Information architecture is a central part of business architecture. This has the greatest potential for value creation in digitisation. The key to success lies both in the modelling itself, but also in the methods and techniques for organising models. All modelling and documentation is done with clearly defined use cases.
Based on the business
All information models represent the business and reflect the concepts used to manage and operate the business. It can therefore be developed together with experts in the business. Computer experts can participate in the development, but it is not their insight that is the basis for the models.
The most important foundation for a quality information architecture is based on the principle:
“Data is obtained only once, at the source, and is used many times for all authorized needs”.
Wholeness and detail
Our approach is based on one overview model and many detailed models. The development is done with several iterations between wholeness and detail. There are clear rules for how each detailed model fits into the overview model. Likewise, there are clear rules for how all detailed models “connect”. A quality information architecture can be established without the use of advanced modelling tools.
Pure models vs. mixed models
We first develop pure, well-defined models. This applies to both information and processes. When these are stable, they are related in matrices. It gives us compact overviews and a basis for analyses.
Mixed models with illustrations may be relevant for communicating certain messages to people who do not directly participate in architectural or development work.
The information architecture is necessary for many decisions and further design work. Here are some:
– identify well-defined building blocks for databases
– design of the company’s digital platform for all future application development
– identify sources and applications for information (and data)
– define data responsibility
– identify unwanted redundancy of data and functions
– identify dependencies between information, and derive a data-ready sequence for development