Suomi.fi for Service Developers
Go directly to contents.
Best practices for developers
API design in public administration

Find or create an open data model

Find out about regulation

Regulation often specifies in detail

  • what data a public sector organisation may collect and process
  • what purpose the data may be used for.

The collection and processing of personal data, for example, are subject to detailed regulation. This is why it is a good idea to start designing the data model by familiarising yourself with regulation, in other words regulatory interoperability.

Updated: 6/5/2026

Design the data model

In an ideal situation, you will already have considered the data model design, in other words semantic interoperability, before the procurement. In practice, the procurement can start with designing the data model, vocabularies and code sets of the information system in question

  • either for the entire procurement, or
  • to the extent in which the object of the procurement exchanges information with other information systems.

The data model describes the relationships between vocabularies and code sets. It may also include the vocabularies and code sets. An open data model will be freely reusable.

The data model links the vocabularies, code sets and records into a hierarchical structure.

Updated: 6/5/2026

Design the database

Each information system comprises a database. When designing the database, you specify its data types, including text, numbers, or dates.

However, field names do not always unambiguously indicate the meaning of the data, which is why semantic interoperability is also needed. In practice, for example a 'name' field in a database may mean different things, such as the name of a person, animal or object.

Updated: 6/5/2026

Understand data modelling and be aware of information needs

You should note that creating data models, code sets and vocabularies takes time. When creating them, it is advisable to ensure that the publishing organisation has a solid understanding of modelling the data it publishes. A good example of this is the Ministry of the Environment's Ryhti project, which created open and reusable data models.

The better the information needs are identified, the more efficient and targeted development will be. If the requirements are known, draft APIs can be produced even before the tendering process starts.

Well-specified data models

  • reduce work
  • simplify implementations
  • reduce the risks associated with data use.
Updated: 6/5/2026

Develop the data model in stages

You cannot always specify the data model before software development or procurement starts, and the need to develop it may come as a surprise.

In agile development, the data model can be developed in stages. You can initially focus on architecture design, which will speed up development and reduce work later.

Updated: 6/5/2026

Are you satisfied with the content on this page?

Glossary