The modern architecture for data- and AI-oriented companies.

Introduction

The increasing importance of data and artificial intelligence (AI) is challenging companies to realign their data management not only technically, but also organisationally and strategically. The Data Management Model 5.0 offers a contemporary, holistic framework that maps the integration of data, governance, AI and value creation in a clearly structured overall picture.

1. From ‘why’ to ‘how’: linking corporate strategy and data & AI strategy

At the heart of the model is corporate strategy (WHY). It defines the overarching goals and business benefits. Derived from this is the data & AI strategy, which specifies how data and AI contribute to the achievement of these corporate goals.

The data & AI strategy translates into four central control dimensions:

• Data scope – What scope of data is relevant?
• Principles & ethics – What values and guidelines govern our actions?
• Organisation & processes – What structures and processes ensure sustainable data management?
• Literacy – How are data and AI skills developed and promoted within the company?

Accountability and stewardship are firmly established as overarching principles and ensure responsibility and a sustainable data culture.

2. Data governance as a control instrument (HOW)

Data governance is the link between strategy and operational implementation. It defines the ‘HOW’ – in other words, how data management is implemented effectively, ethically and in a value-adding manner. Governance includes:

• Clear roles and responsibilities (accountability partners & stewardship)
• Binding rules and standards
• Promotion of data and AI literacy

The data governance building blocks in the model – classification of white and cross-cutting elements

In Data Management Model 5.0, various central functions are understood as building blocks of data governance:

• The white building blocks in the model – including data classification, meta data management, data quality management and master & reference data management – are examples of central governance functions that are applied specifically throughout the entire data life cycle. They ensure transparency, quality, traceability, documentation and compliance in all phases.
• These are supplemented by cross-functional governance building blocks such as data catalog and data sharing. They are colour-coded in the model but play an equally important role in data governance. The data catalogue is a central repository that ensures transparency, orientation and findability in the data pool, while data sharing enables the targeted, compliant and value-adding use of data across departments and companies.

Together, these functions form the backbone of sustainable, company-wide data governance. Their control and implementation at the relevant points in the life cycle enables effective, secure and innovation-promoting data management.

3. Data Life Cycle Management (WHAT): The operational core

The operational core of the model is Data Life Cycle Management, which is divided into three central phases:

a) Plan & Design

• Architecture: Basic structure for the data landscape.
• Data Modelling & Design: Design of data models and structures.
• Data Classification: Categorisation, evaluation and assignment of sensitivity levels for data (e.g. public, internal, confidential, strictly confidential).

b) Maintain & Enhance

• Meta Data Management: Management of metadata to ensure transparency – e.g. for data catalogues, context information for AI assistants and multi-agents, as well as for systematic documentation, verification and continuous improvement.
• Data Quality Management: Ensuring, monitoring and continuously improving data quality – including documentation of quality standards, verification processes and verification. These measures are particularly effective when they are implemented directly in the operational systems, where they ensure reliable and consistent data.
• Master & Reference Data Management: Management of consistent master and reference data.
• Data Storage & Operations / Data Lake House / Data Fabric: Modern data management at the platform level.
• Data Integration & Interoperability: Ensuring seamless exchange and networking of data sources.
• AI Assistants / Multi AI Agents: Automated support for data management and analysis processes – including automation of routine tasks, process optimisation and decision support through AI-based analyses and autonomous real-time decisions tailored to requirements.

c) Enable & Use

• Business Intelligence / Reporting & Dashboarding: Data-based reporting and decision support.
• Document & Content Management: Management and use of unstructured content – including systematic documentation and compliance with documentation and retention requirements.
• Data Science / genAI: Use of modern analysis and AI technologies.
• Data Products & Add-Ons: Development and operation of data products and extensions.
• Data monetisation: Value creation through the targeted use and marketing of data and data- and AI-based products and solutions, such as prediction, recommendation and support systems.
• Data retention, archiving & decommissioning: Compliant data storage, archiving and deletion.

4. Integration of operational systems and partners

The model explicitly distinguishes between two integration perspectives:

1. Integration of operational systems within the company (e.g. CRM, supply chain, finance, materials, manufacturing, e-commerce). These reflect the real complexity and heterogeneity of modern enterprise IT and are fundamental to internal business processes and value chains.
2. Integration of external partner systems, i.e. data exchange and collaboration with partners outside the company. The focus here is on cross-functional linking, data interoperability and collaboration with suppliers, customers or partners.

Data does not flow in isolation, but is embedded in both internal processes and external value creation networks and partnerships.

5. Data catalogue – the central repository for company data

The data catalogue is the heart of transparency, findability and governance in modern data management. A data catalogue provides a central overview of all structured and unstructured data assets within a company. It enables:

• Consistent and targeted search for data resources
• Visibility and context regarding data origin (provenance), quality and access rights
• Linking to metadata management and governance requirements
• Support for regulatory compliance and documentation requirements
• A basis for efficient data sharing and data-driven collaboration

The Data Catalog thus forms the foundation for a data-driven and collaborative organisation in which both business and IT users can quickly identify, evaluate and use the data that is relevant to them.

6. Data sharing as a connecting element

Data sharing is the central connecting element of the Data Management Model 5.0. It describes the ability and willingness to share data both internally (between departments and systems) and externally (with partners, customers, suppliers) in a secure, transparent and efficient manner.

Key aspects of data sharing:

• Openness and interoperability: Data is provided in such a way that it can be used across organisations.
• Rule-based access: Access is granted according to governance guidelines, roles and access rights.
• Transparency and traceability: Data catalogues and metadata management make it possible to trace which data is available and how it may be used at any time.
• Trust-based collaboration: Data sharing enables new forms of collaboration and value creation without incurring risks such as data breaches or loss of control.
• Enabling self-service and AI: Data sharing is what makes self-service analytics, AI-based processes and innovations possible across the entire company.

Data sharing bridges the gap between operational systems, specialist departments and external partners and makes a decisive contribution to leveraging data potential comprehensively and responsibly.

7. Data & AI risk management as a central layer of protection

All data management activities are protected by a comprehensive data & AI risk management system. The focus here is on:

• Security
• Privacy
• Compliance

This is essential for managing risks and complying with regulatory requirements, such as those set out in the EU AI Act and GDPR.

8. Key success factors for modern data management

• Holistic approach: Linking strategy, governance, lifecycle and operational implementation.
• Interdisciplinarity: Collaboration between business, IT, processes and data management.
• Clearly defined responsibilities and roles.
• Focus on data quality and value creation.
• Integration of AI as a lever for efficiency and innovation.
• Literacy & culture: Continuous development of data and AI expertise within the company and a data-inspired leadership culture.
• Ensuring data protection, security and compliance.

Conclusion

The Data Management Model 5.0 stands for a modern, strategically integrated and AI-enhanced understanding of data management. It makes it clear that only those who consistently combine data strategy, governance, lifecycle and value creation and embed them in their organisation can use data and AI as a real competitive advantage and enable sustainable innovation.

Further interesting articles:

• Advisory | Impulse Talks | Trainings
• How to build a data mesh ready platform
• Maximizing ROI in Data Driven with ARDPS
• Processes and data must go hand in hand
• Be a Data Punk
• The Hub & Spoke Model: A Roadmap to a Data-Inspired Organization