Acknowledgements

I am incredibly grateful to the Local Government Association, our research partner who supported this work and provided guidance and insight throughout. Thanks in particular to Gesche Schmid and Juliet Whitworth for their help with the project from the outset.

I would like to thank all those who gave up their time to be interviewed for the project. In particular I would like to thank Andy Graham (Newcastle City Council), Sarah Henry (Manchester City Council), Katherine Rooney (Bristol City Council), Stephen Blackburn (Leeds City Council), Gerrard Abi-Aad (Kent County Council), Michael Webb (London Borough of Camden), Roy Elmer (Suffolk County Council) and Raj Mack (Birmingham City Council) for their tireless help in organising the research and compiling the case studies.

I am grateful to the project's Advisory Group who were a great source of knowledge and insight. They helped shape the research programme and have steered the direction of this report. I am thankful to Chris Francis (SAP), Jen Hawes-Hewitt (Accenture), Duncan Ross (Datakind), Gary Forster (North Tyneside), Ed Parkes and Nick Halliday (Government Digital Service), along with Sarah and Gesche already mentioned above.

There are a number of people at Nesta who have helped steer this research and produce the report. In particular, I would like to thank Julie Simon, Eddie Copeland, Tom Saunders, Oscar Nowlan, Helen Durham, Geoff Mulgan and Stian Westlake.

The data maturity framework benefited greatly from the feedback and insight of Anthony Townsend (Bits and Atoms), Emma Prest (DataKind), Sian Basker and Madeleine Spinks (Data Orchard), and Lauren Haynes (Chicago University - Data Science for Social Good).

There are many other people who I have spoken to throughout the research, who presented at our workshops, and who have contributed to this report. There are too many to mention individually but I am extremely grateful to them all for sharing their experiences, knowledge and perspectives.

Any omissions and errors are those of the author.

Tom Symons

Local Government Association

The LGA is the national voice of local government, working with councils to support, promote and improve local government

www.local.gov.uk

Nesta...

Nesta is an innovation charity with a mission to help people and organisations bring great ideas to life.

We are dedicated to supporting ideas that can help improve all our lives, with activities ranging from early-stage investment to in-depth research and practical programmes.

www.nesta.org.uk

©Nesta 2016

Foreword - LGA

The common observation that 'information represents power' is never truer than in the area of local government. As providers of over 900 unique local services for the people and places we serve, councils assemble vast quantities of business and transactional content as part of day-to-day operations. This mountain of data provides the potential for impressive analytics, further innovation and a powerful evidence base.

It is only in recent years, however, that the potential for making better use of data within local government has been considered in earnest. The challenges of austerity, along with a drive towards improved efficiencies and transparency in the public sector have encouraged the creation of a more reliable evidence base from which decisions for future directions can be taken reliably.

Whilst local government has assembled these huge quantities of data, data base connectivity has been confined to a smaller number of forward thinking authorities. Silos of data are widespread in local government and, whilst they support their primary functions successfully, opportunities have been missed because data between councils (and even within councils) are often incompatible for connection, sharing and wider re-use for other purposes.

But information management practices within local government have improved substantially over the last decade, largely driven by the move towards increasing electronic delivery of services, a drive for productivity gains and improvements in the capabilities and skills of IT resources. It is arguable that local government is one of the most advanced sectors in relation to the work and progress made in the area of information standards and data collection.

When done well, information management is not restricted to IT standards and process monitoring. A whole raft of other important considerations are essential for a fool-proof local information infrastructure. Examples include: quality control measures, security, governance, data cataloguing, sharing protocols, timeliness of update, licensing and release.

There are a number of impressive success stories where pioneers have addressed the essential considerations above and have used data to support decisions and improve services or reaped the other benefits that better use of data can bring. These examples all required navigation of a series of barriers, direction changes, mistakes and strategy refinement along the way. However, the experience and lessons learned can encourage other local authorities to engage and move forward in similar ways.

During 2016, the Local Government Association teamed up with Nesta on their Local Datavores work, a research programme which seeks to explore how local authorities can use data better to improve the lives of people and communities. This involved engagement and research with groups of experts (data analysts, project managers, IT specialists and senior leaders) in local authorities and their partner organisations. From the information gleaned, the project team was able to review and analyse councils' uses of data. The result is a powerful and impressive list of innovative and successful projects. This report summarises the key findings and bases its content on a series of case studies drawn from widespread geographical locations and very different areas of local government responsibilities.

