2020 Shortlist

Locked Out

Category: Innovation (small and large newsrooms)

Country/area: United Kingdom

Organisation: The Bureau of Investigative Journalism, members of the Bureau Local Network

Organisation size: Small

Publication date: 10 Apr 2019

Credit: Maeve McClenaghan, Charles Boutaud, Tom Blount, Charlotte Maher

Project description:

Our #LockedOut investigation revealed just how hard it was to find an affordable property to rent while on housing benefits. We dug into data from across Great Britain to show, not only the state of affairs nationally, but also the local, granular picture. This collaborative project drew in journalists from across the country to tell an important story about the housing crisis and a leading cause of our spiralling homelessness epidemic.

Impact reached:

The work was welcomed by campaigners and lawyers across the country, with some officials saying it would help them create local policy. We heard of at least one legal case where lawyers used our data to prove their client and his family were unable to access an affordable home.


The work also prompted the UN Special Rapporteur on Housing to speak out on the issue and, inspired by our work she said she planned to look into affordability as an issue elsewhere. 


Three months after publication the government announced an increase in the rate of Local Housing Allowance (housing benefit), though we have since shown this increase does little to improve things.

Techniques/technologies used:

We captured the details of 62,695 two-bed properties advertised on 15/09/2019, using the Nestoria API (which aggregates property data from online property listing sites.)


We used the latitude and longitude coordinates of each property found through the API to localise them within the right Broad Rental Market Area (BRMA) using shapefiles obtained from the Valuation Office Agency. We then analysed the data to ascertain the total number of affordable properties in each BRMA, as well as the increase in LHA that would be necessary to make the 30th percentile affordable.


We also knew that refusal to let to those on benefits makes the shortage even worse. Reporters contacted the landlords of 180 two-bed properties posing as a single mother and asked whether the landlord accepted people on benefits.


We also went further to ensure wide accessibility to our findings, building an interactive online tool which allowed anyone to find out their local situation. 

As the housing market is ever evolving, it was clear that we needed to ensure others could reproduce our investigation later, so we published the code we used to collect and analyse the data in a Github repository.

What was the hardest part of this project?

We knew this was an investigation with regional and local nuance, and so it was imperative that the findings were shared as far and wide as possible.


We worked on this story with our Bureau Local network. We wrote a public-facing Reporting Recipe and opened all relevant data at BRMA level (which is not publicly available) as well as a list of all the affordable properties. As well as publishing the story with HuffPost UK, our local reporter network produced more than 20 stories detailing the situation in their area.


We also held StoryCircles across the country – presenting our data and findings to people with real-life experience in physical settings.

What can others learn from this project?

This project combined innovative data gathering and mapping techniques, paired with a real focus on making the findings accessible and useful to as many people as possible. This collaborative approach to data-journalism is breaking ground in the UK. The Bureau Local project (part of the Bureau of Investigative Journalism) has developed a clear framework for how these investigations can be done: we publish Reporting Recipes to help even someone with no journalistic experience dig into the story, we open up our data with clear guidance on how to read it, we have published the code we used so others can re-produce the investigation at a later date and we have even produced a guide to help people should they want to replicate our model for themselves.

Project links: