This handbook is about policy-related open data standards. These cover both content (what to publish) and representation (how to publish).

The handbook describes maintaining data specifications that can support the realization of policy goals.


A campaign calls for organizations to publish their pay ratios between men and women, and highest and lowest paid employees, in order to highlight and address pay inequality. This is the policy goal.

A schema is developed and documented, describing how to publish CSV files on an organization's website that provide these figures. This is the data specification.

Further documentation, data-quality frameworks, and validation models are developed that can be used to check that the data provided using the specification is accessible, meaningful and actionable. Together with the specification, this constitutes the open data standard.

The success or not of this standard rests first on data availability and interoperability, and then on the data being usable, useful and ultimately put into uses that can drive policy outcomes.

The originators of a standard may have one or more objectives in mind, but in many cases a standard has a greater chance of being adopted when it responds to a wide range of user needs.

Through an iterative development approach, looking at both the 'supply side' and 'demand side', it is possible to develop standards that will work for a wide range of different stakeholders.

Most policy-related open data standards, once released, need ongoing maintenance. The maintainers of a standard can make improvements based on learning or in response to policy developments. Maintenance, and a process for gathering and acting on feedback, also ensures that implementation guidance reflects the experience of standard users. The maintenance system will often require clear governance.