We hope that the materials provided in this report will encourage other local authorities (and their local partners) to begin to move into this space and feel confident that there are good sources of experience and support to help them make better use of data. For everyone, we can take heart that highly skilled data practitioners are in post, working daily to improve the use that is made from local data sets; and that, by working together, all authorities can benefit from the learning and experience of these pioneers and improve the use they make of data for local and regional decisions and service transformation.

Councillor William Nunn Chairman, LGA Improvement and Innovation Board

Executive summary

Think of a local government service area and the chances are that data can be used to improve it. Whether it's protecting vulnerable children, supporting older people to return from hospital independent and healthy, collecting the bins, fixing housing repairs, gritting the roads, or helping local businesses to grow, data can help councils to perform these tasks better. This report explores the many and various ways in which councils are already using data to make their services more effective, manage budget cuts and support the local economy.

This report, the second in the Local Datavores series, uses case study research to profile some of the most pioneering uses of data in local government. It features a wide range of councils that have been using data to inform decisions about adjusting to a harsher financial climate while maintaining high quality service levels. The research has explored the specific ways in which data can benefit people and communities, and how councils can make this happen. The report is aimed primarily at helping local public sector staff, from senior commissioners through to frontline professionals, get more value from their data.

The research has involved nearly 40 interviews with data analysts, project managers and senior leaders in local authorities and their partner organisations. Through these interviews and a literature review, the research analysed councils' uses of data, impact, their mindset and methodology, the challenges they encountered, the critical success factors they identified, and how these innovations could be scaled-up.

How and why councils are using data

The case studies revealed myriad ways in which data and analytics are helping councils across a wide range of statutory and discretionary services and responsibilities. The research found that the benefits are also many and numerous, helping councils to save money, make their staff more productive and their services more effective, support local businesses to grow, and reduce environmentally harmful activities. The ways councils are using data to realise these benefits – the key use cases fall broadly into 11 areas of local government activity:

Data-led innovation - making it happen

Realising these benefits is neither straightforward nor insurmountable. Any council undertaking data-led innovation should expect to encounter a least some of the challenges the case study councils contended with. For instance, they are likely to find they need to spend time making their data of useable quality, that legislation about information governance will offer few black and white answers, or that pulling data together from legacy IT systems will feel harder than it should. Councils undertaking data work are likely to share the case studies' experience that embedding data and analytics in council services and decision-making can be challenging for some working cultures and staff.

The case studies covered different types of data work in different council service areas, but there were striking similarities in what they told us about how to make data projects work.

Many reported that they had sought expert legal guidance and invested significant staff time in developing highly specific information sharing agreements. Data quality was usually tackled head on, by attempting to use the data, exposing and fixing quality problems. To help their colleagues and working cultures be more receptive to using data they engaged with end users about their needs and requirements. They built data models and analysis that responded to these needs, for instance saving staff time or helping them do their jobs better. Their working methods involved starting small, being flexible, testing and iterating, often without a fixed idea of what the final product would look like. Finally, they all worked to bring senior managers and leaders on board as champions of their work, helping them push through barriers such as a reluctance to share data or use data tools in day-to-day work.

Taken together, the experiences of the case studies can be seen as a how to guide to undertaking data-led innovation in local government. The lessons they learnt can be lessons for the whole sector. Collating these insights offers a ten-step guide to data-led innovation in local government:

1. Start with a clear problem to be solved, for which data can offer impactful and actionable insight.
2. Gauge the level of support for data-led work in senior leadership and work to convince them of the importance of the project.
3. Start small, engage with end-users to find out where and how data could be used to make their day-to-day work easier.
4. Be clear about ultimate objectives and how these will be measured.
5. Ensure there are realistic financial and staff resources allocated to the project.
6. Approach the work through a series of short, repeatable work cycles, which enable rapid development, testing and iteration.
7. Secure dedicated expertise for information governance and be specific about the purposes of sharing data.
8. Test the product with end-users and take on board their feedback.
9. Be receptive to making decisions informed by data.
10. Evaluate the overall impact of the work against the original objectives.

The work profiled in this report has all happened recently, in the period from 2011 onwards. Even in that short period, analytical capabilities have increased massively, as has the ambition of local authorities to use their data. The opportunities to get more value from data are increasing all the time.

There is a range of ways in which progress can be accelerated:

• Opportunities to embed more data analysts with frontline staff and in management teams should be explored to promote further data-led innovation.
• There is a need for a programme which can spread the knowledge and expertise of the people involved in these and similar projects.
• The integration of data across local authorities brings huge potential and a quicker route to impact. This should be a feature of devolution deals and other local partnerships.
• More systematic evaluation is required to ensure we are continuing and replicating the right data tools and analysis.
• And finally, there is scope for the learning about how to use data better in local government to be turned into practical tools and resources to help other councils.

