Result

Simpler and more systematic business-model development

The “Data-Driven Business Models” workstream focused on defining and describing the capabilities a company needs to have in general, and enterprise architecture needs to have in particular, in order to successfully develop and implement data-driven business models (DDBM). Among other things, the workstream produced a guide that will help CBA Lab member companies develop their own data-driven business models and/or enable them to work with other companies, or in other ecosystems, that utilize data-driven business models. 

The workstream addressed four questions in this regard:

• What is a DDBM?
• How can companies develop a DDBM?
• Which capabilities are needed to ensure a successful DDBM?
• What needs to be done in order implement and operate a DDBM?

In order to make the most precise determination possible as to which capabilities are needed, the workstream participants first identified the trends that make data-driven business models so relevant these days. They then described what such a business model actually is. The most important trends identified in the first step involve the increasing possibilities for collecting data (e. g. via sensors and social media), for processing data (e.g. via analytics technologies), and for exchanging data via platforms such as Bosch IoT Suite. 

The workstream described a data-driven business model as a “business model that relies on data as a key resource and/or the core of a company’s value proposition, whereby data collection, processing, dissemination, or vending are the company’s main activities.”  

Examples of such data-driven business models include those used by Check24, Uber, and Travelbasys.

With regard to capabilities, the workstream focused on key data-specific development and management approaches: 

• Data strategy development 
  Which types of data are important? A DDBM needs to have a strategy for using data both in an internal context and within the framework of its relationships with customers. 
• Data model management
  Before a DDBM can be developed, a decision needs to be made regarding how data models will be managed across all key business objects, utilization scenarios, contexts, and classifications. This also means that a determination needs to be made as to whether special concepts for data models (e.g. digital twins) should be established.
• Data provision management 
  A decision must be made regarding which types of data are to be made available for both data exploration and subsequent prototypes or finished systems, as well as the form in which the data is to be provided. Rules also need to be formulated that differentiate between those types of data that will be made available internally and those that will made available externally. 
• Data governance management
  There also needs to be another set of rules that name specific individuals who will be allowed to use data, whereby these rules must also define what these individuals can actually do with the data. Decisions also need to be made regarding the design of data governance processes and the structures that will be implemented for the governance bodies that will decide how data is to be used.
• Data (IT/reference) architecture management
  Those who develop a DDBM must decide which technical platforms, orchestration possibilities, etc. will be needed to implement the business model.

A distinction was made by the workstream with regard to capabilities relating to existing conditions at a company, the identification processes used, and the methods for implementing and operating a DDBM. The closer a company gets to the operational stage, the more operation-focused its capabilities need to be. For example, in the case of identification and implementation, a company needs to be able to effectively conduct data exploration processes and safeguard data, while in the operational stage, capabilities must be geared more towards testing and monitoring.

The association formerly known as Industrial Data Space (now known as the International Data Spaces Association) plays a role here, whereby CBA Lab is already very familiar with this organization.

International Data Spaces is dedicated to ensuring the secure exchange of data between companies in a system in which the data provider always remains the owner of the data and thus maintains control over how it’s used. The organization has proposed an architectural approach for depicting key DDBM capabilities such as data exchange standards, (meta) data management, data modeling, monitoring and collection/storage/provision of data, data quality management, compliance and governance.

“As architects, we don’t claim to be able to develop data-driven business models,” says Workstream Coordinator Miriam Suchet. “Nevertheless, the close connections between technology, business operations, and data mean that specialist and IT departments need to be just as closely connected. That’s the only way we can make use of a rapid and iterative approach that will enable us to determine which data, processes, legally compliant structures, and technical platforms will be needed for a particular business model. This in turn makes it possible to implement the business model very quickly after such a determination has been made.

In the future, those companies will enjoy a competitive advantage that are able to most effectively manage the development of forward-looking business models and, similar to the current situation with software development, are able to do this in such a systematized manner that subsequent implementation of the model will be extremely rapid and display very low friction losses.” 

The workstream participants also took these aspects into consideration and then analyzed methods such as design thinking, Business Model Canvas, and Operating Model Canvas to determine their suitability for use in business model development.

“Whereas Business Model Canvas illuminates all aspects of a business model, Operating Model Canvas specifies the required resources,

activities, and partnerships in more detail,” Suchet explains.

In general, the workstream searched for ways to systematize and simplify the development of data-driven business models – and make the associated development processes repeatable.