Sovereign Network Group was Highly Commended in the Innovation category of the Housing Technology Awards 2026.
When you commit to investing £100 million in communities over the next decade, you inevitably stop and ask yourself: how do we make sure this really changes lives? At SNG, that ambition comes with an equally strong sense of responsibility. The community team knew they didn’t just want to spend wisely, they wanted to understand their neighbourhoods deeply enough to invest meaningfully.
Understanding community needs
Anyone working in housing knows that understanding community need is never simple. National datasets can point you in the right direction but they don’t capture the nuances, such as the difference between one street and the next, how a single service closure affects an estate or the quiet pressures only visible to those who live or work there every day. Staff across SNG held that insight but turning lived experience into strategic, scalable evidence wasn’t easy.
This led to a deceptively simple but vital question: how do we see our communities clearly enough to help them fairly? The answer became our Community Indicator Model.
From the beginning, the aim was to bring humanity and data together. SNG needed more than dashboards; it needed a single, shared picture built from national datasets, internal operational knowledge, colleagues’ on‑the‑ground expertise and a long-term understanding of residents’ realities.
The result is a tool which shows how communities are experiencing financial stress, health inequalities, limited access to services, digital exclusion, safety concerns and more, all updated with real-time responsiveness. Our staff often describe using the model as finally being able to articulate what they’ve sensed for years. It’s the blend of “we knew this” and “now everyone can see it” that makes the tool so powerful.
Finely-focused information
The model acts like a lens: where SNG once held many fragmented pieces of information, it now has a clear, joined-up view. Our teams can see where needs cluster, where patterns emerge and where early interventions will matter most.
Behind that clarity is a substantial technical engine and a deeply collaborative build. Working with Intuita Consulting, SNG developed the model using Microsoft Azure, Azure ML Studio, GIS and Power BI, with its foundations shaped by our colleagues across the business. Workshops identified which indicators genuinely mattered in practice, while subject-matter experts refined weightings so that the model reflected lived experience as well as data. Everything was documented to guarantee transparency and flexibility for the future.
Technically, the model was built through a multi-stage analytics pipeline designed to integrate large national and internal datasets into a unified scoring framework. The process began with extensive data discovery workshops, feature identification and the acquisition of open and internal datasets. These were rigorously cleaned and matched across multiple geographies (property, LSOA, MSOA & local authority) to ensure perfect alignment. Once harmonised, indicators were transformed onto a common analytical scale with standardised directionality so higher values always reflected better outcomes.
Statistical rigour
Correlation analysis and clustering on the absolute correlation matrix grouped indicators into statistically-coherent structural themes that remained consistent across geographies. Weighting followed a hybrid method – external benchmarks, such as the Cambridge Crime Harm Index, informed weightings where appropriate, while factor analysis generated data‑driven weights for areas where benchmarks didn’t exist. To ensure comparability across hundreds of features, scores were normalised using z‑score techniques and mapped onto a bounded 10–90 scale. Finally, national‑level features and SNG’s operational data were merged through a configurable weighting scheme, producing a flexible, transparent and reproducible model capable of handling hundreds of indicators across varying spatial granularities.
Even though the tool is still relatively new, its impact is already tangible. For example, our community investment leads use it as the starting point for their project planning, and colleagues preparing funding bids use it to strengthen their cases (and many have already seen greater success). Our frontline teams say it boosts their confidence because it supports their instincts with clear evidence, and when SNG meets local charities, councils or health partners, the model provides an instant, shared foundation for conversations.
Non-data evangelists
Some of the most enthusiastic feedback has come from people who never saw themselves as ‘data people’. They describe the tool as intuitive, approachable and even enjoyable to use. That accessibility is helping embed a cultural shift that SNG has long aimed for: decision‑making that is both compassionate and evidence‑based, human and data‑informed.
Most importantly, the model is already improving outcomes for residents. With a clearer understanding of needs, we can direct our investments more fairly and effectively. Communities facing multiple pressures aren’t overlooked and emerging problems are identified sooner. Instead of reacting to crises, we can act proactively, intentionally and transparently.
It’s no exaggeration to say the Community Indicator Model is becoming the backbone of SNG’s long-term community strategy. And while it is a sophisticated technical achievement, it also represents our commitment to listening deeply to communities and responding with care.
Wider applicability
The wider housing sector has taken notice. Many organisations use national datasets and many have internal dashboards, but very few have brought these elements together in a model as flexible, locally precise and transparent as this. Already, it is being viewed as a blueprint by others – a sign that our approach could inspire broader sector change.
Looking ahead, SNG will use the model to guide its £100 million investment, tracking not just spending but impact: wellbeing, financial stability, access to services and strengthened trust between partners. The early signs suggest the model will exceed expectations, not only in outcomes but also in how it brings our teams together around a shared purpose.
At its heart, the Community Indicator Model isn’t a data project. It’s a fairness project, a clarity project and, more than anything else, a people project. It helps us to see communities not as statistics but as living, evolving places and to act with precision, empathy and confidence.
For SNG, the model isn’t the destination, it’s the beginning of a more responsive, thoughtful way of supporting communities. And for the people living in those communities, it promises something powerful: investment that sees them, understands them and truly meets their needs.
Claire Hyland is the data, analytics and AI director at Sovereign Network Group. The housing provider was Highly Commended in the Innovation category of the Housing Technology Awards 2026.
