Building a collaborative R and Python analytics community in Suffolk

The Suffolk Office of Data and Analytics (SODA), a collaboration between Suffolk’s Local Authorities, Suffolk Constabulary and local NHS Organisations, has developed a collaborative, analyst-led community to support the use of coding tools such as R and Python across local government and partner organisations. The community provides a practical forum for analysts to share knowledge, troubleshoot issues, and improve the quality, efficiency and reproducibility of data analysis. By working closely with IT colleagues and taking a gradual, relationship-based approach, Suffolk has enabled secure use of advanced analytical tools while reducing duplication of effort and analyst isolation.

Like many councils, Suffolk County Council had pockets of analytical expertise using coding tools such as R, often concentrated in individual teams or relying upon a small number of highly skilled individuals. While this had enabled sophisticated analysis, it also created risks. Knowledge was siloed, support was informal and when individuals moved roles or left the organisation, capability gaps emerged.

At the same time, analysts across Suffolk County Council, district councils, and partners such as the Integrated Care System were increasingly looking to move beyond traditional business intelligence tools alone and towards ‘analysis as code’. This analysis as code approach allows analytical methods to be documented, reused, quality assured and updated more easily.

Support for these tools was variable. R and Python were not core systems owned by IT teams, and security, access to packages, and system configuration often relied on analysts raising ad hoc requests. Analysts frequently found themselves solving the same technical problems independently, leading to inefficiency and frustration.

Recognising both the opportunity and the risk, analysts within SODA took a bottom-up approach to building a community of practice that could support safe, effective and sustainable use of R and Python across organisations.

R and Python are open-source programming languages widely used for data analysis, statistics and automation.

• R is particularly strong for statistical analysis, data manipulation and producing repeatable analytical outputs. It is commonly used in public health and research, and has a large library of specialist packages for analysing population and outcomes data.
• Python is a more general-purpose programming language with powerful data analysis capabilities. It is widely used for automation, data processing and increasingly for machine learning and AI applications.

In simple terms, R and Python allow analysts to write code that:

• cleans and prepares data in consistent way
• applies analytical methods transparently
• produces outputs that can be rerun automatically when new data is available.

This contrasts with more manual tools where analytical steps can be harder to inspect, reproduce or quality assure.

The Suffolk approach has been deliberately light-touch and collaborative, rather than creating a formal new team or governance structure.

The initial step was the creation of a Microsoft Teams channel open to analysts across Suffolk County Council, district councils and partner organisations who were interested in coding, particularly in R. This reflected how analysts already communicated day-to-day and avoided the need for new platforms or complex setup.

The purpose of the community was simple:

• share solutions to common technical problems
• reduce duplication of effort
• create a space where analysts could ask questions without feeling isolated.

Rather than positioning the community as expert-led, it was framed as a peer support network. Analysts at different skill levels contribute, and no expectation is set that individuals must be experts. This has helped broaden participation and surface ‘hidden’ expertise across organisations.

Close working with IT has been a critical enabler. While IT teams retain responsibility for security, infrastructure and access controls, the community helps bridge the knowledge gap around how analytical tools are actually used in practice. A named IT contact provides continuity, and more senior IT colleagues are brought in when complex questions arise.

The main impact of the community and use of R and Python has been improved efficiency, quality and resilience of analytical work.

Analysts report that having a shared forum means issues are resolved faster, particularly where problems have already been encountered by others. This reduces time spent troubleshooting individually and helps analysts focus on value-adding analysis.

The approach supports reproducible analysis, where code can be shared, reviewed and reused. This makes it easier to:

• update analysis when new data becomes available
• replicate good practice seen in other organisations
• quality assure methods before results are shared with senior leaders.

The community also tackles analyst isolation. Coding can be a solitary activity, particularly in smaller organisations where there may be only one or two specialists. Having a network of peers across organisations provides reassurance, learning opportunities and informal mentoring.

From an organisational perspective, the approach supports better value for money. Analysis written once and reused avoids repeated manual effort, reduces reliance on re-building dashboards from scratch, and helps organisations respond more quickly to emerging questions.

One of the main challenges has been security and access, particularly in relation to Python. Due to its flexibility and wide range of capabilities, Python was initially seen by IT colleagues as a higher security risk. In response, Suffolk took a balanced approach, including the use of secure virtual machines rather than local installations.

Access to external packages has also required careful handling. Analysts cannot simply download packages directly; instead, requests go through IT security checks before being approved. While this can slow progress, it ensures risks are managed appropriately and builds trust between analysts and IT teams.

Another challenge is capacity. The community is not anyone’s full-time role and relies on goodwill and shared ownership. Maintaining momentum requires ongoing effort to keep the space active and inclusive.

Finally, there is a learning curve. Coding tools take time to learn, and a rapid, organisation-wide rollout would risk overwhelming teams. Suffolk deliberately avoided mandating use of R or Python, instead allowing interest and capability to grow organically.

Suffolk’s experience suggests several lessons for local authorities considering a similar approach:

• get IT involved early, and at a senior enough level to have meaningful conversations about risk and benefit
• acknowledge security concerns openly – R and Python do pose risks, and these need to be managed, not dismissed
• start small with enthusiastic individuals and build momentum over time
• use existing platforms, such as Teams, to reduce barriers to participation
• avoid over-formalising in the early stages, focus on relationships and shared purpose
• promote peer learning, not expert hierarchies.

Over time, Suffolk would like to develop shared code repositories and strengthen reuse of analytical methods across organisations. However, the community itself is seen as the main success, creating a space where analysts can collaborate, share good practice and solve problems together.

By focusing on people, relationships and practical support rather than technology alone, Suffolk has shown how local authorities can safely unlock the benefits of R and Python while building a more resilient analytical capability.

For local authorities interested in learning more contact: [email protected] SODA Advanced Analyst and Researcher