What is information architecture?
Information architecture is the part of enterprise architecture that defines which information an organisation needs, what that information means, how it is structured, where it comes from, who is responsible for it and how it flows between processes, systems and partners. It bridges the business architecture (processes and goals) and the application architecture (software and systems). Applications come and go, but information often lives on for decades, which is why a sound information architecture is one of the most stable foundations an organisation can have.
Information architecture ensures that the right information is available at the right time, in the right form and to the right person.
🧠 What does information architecture cover?
An information architecture is not a single document but a coherent set of agreements and models:
| Component | Description | Example |
|---|---|---|
| Concept framework | Unambiguous definitions of key terms | What exactly is a “customer”, “notification” or “fault”? |
| Conceptual information model | Which objects exist and how they relate, independent of technology | An installation has measuring points; a measuring point produces readings |
| Logical data model | Attributes, data types, keys and relationships | Entities and relationships, XML or JSON schemas |
| Data flows | How and where information moves between processes, applications and chain partners | From sensor via a historian to a report |
| Metadata and classification | Origin, quality, owner and sensitivity of data | Classification on availability, integrity and confidentiality |
| Standards and frameworks | Shared models for exchange | RGBZ, StUF, STOP/TPOD, Aquo, Linked Data, ISA-95 |
The distinction between the conceptual, logical and physical levels matters. The conceptual model describes reality in the language of the business, the logical model translates that into data structures, and the physical model captures the actual implementation in a database or message. In the Netherlands, the Metamodel for Information Modelling (MIM) maintained by Geonovum is the usual way to describe these levels consistently; the Netherlands Standardisation Forum recommends MIM as an open standard and added version 1.2 to its list in 2024.
🎯 Why is information architecture important?
| Benefit | Explanation |
|---|---|
| Coherence and reuse | Data is captured once at the source and used many times |
| Fewer errors | Clear definitions prevent two systems from interpreting the same word differently |
| Better service delivery | Faster and more reliable information for staff and customers |
| Interoperability | Essential for chain collaboration and integration between organisations |
| Insight and accountability | Supports audits, compliance and reporting |
| Agility | Replacing an application is easier when the data is described independently of it |
That last point is the core of Common Ground, the movement launched by the VNG (Association of Netherlands Municipalities) with Dutch municipalities: data stays at the source and is exposed through standardised APIs instead of being copied into every application package. That only works when the information model is clearly defined.
🔄 How does information architecture relate to other architecture layers?
| Architecture layer | Link with information architecture |
|---|---|
| Business Architecture | Supplies the information requirements derived from processes and objectives |
| Application Architecture | Uses the data models and exposes data through interfaces |
| Technical Architecture | Provides storage, integration platforms, databases and security |
| OT Architecture | Supplies process and asset data from controllers and sensors |
In TOGAF, this domain is formally called data architecture: together with application architecture it forms Phase C of the Architecture Development Method, Information Systems Architectures. The current version is the TOGAF Standard, 10th Edition, released by The Open Group in April 2022. In ArchiMate (version 3.2 since October 2022), information is modelled as a business object in the business layer, a data object in the application layer and an artifact in the technology layer.
🏛 How does information architecture work in Dutch government?
The Dutch government uses a system of reference architectures in which the information layer has a fixed place:
| Architecture | Audience | Information component |
|---|---|---|
| NORA | All of government | Concepts, information models, data principles and the system of ten base registers (including the BRP population register, the BAG addresses and buildings register and the Trade Register) |
| GEMMA | Municipalities | RGBZ (case data), the case type catalogue (ZTC), Common Ground and information models for municipal domains |
| WILMA | Water authorities | Aquo for water data, DAMO for objects such as watercourses and flood defences |
| PETRA | Provinces | Information models such as IMRO (older spatial plans), STOP/TPOD (Environment and Planning Act) and IMEV (external safety) |
| MARIJ | Central government | Government-wide agreements on data and exchange standards (since succeeded) |
Here, information architecture is the foundation for standards, data quality and data exchange. Many of these models appear on the lists of the Netherlands Standardisation Forum; standards on the “comply or explain” list must be applied by public bodies in procurement, or any deviation must be justified.
🏭 What does information architecture mean in an OT environment?
