Our analysis of Birmingham City Council’s outcomes and service use data demonstrated how data can be better used to produce novel insights and a more complete picture of need in the early years. Our analysis of the data found that, when judged by a combination of performance on early years outcomes and take-up of early educational entitlement, the most in-need areas were not on the radar of the services in a position to provide any additional support. We also found that the most in-need areas may require different solutions. By combining locally held data with census data, we found that these areas were not homogenous in terms of their sociodemographic characteristics. This has important implications for how solutions and services are designed in response to findings.

Our co-design methodology enabled Birmingham City Council and its health and care partners to agree key use cases and a process for creating an early years data tool which combines the most up-to-date early years outcomes and service use data and enables a wider range of people access to this data. Nesta worked with the council and its partners to ideate, test and prototype an early years data tool through a series of interviews and co-design workshops, building on learnings from the data analysis. The team focused on several key use cases and functionalities for ‘version 1’ of a locally held tool. This included bringing together different and currently siloed data to provide a holistic view of early years outcomes and service use as well as detailed profiles and maps to identify areas of highest need.

Building this early years data tool would be one step closer towards harnessing the power of early years data already held by different parts of Birmingham’s system. This tool is now being built by Birmingham City Council and deployed locally. However, to maximise the value of the tool and ensure its sustainability, a rigorous evaluation of its implementation and further development of the tool with a user-led approach will be crucial.

About the project:

Birmingham City is the third most deprived city in England and has stark levels of educational inequalities and child poverty. It is also the largest local authority in Europe: 15,000 children completed reception year in Birmingham in 2022 (GOV.UK, 2022). Of these, 4,937 children were eligible for free school meals (around 33% of the cohort) and only 55% had a good level of development at age five, as measured using the Early Years Foundation Stage Profile (1) (GOV.UK, 2022). This is substantially lower than the national average for all children, with 67% of all children in England reaching a good level of development in 2023 (GOV.UK, 2022). If the outcomes of children in Birmingham on free school meals were brought in line with the national average, then an additional 576 children would be ‘on track’ at the end of their reception year.

This project explored whether better linkage and use of local administrative data offer insights which support action to address these challenges. We identified where and how locally held data can be better used to build a more detailed understanding of areas in Birmingham that have the poorest early years outcomes and lowest take-up of early years services for children aged 0-5 years old. We also explored whether such analysis might be supported long term by a bespoke early years data tool that can be built in Birmingham’s data systems and process the most recent outcomes and service use data to provide meaningful city-wide insights.

There were two strands of work with two subaims:

• Strand 1: To understand what can be learnt from single snapshots of existing data about areas where children aged 0 to 5 years old are experiencing the poorest early years outcomes and have the lowest take-up of early years services.
• Strand 2: To co-design and produce a ‘design brief’ for an early years data tool that more regularly brings together data about early years outcomes and service use to inform service improvements in the longer term.

A more detailed description of the methodology can be found in the Appendix.

Data has a crucial role to play in the delivery of public services and particularly in how we make more effective decisions in the early years. These decisions could include whether or not a service should be commissioned, re-commissioned or re-designed or whether a family needs additional support that could improve children’s outcomes. Data could also be used by caregivers to make evidence-informed decisions about the care they seek for their families, for example by identifying where there is high-quality early years provision available locally.

At present, most local authorities find that their data across the early years is fragmented and incomplete. This is partly due to the design of the early years ecosystem – most children and families will engage with many services and organisations before a child even enters primary school. These organisations collect their own data and often store it independently, despite the fact that they are often seeing and working with the same families. This leads to an incomplete picture of a family and child’s needs, making it difficult for service planners and frontline professionals to provide integrated support that takes a holistic, family- and child-centred approach. This also makes it difficult for councils to share their data with families and caregivers.

Building a data tool to address these challenges presents an opportunity for councils to do more with the same resources and increase data accessibility for caregivers. The power of data tools could be better harnessed in the early years in order to join up public and private data safely and securely whilst increasing the accessibility of data. For example, in England, Policy in Practice have created the Multi Agency Safeguarding Tracker (MAST). MAST is a data tool that joins up NHS, social care, fire and rescue, and police data in order to enable professionals to make more informed decisions around safeguarding society’s most vulnerable. Yet, such data-driven measures are rarely taken in the early years. This is despite the evidence that educational inequality before the age of five has profound and long-term impacts on a child’s health, well-being, job prospects and life chances.

