Unlocking the value of fragmented public data

As a joint project between mySociety and the Centre for Public Data, we have written a set of simple principles for how to get the most impact out of publishing public data.  You can read the report online, or download it as a PDF

Fragmented public data is a problem that happens when many organisations are required to publish the same data, but not to a common standard or in a common location. Data is published, but without work to join up the results, it rarely has the intended impacts. 

The results of this are frustrating for everyone. Data users cannot easily use the data, policy makers do not see the impact they want, and publishers in public authorities are required to produce data without seeing clear results from their work. 

Better and more consistent publication of data by local authorities helps enable understanding and action at scale across a range of areas. At the same time, we recognise that the technical advice given has assumed higher levels of technical capacity that in practice is possible for many data publishing tasks. Our goal has been to make sure our advice makes data more accessible, while having a realistic idea of technical capacities and support needed for data publishing. 

This report recommends three minimum features for a data publishing requirement to be successful: 

  1. A collaborative (but compulsory) data standard to agree the data and format that is expected.
  2. A central repository of the location of the published data, which is kept up to date with new releases of data.
  3. Support from the data convener to make publication simple and effective – e.g. through validation and publication tools, coordinating returns, and technical support.

We recommend that:

  • Whenever government imposes duties on multiple public authorities to publish datasets in future, it should also provide the staff and budget to enable these features.
  • The Central Data and Digital Office should publish official guidance covering the above.

You can read the report online, or download it as a PDF

Better data publishing helps climate action

This project is informed by recurring frustrations we have run into in our work. Projects such as KeepItIntheCommunity, which mapped registered Assets of Community Value, were much more complicated than they needed to be because while transparency was required of authorities, coordination was not – meaning the task of keeping the site comprehensive and updated was enormously difficult. In principle, we could build a tool that empowered communities in line with the intentions of the original policy makers. In practice, a lack of support for basic data publishing made the project much harder than it needed to be.

This problem also affects our work around local government and reducing emissions. Local government has influence over one third of emissions, but much of that is indirect rather than from the corporate emissions of the authority directly.  As such, many activities (and datasets) of local government have climate implications, even if the work or data is not understood as climate data. For instance, the difficulty in accessing the asset data of local authorities makes it harder for civil society to cross-reference this information with the energy rating of properties, and produce tools to help councils understand the wider picture. 

In future we will be publishing in more detail the kind of data we think is needed to support local authorities in emission reduction – but emissions reduction cannot be isolated from the general work of local authorities. Improving the consistency of the data that is published helps everyone better understand the work that is happening, and makes local government more efficient. 

Sign up to hear more about our climate work, or to the monthly mySociety newsletter

Photo credit: Photo by Olav Ahrens Røtne on Unsplash