Information architecture is becoming increasingly important in operational technology (OT) too. A factory or treatment plant produces vast amounts of data, but without a model a tag such as PT-101.PV is only meaningful to the engineer who created it.
| OT application | Information aspect | Useful standard |
|---|---|---|
| Sensors and measurement data | Structuring telemetry: what, where, when, in which unit? | Naming conventions, Unified Namespace |
| SCADA and PLC data | Mapping signals onto an information model | OPC UA companion specifications |
| IT/OT integration | Exchanging orders, equipment and production results between ERP and MES | ISA-95 (IEC 62264), B2MML |
| Asset information management | Data on machines, maintenance, location and status | Asset Administration Shell (IEC 63278-1), ISO 55000 |
| Process industry | Making field data securely available for optimisation | NAMUR Open Architecture |
| Configuration management | Which systems, firmware and connections exist? | CMDB and asset inventory |
The hierarchical equipment model of ISA-95 (enterprise, site, area, work centre such as a production line or process cell, and work unit such as a unit or work cell) is often the backbone in practice: it serves as the topic structure in a Unified Namespace, as the asset hierarchy in a historian and as the structure of a digital twin. That way IT and OT speak the same language, and data from the Purdue Model can flow reliably to dashboards, BI and analytics.
🔧 How do you set up an information architecture?
A pragmatic approach in six steps:
- Map the information needs - which decisions must be made and what information do they require? Start with the processes, not the systems.
- Define a concept framework - define the twenty to fifty core terms and make sure business, IT and OT all endorse them.
- Reuse existing models - choose sector standards such as ISA-95, Aquo or the GEMMA models before designing your own.
- Assign owners - every data object gets an owner who is accountable for its quality and definition; that is the essence of data governance.
- Classify the data - determine for each type of data the requirements for availability, integrity and confidentiality according to the CIA Triad, and record whether it contains personal data.
- Model the flows - draw which data travels over which connections between which zones, including direction and frequency.
For the professional management of data there is also the DAMA-DMBOK (second edition, 2017), which treats data architecture, metadata, data quality and data governance as separate knowledge areas.
🔐 What role does information architecture play in security and privacy?
- Governance ensures ownership, quality and lifecycle management of data.
- Privacy by Design requires you to know where personal data is stored and what it is used for.
- BIO and AVG impose requirements on data minimisation, logging and transparency; the AVG (the Dutch name for the GDPR) also calls for a record of processing activities.
- Security by Design requires insight into which data flows where.
- Data integrity often matters more than confidentiality in OT: a manipulated reading can lead to a wrong control action.
Under NIS2, implemented in the Netherlands by the Cybersecurity Act (Cyberbeveiligingswet) in force since 15 August 2026, essential and important entities must manage the risks to their network and information systems, which requires knowing which systems and data they have. An up-to-date information architecture that shows the data flows between IT and OT makes that security risk assessment much easier.
❓ Frequently asked questions
What is the difference between information architecture and data architecture?
The two terms are often used interchangeably. Information architecture emphasises meaning, concepts and the information needs of the business, while data architecture focuses more on data models, storage and integration. TOGAF calls the domain data architecture; Dutch government architectures usually speak of the information layer.
Is enterprise information architecture the same as website information architecture?
No, these are two disciplines that share a name. Information architecture in UX, popularised by Louis Rosenfeld and Peter Morville’s 1998 book, concerns the structure, navigation and labelling of websites and apps. Information architecture in enterprise architecture concerns an organisation’s data and how it fits together.
Who is responsible for information architecture?
An information architect or enterprise architect designs and safeguards the information architecture. The content, however, belongs to the business: data owners decide on definitions and quality requirements. In IT/OT environments it is wise to give an OT engineer a say in the models for process and asset data as well.
Which standards are used for information architecture in OT?
For information architecture in OT, the most common standards are ISA-95 (IEC 62264) for the link between ERP and MES, OPC UA companion specifications for machine and device data, and the Asset Administration Shell (IEC 63278-1) for digital asset information. The process industry adds NAMUR Open Architecture. A Unified Namespace often uses these models as its structure.
How do you model an information architecture?
An information architecture is usually modelled at three levels: conceptual, logical and physical. ArchiMate is often used to show the relationship with processes and applications, and UML class diagrams following MIM for the information models themselves. More important than the notation is that terms are unambiguous and owners are known.
Does a small organisation need an information architecture?
Yes, even a small organisation benefits from a simple information architecture. A glossary, an overview of the most important data with owner and classification, and a sketch of the data flows between office and production systems are often enough. That prevents duplicate data entry and makes security risks visible.
📌 In summary
Information architecture defines which information an organisation needs, what it means, who owns it and how it flows between processes and systems. With concept frameworks, information models and standards such as MIM, ISA-95, OPC UA and the Asset Administration Shell, it makes data reliable and reusable in both IT and OT environments.