Over the next year, Nesta will be exploring these opportunities and others to help government - central and local - get more from the data they hold. This report features ways of using data which are having hugely positive impacts for people, communities and councils. We want these to be the start of the story, not the end.

Introduction

At the outset of government austerity in 2010 there were two dominant ideas about how to address public sector funding and service pressures; streamlining and sharing back office services to reduce waste, while integrating and coordinating frontline services across the whole local public sector so that they are more effective. As councils have followed these two approaches, it has become clear that one of the essential ingredients in making both work is the better use of data and analytics.

Local authorities sit in the middle of a web of information. Everything from social care for vulnerable children, waste collection, procurement, council tax collection to planning applications produces huge quantities of data. This data is sometimes garbled, hard to analyse or personal and sensitive. But it is potentially very useful in helping councils to make services more targeted and effective, to allocate resources to where they will have the biggest impact, to save officer time in front and back office processes, and to provide insight into the causes and solutions to costly social problems.

Running a city or a local authority is to a great extent about managing and responding to information. Increasing digitalisation of services, the use of sensors and other forms of data collection mean that there are emerging data sets which capture the wide variety of activities performed by councils. And while big data presents opportunities for local councils, there are equally important opportunities presented by smaller data sets already available to councils. Whether the data sets are big or small, there are major benefits to be had from using them more intelligently, sharing them more widely and making them more open.

The need to do this has scarcely been greater. Local government is entering its seventh year of austerity, with spending cuts set to continue until 2020 at the earliest. The central government grant is forecast to be reduced by 54 per cent between 2015-2020, to just £5.4 billion. Newly permitted increases in council tax rates may mitigate some of these reductions, keeping funding levels broadly constant in this period, though there are concerns this will not be enough for all councils and may not cover increases in wage costs through the rise in the national minimum wage.¹ Furthermore, set against increases in demand, such as pressures in adult social care caused by an ageing population, it is clear councils will need to keep finding new ways to reduce costs while maintaining service quality and meeting their statutory obligations.

This research focuses on the specific ways in which data can help councils to meet these challenges. This is not to argue that data is a simple or complete solution. It is just one tool of many that councils will need to use. But while data on its own may not be enough, without making the most of the data that is available, an already tough task will be a lot harder.

A note on terminology

Nesta developed the concept of a 'datavore' in the 2012 report Rise of the Datavores. This defined datavores as companies which:

“ Gather online customer data intensively, subject this data to sophisticated analyses (such as controlled trials and data and text mining), and use what they learn to improve their business. They also report that they are more innovative than their competitors, in products as well as processes.”²

For the purpose of this research programme, we define local datavores as councils which:

"Intensively gather data about people, communities, places, businesses, council processes and services, infrastructure and the environment and subject this to sophisticated analyses. They also seek to share, integrate and use data where possible for the improvement of services and operations. They use what they learn to inform decisions about improvements to council operations, processes, services and infrastructure, and to ensure they meet the needs of their residents.”

Section 1 introduces the case studies from the UK and overseas. The case studies cover a range of data uses, social issues and benefits.

Section 2 outlines the benefits of using data as a tool of improvement in local government. These use cases are grouped together under 11 areas of council activity or service.

Section 3 explores the eight in-depth case studies and the detailed accounts of their work, methods and impact. The challenges they encountered, their strategies for overcoming these are considered. The section ends with a step-by-step guide to undertaking data-led innovation in local government, based on the lessons offered by the case studies.

In the conclusion, recommendations for increasing the use of data in local government, based on the insights of the work, are outlined. These include a programme to promote further data-led innovation, and a twin programme to help scale and replicate the most effective data-led projects from across the UK. The need for better cross-authority collaboration as explored in Nesta's work to establish Offices of Data Analytics - is also outlined.

Section 1 - The case studies

This report uses case studies to showcase leading examples of data-led innovation and articulates the value of better data use and analytics by councils. The case studies chosen reflect a range of council service areas, types of data use, types of council, region and geography. The case studies are grouped across four themes which represent the core of local government activity, both in terms of service and funding pressures:

• Children's social care and troubled families.
• Adult social care, health and public health.
• Smart places and local economic growth.
• Public service transformation and open data.

The eight case studies

Full summaries of the case studies can be found in Appendix 1. Below is a short summary of each.

