Companies often know a great deal about individual assets, processes, and activities, yet surprisingly little about how they relate to one another.
A production system knows the operating status of a machine. The maintenance department documents repairs. The ERP system stores spare parts and costs. Another application manages the employees responsible for the equipment. All the information is available. But anyone trying to understand why a machine repeatedly fails and how those failures affect ongoing orders must bring together data from several applications.
The problem lies in the structure of the IT landscape. Each system manages its own fragment of reality.
Cluu, the digitalization platform developed by Softwarehelden, takes a different approach. Its technological foundation is an ontology-based virtual graph database that connects existing information and places it within a shared business context.
What may sound abstract at first fundamentally changes how companies can use their data.
An Ontology Describes How Things Are Connected
The term ontology originally comes from philosophy. In computer science, it refers to a structured model of concepts, properties, and relationships within a particular domain of knowledge.
Consider an industrial company. It has machines, components, employees, projects, test orders, and suppliers. A conventional database stores information about these objects in tables.
An ontology goes a step further. It describes what these objects mean and how they relate to one another.
A machine consists of components. A component is supplied by a supplier. A machine is used in a production process. An employee is responsible for a maintenance task.
These relationships form a business-oriented model of the company's operations. Information from different systems can be associated with the same object.
The advantage is that applications no longer have to reconstruct these relationships from scratch whenever a new question arises. They can draw on a shared understanding of the data.
From Data Collections to a Knowledge Graph
While an ontology defines the conceptual model, a knowledge graph represents the actual objects and their relationships.
Think of it as a network. Nodes represent machines, projects, people, or other entities. Connections describe how they relate to one another.
What makes this approach particularly useful is the ability to explore information by following these relationships.
Consider an example from automotive development. A test vehicle runs a particular software version. That version has been used in several test orders. One of those test orders is linked to measurement files and a documented defect.
In isolated systems, this information is scattered across separate records. In a knowledge graph, it becomes part of a connected information network.
This makes it possible to answer questions that go beyond simple database queries: Which other tests were performed using the same software version? Which defects occurred in comparable vehicle configurations? Which measurement results belong to those tests?
The real potential lies in making relationships across systems and departments visible and usable.
Cluu Connects Data Without Replacing Existing IT Systems
A knowledge graph alone, however, does not solve the challenges of complex IT environments. Data still resides in different databases, applications, and file systems.
Many integration architectures therefore rely on collecting data in a central repository. This can be a sensible approach, but it also creates additional work involving data transfers, synchronization, and updates.
Cluu uses a virtual graph database. It allows existing data sources to be accessed through a shared business model without requiring all information to be migrated into a new central database first.
Existing ERP, project management, asset management, and testing systems can continue performing their respective tasks. Cluu establishes connections between the relevant information and makes it available to applications that work across these systems.
This is particularly valuable for companies whose IT landscapes have evolved over decades.
A proven system does not need to be replaced simply because new digital processes require information from multiple sources. Instead, new applications can build on data that is already available.
Of course, interfaces still need to be configured, data models aligned, and business relationships defined. A virtual architecture does not eliminate this work. But it enables integration without fundamentally rebuilding the entire IT landscape.
What Companies Can Actually Do With It
The value of an ontology becomes apparent in everyday operations.
In project management, tasks, resources, contracts, and responsibilities can be linked. A project manager can identify which work packages may be affected by a delayed delivery.
In asset management, equipment can be connected to its maintenance history, spare parts, technical documentation, and responsible personnel. Employees no longer have to search multiple applications to find relevant information.
In automotive testing, Cluu Assisted Testing connects test orders, test objects, specifications, measurement data, and defects. Engineers can examine test results in the context of the corresponding vehicle configuration and test conditions.
A shared data foundation also makes it possible to support different business processes on the same platform.
Softwarehelden provides more than 50 applications based on Cluu. These applications can share data models, functionality, and access permissions. New use cases therefore do not necessarily require another isolated data repository.
Why Knowledge Graphs Matter for Artificial Intelligence
Large language models can analyze documents, answer questions, and summarize information. But they do not automatically understand the specific relationships within a company.
An AI system might be able to read a maintenance instruction. Whether that instruction applies to a particular installed machine, its year of manufacture, and its current technical condition depends on additional information.
A knowledge graph provides structured context for these decisions.
It can establish connections between a machine, its configuration, relevant documentation, and previous maintenance activities. An AI application can use these relationships to retrieve more relevant information and support its answers with traceable sources.
Cluu goes a step further by connecting data with business processes and applications. Depending on its configuration, the Cluu Assistant can retrieve information, support tasks, and initiate actions within connected applications.
A knowledge graph does not automatically make AI error-free, however. The reliability of its answers still depends on the quality and completeness of the underlying data, how up to date it is, and how the AI application is implemented.
A Shared Model Needs Clear Rules
As information becomes more interconnected, access control becomes increasingly important.
The fact that information is technically connected does not mean every employee should be allowed to see all of it.
Cluu therefore includes a rule-based authorization system. Access to data and applications can be managed through defined permissions.
The quality of the business model matters just as much. If different systems use different names for the same object, these terms must be mapped correctly. Otherwise, even a knowledge graph can produce misleading connections.
An ontology must therefore reflect actual business processes and evolve in collaboration with the people who understand them.
The Foundation for a Connected Enterprise IT Landscape
The value of an ontology does not come from connecting as much data as possible. It comes from using those connections to answer meaningful questions and support real business processes.
Cluu turns this principle into a platform architecture: Existing systems provide information, the virtual graph database establishes relationships between it, and applications built on top make that information useful in everyday operations.
Companies can start with a manageable use case and gradually expand their connected information landscape.
This also changes how organizations approach their IT architecture. Rather than solving every new requirement by introducing another isolated application, they can increasingly build on existing data, relationships, and functionality.
Making this possible requires a shared understanding of what the data means and how it is connected.
That is precisely what the ontology behind Cluu provides.
