Data Standards
Data standards are often considered a dry topic. They become relevant, at the very latest, when it becomes clear that while your own data works within your own system, it can hardly be shared with other systems or organizations. What follows is usually a time-consuming process: manual modifications, errors, and dependencies.
What is a standard—and what isn't?
When people think of data formats, many think of file types like Excel, PDF, or Word. Formats describe how data is stored or displayed—in other words, its external form.
A data standard governs something else: It defines what the data means and how it is structured. Two organizations can use the same format and still fail to understand each other because they define “address” or “date” differently. The format is the shell; the standard determines what’s inside.
Here’s a simple example: An Excel file with a “Date” column is a format. But whether 01.03.26 means March 1 or January 3—that’s determined by a standard.
Three Types of Data Standards
1– Technical standards govern how data is structured and transmitted. These include formats such as CSV, JSON, and XML, as well as protocols and interface standards.
2 – Semantic standards define meaning: What do we mean when we write “address” or “diagnosis”? A semantic standard ensures that everyone understands these terms in the same way—regardless of which system processes the data.
3 – Governance standards set the rules: Who is allowed to access data? Under what conditions may it be shared? How long is it retained?
In practice, all three levels are interrelated.
What Standards Can Do—A Concrete Example
A municipality collects resident data: name, address, and date of birth. The canton needs the same data for tax collection. The health insurance company needs it to collect premiums, as does the Federal Statistical Office.
Without standards, every organization uses its own terminology. This means that before data can be shared, it must be manually converted each time. Errors creep in and extra work is created—over and over again, always in the same places.
With standards: Data is entered correctly the first time and can be used directly by all parties involved. No manual conversion, no data entry errors, and no duplicate work.
Data standards are, therefore, infrastructure: They’re not very visible in everyday life, but they form the foundation for seamless data exchange.
The advantage of standards: automatic verification
An often-overlooked advantage: When data is structured according to a standard, it can be checked automatically to ensure that the rules have been followed: Is a required field missing? Is the format incorrect? Was an unknown term used?
The result is simple: fewer errors and a solid foundation for more advanced applications such as searches, analyses, and data exchange.
Standards and Digital Sovereignty
Data standards are not merely a technical means of making processes more efficient. They are also a prerequisite for ensuring that data can be used as a societal resource—and not just as an asset of individual companies. Those who view data as a common good need standards that enable access, create transparency, and counteract the concentration of power. Open, collaboratively developed standards provide a structural foundation for this.
Data standards are a key lever. They can strengthen or weaken oversight, depending on how they are developed.
Standards strengthen sovereignty when they:
are open and accessible to everyone (“open standards”)
be developed collaboratively by many stakeholders—not by a single company
enable switching between different providers (“interoperability”)
Standards undermine sovereignty when they:
are proprietary, meaning they are owned exclusively by a company
Create a lock-in: Anyone who stores data in a proprietary format will find it difficult to switch away—and is therefore dependent on a single provider
are developed in a non-transparent manner, without input from those affected
This is also evident in the example of the municipality: If you store administrative data in a proprietary format from a single software provider, you can hardly switch providers. The data is effectively locked in. Open standards prevent exactly that.
Who sets the standards?
Standards are developed in various ways:
Standardization bodies such as ISO (international), CEN (European), or SNV (Switzerland) develop formal standards
Regulations set certain standards—such as the EU’s Data Act or the European Data Strategy
Market and Practice: Sometimes a standard simply becomes established because everyone uses it. The key factor then is whether it remains openly accessible and whether others can help shape it—or whether it effectively remains the property of a single company. One example is the PDF format: It was originally developed as a proprietary format by Adobe and later became an ISO standard.
Collaborative Development: Open-source communities and industry associations work together to develop standards and make them available. In Switzerland, for example, this is done through the eCH association, which develops e-government standards—such as those for the exchange of resident data between municipalities and cantons or for the standardization of parliamentary data.
It is important for Switzerland to note that many key data standards are developed internationally—for example, at the European level or within global standards-setting bodies. The question of which standards to adopt and how Switzerland should participate in their development is therefore relevant not only from a technical perspective, but also from an economic and political standpoint.