Leeds City Council - opening data to drive innovation and transparency

Theme: Public service transformation and open data

Leeds City Council has created one of the largest open data portals in the UK. The objectives were to:

• Increase transparency about the council's work, decisions and spending.
• Reduce the amount of time officers spent dealing with Freedom of Information requests.
• Bring businesses, developers and citizens together with the council to collaboratively solve problems.
• Support local economic growth by giving business new sources of intelligence and identifying where the council's resources could support the local economy.

Data Mill North hosts over 300 data sets from over 40 different organisations. The council has used the Data Mill to proactively engage with Leeds' community of developers and civic enthusiasts through Innovation Labs. These are day-long events which address specific problems and award funding to the idea which best uses open data to tackle them.

Benefits

• The council is implementing new procedures which will increase the number of published datasets to reduce the number of FOI requests.
• New product and services have been created which benefit the people of Leeds, such as an app which helps residents to put the right bins out on the right collection days.
• Open data has enabled analysis of the areas in Leeds which had the highest cycling accidents, informing plans to improve cycle safety.

London Borough of Camden - Camden Residents' Index

Theme: Public service transformation and open data

Camden has created the Camden Resident Index and data dashboards to make its public services better. The objectives were to:

• Identify and eliminate instances where staff activity was duplicated.
• Make delivery of services such planning and housing repairs quicker, higher quality and less costly.
• Identify where fraud and error were taking place.

The Camden Resident Index brings together data from 16 council IT systems, covering 123 fields of primarily demographic information. This is used to streamline processes and provide a more responsive service to citizens. Bespoke data dashboards give managers a view of operational data which is geared towards customer satisfaction, making it easier to improve services.

Benefits

• Data has been used to improve public service delivery and back office processes, required to meet a funding shortfall of £70 million.
• Identified fraud, such as housing tenancy, blue badge or schools applications fraud.
• Supported changes to services which improved customer satisfaction and reduced how much officer time was used in delivery.

Newcastle City Council - data-led transformation of children's social care

Theme: Children's Social Care and Troubled Families

Newcastle has used data and analytics to improve the way that children's social work is delivered. The objectives were to:

• Improve the way that social workers deal with complex family needs by giving them feedback and research about what works best.
• Make is easier and less time consuming for social workers to manage their caseloads.
• Identify the factors which are predictive of poor and costly outcomes.

Data was used to change the way Newcastle delivers long-term children's social work. Social work units now specialise and work with a caseload of families who all have similar needs and circumstances. The family groupings are based on data analysis of 'concern factors' when children become known to social care services. Data analysts are embedded within the social work units and develop data dashboards for social workers, review team performance, undertake research and analysis about 'what works', and provide deep-dive analysis into the needs of children and parents. These insights feed directly into how social workers help some of the most vulnerable families in the city.

Benefits

• Social workers become more specialised, improving the quality of the support they provide to children and families.
• Data dashboards make it quicker and easier for social workers to manage their caseloads.
• Data analysis provides rich information about how social workers can best help children and families.
• Cases are now closed more quickly, in comparison to traditional social work methods.
• Fewer additional services are required for families known to children's social care.

Manchester City Council - using integrated data to improve family support services

Theme: Children's Social Care and Troubled Families

Manchester has developed a data warehouse for children's social care that integrates 16 different data sets from multiple agencies. The objectives of the project were:

• Identification of families in need of additional support.
• Quicker access to better quality information about families at the point of assessment.
• Testing whether services are achieving their stated objectives.

The system is used by frontline workers to make assessments about the type of support children and families need. Data integration provides instant access to information about family genealogy, involvement with different agencies such as the police or schools, and relationships to other families also known to local services. Manchester can also use this data to make decisions about which types of service are needed, for whom, and at which points.

Benefits

• Increases staff productivity, saving an estimated two weeks a year for each key worker.[^4]
• Improves the quality of decision-making by key workers on the frontline.
• Gives decision-makers quality information about whether to commission or stop services.
• Identifies factors which are predictive of poor and costly outcomes, such as involvement with the children's social care system.

Bristol City Council - the citizen-centric smart city

Theme: Smart places and local economic growth

Bristol runs a smart city programme which is open, collaborative and responsive. The objectives were to:

• Improve how Bristol manages factors such as congestion, emergency incidents, weather and issues that are important to the public.
• Collect data about citizen needs and preferences, and to respond to these.
• Make all data that isn't confidential open to the public.