Data tools could ensure data is better utilised to further our understanding of early years outcomes and service use, and increase data accessibility. For example, algorithms could be developed that enable the linkage of data from different systems in real-time or tools could be built that help local authorities and families visualise the geographical distributions of early years outcomes, provision and service use across an area. Data tools in the early years could also support local authorities to understand trends in outcomes and service use over time or compare their regional performance with their regional and statistical neighbours. This knowledge could then be used to make better informed decisions about services to improve outcomes and reduce educational inequalities. Data tools in the early years could also empower caregivers to make more evidence-informed decisions about the early years provision they seek for their families.

Nesta worked with Birmingham to link data on early years outcomes, take-up of early years services, location and quality of early years settings and other publicly and locally held data sources. This was the first time these have been brought together in Birmingham, linked at an area level and analysed concurrently.

Our analysis focused on five key questions:

• Where in Birmingham are outcomes the poorest and what are the characteristics of these areas?
• How does the size, range, location and quality of support offered through local early years services vary across Birmingham, and how does it compare with early years outcomes in each area?
• What are the rates of take-up for early education entitlement offers for areas across Birmingham and does take-up differ between areas with differences in outcomes?
• Given answers to the above, which areas are most in need of additional support?

By linking this data together and analysing it for the first time, the project highlighted the benefits of creating a tool which can do this in real-time and be integrated into Birmingham’s systems. Our analysis enabled a more accurate and complete picture of need in the early years and produced novel insights. These insights have the potential to improve how council resources are directed to offer additional support for families who need it the most.

Insight 1: The areas most in need of additional support were previously overlooked

We analysed the data to identify six areas most in need of additional support due to outcomes and take-up of early education entitlement being particularly poor. We were then able to compare this list with areas typically considered to be most in need by the various services participating in the project. This comparison highlighted that, when judged by a combination of performance on outcomes and take-up of early education entitlement, the most in need areas were not usually those which were prioritised by services in a position to provide additional support. This support sometimes included sending additional letters and information about early education entitlements to eligible parents and families (such as to groups of families in temporary accommodation), or increased community outreach activities to encourage take-up of early education entitlement in a local area (for example, in supermarkets).

The six areas (2) identified by our analysis were Sparkbrook and Balsall Heath East, Alum Rock, North Edgbaston, Glebe Farm and Tile Cross, Allens Cross and Gravelly Hill. Conversely, existing activities to drive up engagement with services and improve outcomes have been predominantly focused on Perry Barr. While this is a deprived area, it fared better on outcomes and take-up of early education entitlement in 2022 and the spring term of 2023, respectively (3).

This finding illustrates the value of joining up and better utilising this data to improve understanding of the early years landscape in a local area. This analysis and novel use of early years data has shed new and interesting light on where outcomes and take-up of services are poorest in Birmingham.

Insight 2: The most in need areas may require different solutions

By combining locally held early years data with 2021 census data on topics such as deprivation, language, religion and ethnicity, our analysis found that these six areas were not homogenous in terms of their sociodemographic characteristics. This has important implications for how solutions and services are designed in response to findings.

For example, Sparkbrook and Balsall Heath East – an area near Birmingham City Centre with the highest number of children not reaching expected outcomes at age five in 2022 – had a relatively high proportion of people identifying as Asian and Muslim when compared to the rest of Birmingham, according to the 2021 census. In contrast, Glebe Farm and Tile Cross – an area with the highest number of assessed children not at or above expected levels of development at age two in 2022 – had a relatively higher proportion of people identifying as white and Christian, compared to the rest of Birmingham. This suggests that efforts or activities to improve outcomes and take-up of services might need to be tailored to these areas, possibly through resources and advertisement materials in languages other than English or by working with community groups to further understand barriers to engagement with services. These insights would not have been possible without joining up and better utilising local data in the early years.

To build on the data analysis conducted as part of this project, Nesta and key stakeholders from across Birmingham and its partners worked to ideate, test and prototype an early years data tool in three co-design workshops. The most developed prototype for the tool was an interactive data report created using the previously analysed static data. This report allows users to explore outcomes, service use and other related datasets in more detail, deep dive into areas or data sets of particular interest and draw their own conclusions from the data.

