7 ENTERPRISE ARCHITECTURE PRINCIPLES: MAXIMIZING SUCCESS

Should we follow enterprise architecture principles while working on an MVP?

There are a few core architecture principles that need to be adhered to alongside the more frequently used software architecture principles. Product Managers aren’t typically aware of the long-term value and tend to push them into a backlog for later release or sprints. There is nothing wrong with this approach except for the unknown constraints and risks that result in the tasks never getting implemented.

From working on a few Products, I would like to highlight 7 core principles that any product architecture must make a priority:

· DATA LIFECYCLE MANAGEMENT
· ARCHITECTURE FRAMEWORK
· API INTEGRATION/MANAGEMENT
· BUILD TO SCALE (NOT ONLY DEPLOY TO SCALE)
· IDENTIFY TARGET/FUTURE STATE AND TRANSITIONAL ARCHITECTURE
· PRIVACY AND LAW COMPLIANCE
· KNOW HOW MANY “SERVICES” ARE JUST ENOUGH

Data lifecycle management

Frequently, I have observed how we seem to prioritize decisions on data management, role/group based access and backup.

It’s a good place to start as we seek, persist and manage but we can’t forget data lifecycle management or DLM that has a much broader scope.

Imagine working for a healthcare service for example. The volume of data that could feed into their system will be in the TB range, depending on their services, network presence and a number of hospitals. There are also specialized services within a hospital that need different formats, size, and types of data in order to ensure a proper diagnostic of the patient.

However, there are many issues with the use of data, such as:

· What happens to the data once the patient is discharged?
· Do we only want inpatient or outpatient data or both?
· Do we need to remove the data from our systems?
· How can we make data lifecycle management cost-effective?
· Can one data set have a reduced life cycle compared to others?

These questions usually depend on the sector you work in and the corresponding government policy. However, as architects, we need to be more aware of data lifecycle management by getting in touch with officials to find up-to-date regulatory information.

For example, in the Canadian province of Ontario, the law was changed from a seven-year data lifecycle to an indefinite period and with an exemption for clinical data. This had a major impact across the hospital service, insurance industry, and clinical care providers.

Remember, this awareness around data will also help identify your technology strategy and road map, which could result in less (unpleasant) surprises for the future.