Optimizing product management with data What do companies need to do to ensure that their products continue to be successful in the future? Product management deals with this question. These experts are responsible for planning, managing and controlling their company's products and services. They often rely on their gut feeling, because relevant data is usually scattered across different departments. The 'Data-based Product Management' project is working out how product management can be improved on the basis of data. In the project, the companies Diebold Nixdorf, DMG Mori, Isringhausen and Schmitz Cargobull are to be guided to carry out classic product management tasks, such as planning new product features, more objectively and successfully by using modern data analysis. The idea is to use data from a variety of sources, such as operational data, internal data from marketing and sales, and external data from social media, for example. The companies are supported by the research partners Fraunhofer IEM and Heinz Nixdorf Institute.

Product management (PM) is currently facing a major challenge: In the course of the dual transformation, which describes the joint exploitation of the potential of the megatrends of digitization and sustainability, both existing and future market services have to meet entirely new, increasingly stringent requirements. As a result,

market services must be drastically changed within a short period of time in order to minimize the negative

impact on the climate, for example through new functions to reduce energy consumption. Due to limited financial and time resources, it is essential for

product management in this context to choose the right points of attack and, above all, to tap the potentials that promise a high leverage effect.



In order to identify these potentials, extensive information about the entire life cycle of the product is required. One promising way of obtaining this information is

in particular the analysis of product-related data in product management, which is expected to become much more important in the next

years. For example, operating data of the products in the field, market data or social media data can be considered. With the help of data analyses, weak points of products as well as behavioral patterns of users can thus be identified, for example. These can subsequently form the basis for the data-based optimization of future product generations. 



Decisions in product management are often made on the basis of empirical knowledge and the gut feeling of individual players. Extensive product-relevant data is usually not used for decision-making because it is often scattered across numerous departments, e.g., research and development, production, or sales. In addition, product management usually lacks an overview of what data is fundamentally available in the company. Only around ten percent of companies consider themselves to be very well positioned in terms of their data and systems, and around one third describe their own company as little to not at all prepared for the coming increase in complexity [BMF19]. Product management is thus currently failing to systematically tap the potential of digitization and base decision-making on objective data.

Test





The aim of the planned project is to develop a system for data-based product management in manufacturing companies. The system should enable companies to carry out classic product management tasks (e.g., planning new product features) more objectively and successfully by using modern data analysis approaches. The idea is to aggregate data from various sources and evaluate it together, e.g., operational data from networked systems, internal data from marketing/sales, or external data from social media. The target picture is shown in the image (right). To achieve this goal, the following subgoals are pursued:



1) Reference model for data-based product management: The basis for data-based product management should be a reference model that is both scientifically sound and application-oriented. The model has a prescriptive character and thus specifies how data-based product management should basically proceed and be organized. The model is to be derived both with the help of the literature and with empirical investigations in manufacturing companies. Components of the reference model are, for example, processes, methods, (technical) building blocks and architectures that are needed for data-based product management.



2) Prototypical product.Intelligence Tools: Based on the reference model, six prototypical tools will be created that implement the conceived functionalities in the respective pilot companies. Each tool will be individually adapted to the requirements and use cases of the companies and implemented in the existing enterprise architecture. A systematic comparison of the tools and their use will also provide important insights that will be used to further develop the reference model. In addition, they are made available generalized to other companies in terms of result transfer. 

.