Bristol has created a digital and data infrastructure that enables collection of data from a wide range of sources. An Internet of Things mesh network enables citizen generated data, along with data from sensors, CCTV cameras and operational services, to be used by the council. Harnessing this data allows the city government to focus on everything from reducing congestion to how best they can respond to citizen concerns about household damp. Bristol uses an open data portal to make as much data as possible available to the public.

Benefits

• The open data portal has supported the creation of a number of apps and services which respond to needs identified in Bristol.
• Collecting citizen-generated data can enable the council to give people the tools and know-how to solve problems they face, such as household damp.
• Bringing data about the city into an operations centre will make Bristol run better, from quicker journeys to improved responses to emergency incidents.

BirminghamCity Council - Digital Birmingham

Theme: Smart places and local economic growth

Digital Birmingham is a specialist unit tasked with crafting and implementing a smart city strategy for Birmingham. The objectives were to:

• Help the city adapt to future challenges, using technology and data to help people live, learn and work better.
• Make the city better able to respond to problems such as adverse weather or congestion.
• Be a test-bed for new digital technologies which can improve the lives of citizens.

Birmingham has undertaken smart city initiatives to improve city management, such as temperature sensors in roads so that gritting can be targeted when it snows, and motion sensors in 5,000 free bikes provided to the community to help plan new cycling infrastructure based on the most common journeys. Birmingham also has an open data portal which local developers and businesses can use to create new products and services.

Benefits

• Birmingham can provide more efficient and targeted services by using data, such as only gritting roads which temperature sensors indicate have snow or ice.
• Collecting data can bring insight into what citizens need, such as where new wheelie bins should be located.
• Data collected can test the impact of experimental schemes, such as whether free bikes improve people's health outcomes.

Kent County Council - informing commissioning with the Kent Integrated Dataset

Theme: Adult social care, health and public health

Kent has developed one of the largest integrated health and social care datasets in the UK, covering a population of around 1.5 million people in Kent. The objectives were to:

• Help Kent council and health services understand the impact of changing the services provided in the county.
• Rigorously test whether services were achieving their stated objectives.
• Identify where there are variations in performance levels in the county.

The data set is comprised of pseudonymised data from acute, community, mental and primary health services and adult social care. This data set enables the public health team in Kent to conduct sophisticated data analysis about what would happen if changes to services or policies were made. Having an integrated data set of this size also means Kent can test 'what works' in the county by running matched cohort analyses. These compare the outcomes of people who receive a service or intervention with a statistically similar group who didn't. This indicates whether differences in outcomes are attributable to the programme.

Benefits

• Decision-makers get high quality information about the possible impact of changes to health and social care services.
• Decisions to reduce or expand services can be made on the basis of rigorous evidence of effectiveness.
• Analysis can highlight where remedial action is needed because services are not performing as they should.

Suffolk County Council - integrating health and social care using data

Theme: Adult social care, health and public health

Suffolk council and local health services are using data to help them create an integrated service. The objectives were to:

• Shift funding towards preventative work and away from costly reactive services.
• Improve the culture and effectiveness of performance management.
• Support the co-location and integration of frontline teams in health and social care.

To do this Suffolk has integrated data from health and social care, changed the performance metrics used by managers, and undertaken analysis of future service pressures and customer journeys.

Benefits

• Suffolk can diagnose problems in the system more easily, such as why there are high numbers of people who have Delayed Transfers of Care from hospitals to residential care.
• Data integration means new preventative programmes can be evaluated better and supports planning for further preventative work.
• Performance management is now based on meaningful outcome measures such as wellbeing, rather than crude input or output measures.

Additional case studies

Alongside the eight in-depth case studies, a range of innovative approaches to data use by councils in the UK and abroad are used to illustrate use cases and benefits.

We also feature case studies from Copenhagen, Jakarta, Helsinki and Allegheny.

Section 2 - The benefits of using data

Data is a tool which helps councils perform a range of activities better. Across the case studies, we saw some of the many ways this can happen. This section collates these examples of data use and describes the ways in which this benefits people, communities and local businesses.

In our earlier discussion papers [^5] and background research for this paper we identified three basic principles for getting value from data:

• Be driven by a problem to be solved or hypothesis to be tested.
• Seek data and analysis that will lead to actionable and impactful insights.
• View data as one tool of many that will help outcomes or objectives to be achieved.

Using these principles as a starting point, we analysed the case studies to see how they had used data and how this had created value. The ways in which data has been used - referred to as use cases - are grouped together according to different kinds of activity councils can undertake. For this report, a use case is defined as 'a process, action or task that can be improved using data'.