Skip to content

Report of the Digital Government Review


Open address data: an opportunity missed

So far, there is one comment on this section. Jump to comments

A good example to help us explore this problem in more detail is address data [42].

At the simplest level, address data can be seen as a list of house numbers, street names, towns and postcodes.

Address data is used in many processes across the UK: posting out and analyzing census returns; ordering a parcel from an online store; despatching an ambulance.

Addresses are created in the UK through a process involving local authorities, the Royal Mail and Ordnance Survey. Royal Mail and Ordnance Survey use the data to create different addressing products that are then sold. Some products are sold back to the public sector. The cost and complexity of licensing these products means that organisations often cannot use them, or use a version that is out of date, leading to difficulties for people in new homes [43].

Much of the effort spent in building or choosing from these different datasets is wasteful and unproductive. More importantly the lack of a single, authoritative set of address data also has an impact on people, thanks largely to the ever-increasing number of automated systems reliant upon address data.

Increased automation is leading to more and more decisions being taken on the basis of these address datasets. The effect of differing datasets could be as simple as a lost parcel or an inability to get home insurance; it could be as complicated as a mistaken statistical analysis of census data; but it could also lead to a misdirected emergency service call, a lost ambulance and a lost life.

This process started with local authorities, Ordnance Survey and the Royal Mail. Until the privatisation of the Royal Mail all of these organisations were public bodies while Address Data, like other geospatial data produced by the public sector, falls within the definition of data that would typically be open.

In other words, we are looking at a spectacular missed opportunity. Government has had both the opportunity and mandate to establish an authoritative open address dataset that would form part of our National Information Infrastructure; a dataset that could link to the open Land Registry data and start to resolve the whole area of geospatial data. Yet it has failed to do so.

This is not for want of advocacy. Many people were requesting this dataset recognizing that it would be appropriate for central government to take action to bring together these organisations and unpick the complex process and licensing models. Other countries, especially Denmark [44] have led the way in pushing such processes through with demonstrable benefits [45].

The time of the Royal Mail privatisation would have been the perfect moment for such an approach but, unfortunately, the opportunity was missed. No action was taken and both the Royal Mail components of the production process and the associated rights to the dataset were sold off. BIS commissioned a report on this topic that was published in January 2014 [46] but, again, no action was taken [47].

The consequences of this inactivity will only become more severe as technology automation increases in the future.

Government is now funding a community-led approach to investigate the feasibility of building a new open address dataset [48]. This approach may be successful, but if it fails then the next government should intervene.

The next census is in 2021. A census requires an authoritative address list so that census collections activities can be targeted where required. This provides a strong and realistic deadline for building a truly open address list.

Recommendation 9

Priority: High

As part of a general move to open up geospatial data the UK should have an open, authoritative and definitive address dataset by 2021. This will increase economic growth, reduce wasted effort and improve access to public and private services by all citizens

[42] Address data is a subset of a larger geospatial dataset, for the sake of brevity we will focus on this part of geospatial data
[43] http://www.bbc.co.uk/news/business-24960746
[44] http://www.theguardian.com/technology/2006/nov/30/epublic.society
[45] http://www.epsiplatform.eu/content/value-danish-address-data
[46] https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/274979/bis-14-513-open-national-address-gazetteer.pdf
[47] At the same time as this was happening, BIS was also working on the Building Information Modelling (BIM) standard (a standard which could form the next part of the chain from “land” to “addresses” to “the structures built on those addresses”).
[48] http://theodi.org/blog/open-addresses-discovery-phase

This page reformats automatically when printed. Print this section

Please note that comments left here are public - you can also make a private submission.

Your email address will not be published. Name, email address and comment are required fields. Please note we may moderate comments.

One comment

  1. Paul Malyon says:

    While I absolutely agree that an Open Address Dataset is required, 2021 is too late. The digital economy can not wait that long – business and social opportunities are being missed now.

    I’d welcome any moves to release an open dataset (whether a new one or the release of an existing product) within 12 months of a new government.