The interactive data report also showcases the art of the possible. It acts as a proof of concept for bringing together early years data to better understand and support service delivery. When shared alongside other prototypes in the co-design workshops, it led the team to decide to focus on four key use cases for version 1 of a locally held early years data tool. These include:

• providing a holistic view of early years services and childcare activity by bringing together different and currently siloed outcomes and service use data held locally
• identifying areas of highest need and their characteristics by providing detailed profiles and maps of areas across the city
• supporting integrated working across early years services by capturing and presenting meta-data (including contact details for a team responsible for that data) to catalyse conversations as a precursor to taking meaningful action in a targeted, evidence-led way
• empowering caregivers to make their own decisions about childcare by linking the tool to a public facing platform (that sits on Birmingham Council’s City Observatory). In the co-design workshops Birmingham City Council and its partners felt this was an important additional use case in order to increase accessibility of early years data for families and members of the public.

One of the key limitations of the data analysis we conducted is that it provides a static and limited snapshot of where early years outcomes and service use are poorest. This means that our findings might quickly become out of date and our analysis would need to be conducted again with each refresh of data to provide the most up-to-date picture of the early years in Birmingham.

A data tool that enabled the above use cases could also overcome this limitation by having two key features. First, it could link the most up-to-date early years data and present a ‘real-time’ view of outcomes and service use across the city. Second, it could be embedded within Birmingham’s data systems to speed up and automate data linkage processes and provide the council with greater ownership of any data processing as well as functionalities and features of the tool.

This data tool would give the council and its early years health and care partners greater autonomy to sustainably draw insights from their own data and therefore would add huge value to the existing early years ecosystem in Birmingham. For example, by better understanding where take-up of funded childcare hours and outcomes are particularly low, this could kick-start conversations between the council, service providers and local families that shed important light on ways to tackle barriers to engagement with services. In addition, knowledge gleaned from this tool could be used to ensure those most in need are targeted with information about the availability and location of funded childcare and to encourage the take-up of childcare and other early years services. This is important as engagement with high-quality childcare and early years services is known to improve children’s short- and long-term educational, cognitive, social and behavioural outcomes.

The insights obtained from our interviews and co-design workshops were used to create a design brief for version 1 of the tool that accounts for the range of desired use cases and functionalities. It also sets out the processes agreed by the team that should be followed for the tool’s development. These processes fall into three broad groupings, with different members of the team responsible for specific steps:

• obtaining the necessary source data in a suitable format on a regular basis
• collating and managing the data in a central place
• developing and managing the tool at the user-facing end

This work demonstrated that building this early years data tool would be a step towards maximising the value of early years data already held by different parts of Birmingham’s system.

It is important to note that data tools that bring together currently fragmented early years data sets at an area level are a starting point for councils and organisations to improve their use of data to support children and families. But these types of data tools are not silver bullets. Early years data is often incomplete, of poor quality or missing crucial information on families’ views or holistic needs. Without early years data linked across settings and services at the level of the individual, we won’t be able to answer questions such as ‘Are children in an area that are not attending childcare the same children with the lowest outcomes?’ Without any representation of children and family voices in the analysis of locally held early years data, questions like ‘How do families feel about local provision, its availability, quality and suitability for their children?’ and, ‘Where do families think services are not meeting needs?’ will remain partially or wholly unanswered.

Some caution should therefore be taken when using aggregated early years data and data tools built using this data in the absence of other sources of information.

There is also a need for constant testing and iteration when knowledge from early years data is used to inform change. Has this knowledge actually improved what we want it to improve? Can the tool be further developed to include additional or more up-to-date data that provides another piece of the puzzle? Such questions should be asked at all stages of the development of a data tool: in the first co-design workshops, in the development phase, after deployment and during subsequent developments.

Decisions made in the early years should therefore be data-informed not data-driven.

What do our findings and reflections mean for other local authorities?

Early years data tools, such as the one being developed as a result of this project, present huge opportunities for other local authorities.

Some areas are already on this journey. In Scotland, The Promise is Doing Data Differently – thinking of new ways to join up data in order to provide a more complete picture of children’s journeys and experiences through care services. Whilst they are working with data from central government, care providers, charities and other key organisations in the system, local authorities are particularly placed to better use their data to support care experienced children and their families. For example, South Ayrshire council are using learnings from work with The Promise to refine and improve their data sets, work more collaboratively across services and ultimately provide more holistic, child-centred support to children and young people throughout South Ayrshire.

If councils and local providers took a similar approach to their early years data, and shared and joined up their data via early years data tools, they too could harness the power of their locally held data to better support the most vulnerable early years children and their families. Whether there is a need to better understand need, outcomes or contexts in which families live, data is a crucial and under-utilised starting point.

