Data analytics Homing in o elessness has proved di icult for govern ents and councils to resolve for decades an data help find a solution he entre for o elessness pact is trying to find out y Liam Kay O ne in every 200 people in the UK is homeless. This stark figure, in housing charity helters analysis of government figures, shows the scale of the problem facing the government, local authorities and charities. The issue is not ust a social and humanitarian one it is also financial. The report At what cost, by homelessness charity risis, estimated that people sleeping rough could cost , a year, depending on the services needed. ata analytics can help councils decide how to use their sometimes scant resources wisely when tackling homelessness. The entre for omelessness mpact, a charity that uses evidence tools to help address homelessness, worked with the ffice for ational tatistics to develop a platform called hare, with the aim of informing how preventing and ending homelessness should be framed and assessed. The charity also runs a ousing ost alculator, which uses councils data to estimate how much it would cost to implement different housing policies for homeless people. ouncils need to have data showing the number of homeless people in their area, the types of accommodation they are in, and those peoples characteristics, such as the support they re uire and whether they have recourse to public funds. Part of its ob ective was to help reduce costs by taking all relevant data into account, providing a high level overview of the solutions available to local authorities to house homeless people such as whether to move them to temporary or permanent housing and using it to address the challenges specific to the local area. The calculator emerged from rethinking the centres priorities and understanding the challenges councils face in tackling homelessness, according to uillermo odr gue u m n, head of evidence and data at the entre for omelessness mpact. The first councils the centre worked with were outhwark, utton and ast yrshire. sing publicly available data, the reporting platform shares more than homelessness indicators, covering a range of sub ects from local authority homelessness applications to house building and public attitudes. n cases where there was a lack of information, the centre used assumptions based on the estimates of council employees. These were especially needed in the escalating and rapidly developing ovid crisis, when thousands of homeless people needed to be housed uickly, o en in hotels, to help stop the spread of the virus. side effect of the use of council estimates has been to underline their employees expertise, says odr gue u m n. hen compared with more recent data published later, the team was pleasantly surprised at their accuracy. The local authorities are in contact with the realities of homelessness, so connecting with them could be a lot more insightful than ust waiting for data on the sub ect to be released three months later, says odr gue u m n. ou need to trust people on the ground. o one was capturing information, so we turned to those we expected to know better and it turns out they did. nce the tool has the basic data it needs from local authorities on the homelessness issue in their area, it creates scenarios and estimates the cost of moving people between different types of accommodation. t does this by combining action areas, goals and indicators, and is based on data and estimates supplied by local authorities. or example, for someone moving into social housing, it would estimate furniture costs as part of the scenario. The aim is to underline the long term savings that could be made by moving someone off the street permanently, and reduce the burdens on multiple services by improving homeless peoples lives. ne of the points we wanted to make was that, if you move everyone to settled housing, you are not only ensuring they have a house for a longer period, but it is also cheaper for the local authority, says odr gue u m n. That was a really important element in how the tool was designed. eveloping the calculator was not without challenges. The benefits system is highly complex, which meant that under s had to be excluded, while other people were sub ect to benefits caps, depending on their personal circumstances. ata uality was patchy across council areas, and there was a lack of harmonisation between the four nations of the . The tool has helped in practice. loomberg ssociates, the consulting arm of charity loomberg Philanthropies, works with cities to improve the uality of life for residents. t worked with the reater ondon uthority, using the centres tool to look at how to move homeless people into permanent accommodation in the city, the likely cost of doing so, and whether costs could be offset. n some cases, using the tool reduced the costs by . 44 Impact ISSUE 31 20_pp44 Data.indd 44 18/09/2020 11:52