical service in the age of industry 4.0.
Wir danken für das Gespräch!
***As an associated partner, the KI Bundesverband e.V. supports the Service-Meister AI project. The association is made up of more than 220 innovative companies, SMEs, start-ups, and experts whose central business purpose is the development and application of technologies based on artificial intelligence.
Keep up to date with our Service-Meister newsletter on all topics concerning the project