One of the planned next steps for Nesta’s work with Birmingham City Council will involve building and deploying the co-designed early years data tool. Birmingham City Council are leading on this work and plan for the tool to sit on Birmingham City Council’s City Observatory so that the tool is available to the public as well as professionals.

To maximise the value of the tool and build an evidence base for its impact, Nesta recommend that Birmingham City Council and its early years partners:

• identify the most important specific use cases for professionals in Birmingham and ensure these use cases are at the forefront of development and deployment decisions. These use cases should be linked to key decisions taken regularly in the early years
• run a series of meetings and webinars to engage different professionals across Birmingham and members of the public around the tool and encourage its use for decision making
• implement an evaluation alongside deployment of the tool to ascertain its impact and link this to the identified key use cases and decisions
• embed the development and use of the tool into ‘business as usual’
• continue to adapt and further develop the tool, and follow and adopt a user-led, prototyping approach when doing this to ensure the tool remains useful and relevant for its users

This project used a mixed-method approach to explore how Birmingham City Council and its partner organisations could better utilise early years data to improve understanding of families in Birmingham and provide more tailored support.

For Strand 1 of this project we brought together a range of data sources in order to better understand areas with the poorest early years outcomes and take-up of EEE. These datasets included:

• data on outcomes at age two, measured using the Ages and Stages Questionnaire (ASQ-3) and obtained from Birmingham Forward Steps (Birmingham City Council’s health and care partner) for 2022, 2021 and 2020
• data on outcomes at age five, measured using the Early Years Foundation Stage Profile (EYFSP) and obtained from Birmingham City Council's Data and Intelligence Team for 2022
• data on take-up of early education entitlement (EEE) at ages two, three and four years old obtained from Birmingham City Council's Early Years team for the spring term of 2023
• data on the location, quality and operating hours of local early years services and settings (where available) obtained from Birmingham City Council's Early Years team and Birmingham Forward Steps
• various other locally held and publicly available data sources to better understand areas, including data on the number of children in temporary accommodation, ethnicity and deprivation data

We selected six wards that we identified as having the poorest early years outcomes and lowest take-up of early education entitlement.

For Strand 2, we conducted 15 interviews with key stakeholders across Birmingham City Council and its partner organisations to understand the early years data currently available, its limitations, desired functionalities for a data tool and technical parameters that might influence its functionalities, development and deployment. We then conducted three co-design workshops to ideate, test and prototype an early years data tool that accounts for needs and technical parameters captured in interviews as well as key learnings from Strand 1. We then created a design brief for the tool that brought together insights and views from across the interviews and co-design workshops and agreed a process with Birmingham City Council for how they could use this brief to develop an early years data tool.

We also developed an interactive data report that showcased the analysis conducted in Strand 1 and acted as a prototype for the co-design workshops in Strand 2.

Perspectives of key stakeholder groups across Birmingham’s early years infrastructure were obtained in the interviews and co-design workshops. These stakeholders included:

• service planners, information managers and analysts in the Early Years team in Birmingham City Council
• contract managers and information and data analysts from Birmingham Forward Steps, Birmingham City Council’s health and care partner with responsibilities for the delivery of health visiting, children’s centres and collecting the ASQ-3
• leads in the Insights, Policy and Strategy team in Birmingham City Council
• leads in the Special Educational Needs Coordinator (SENCo) team in Birmingham City Council
• leads of organisations that support state-funded, private, voluntary and independent early years service providers across Birmingham

Throughout the design process we also considered requirements of four key user groups across Birmingham:

• early years service staff and management in Birmingham City Council
• commissioned partners of Birmingham City Council
• current and potential providers of early years provision
• parents, carers and families

Footnotes:
1. The Early Years Foundation Stage Profile (EYFSP) is a statutory assessment of children’s development at the end of the early years foundation stage (which takes place in reception at age 5). It is made up of an assessment of a child’s outcomes in relation to seven areas of learning: communication and language, personal, social and emotional development, physical development, literacy, mathematics, understanding the world and expressive arts and design.
2. In this analysis, areas were defined as wards. In England wards are electoral districts at sub-national level that are represented by one or more councillors. On average, wards are home to 5,500 people but ward population counts can vary substantially. Birmingham has 69 wards, accurate as of January 2024.
3. Perry Barr may be performing better on outcomes and take-up of early education entitlement due to recent concerted efforts by Birmingham to improve take-up of early education entitlement. However, due to a lack of retrospective data on take-up of early education entitlement, we cannot conclusively say that this is the reason for any differences seen in take-up between areas.