Providing better children‘s services data at lower cost

Since 2020 we have built and supported a community of children‘s services analysts to bring Python data analysis tools to more than 50 local authorities across the UK.

Children’s services analysts have a wealth of data available to them to improve their services, but existing tools, such as Excel, that they have available to utilise this data are limited. 

Whilst useful for some types of analysis – for example, the number of children using a service – more advanced functions, such as demand forecasting or data validation, can prove difficult.

Over the past two years, we have collaborated with Data to Insight, a community of local authority data professionals, to make Python-based analytical tools available to a wider audience.

To do so has meant applying the multidisciplinary skillsets in Social Finance’s Data + Digital Labs Team – including our technical expertise, community-building experience, and user research methodology – to overcome a number of challenges, as detailed below. 

The challenge: How can we deploy a Python-based tool to all 152 local authorities?

These more advanced analytical tools need to be code-based (rather than built in Excel), which has limited the scalability of tools as they require local authorities to either:

1. Share sensitive personal data, requiring lengthy information governance processes; or
2. Install tools / software locally, involving lengthy IT approval processes 

Our solution: Using a software package called Pyodide, we were able to load Python and run code in-browser, but working entirely locally, focusing on standard datasets. This means analysts can access the tool in seconds, like a website, but that no data leaves their machine.

The challenge: How can we ensure ongoing sustainability?

Tools need maintenance and updates (for example when data validation rules are updated, or when bugs are identified). We needed to identify a way to do this which was sustainable within existing local authority funding structures.

Our solution: We upskilled a community of children’s social care data analysts in Python, coached them to be able to code up rules, and built the infrastructure so they can edit and update in future. 

The challenge: How do we ensure the tool is responsive to the needs of all 152 local authorities?

Each local authority will have their data stored in slightly different ways, and have different capabilities in accessing and engaging tools. Additionally, each will have different preferences around the analysis and functionality they would find most valuable.

Our solution: We established a replicable four-tier user research methodology to be used to develop and iterate upon new tools: