Debt Pilot | Portfolio

Using data to intervene earlier in tax debt recovery

Could preventative, data-led support improve outcomes for customers while increasing council revenue?

Screenshot 2024-07-29 at 8.34_edited.png

Overview

Council tax debt recovery is typically reactive, relying on reminders and enforcement once arrears have already accumulated. This approach often leads to low engagement, escalation into crisis, and poor outcomes for both customers and organisations.

I led the design and delivery of a proactive council tax debt support service that used data to identify customers at risk of falling into arrears and offer support before enforcement action was triggered. The aim was to shift the system upstream - improving early engagement, supporting customers to stabilise their finances, and increasing revenue recovery while reducing reliance on court and enforcement processes.

The service was piloted and evaluated through a randomised control trial (RCT) and demonstrated clear impact within four months.

Pilot impact (4 months):

26% customer engagement with proactive outreach
~£400k in additional benefits secured for customers
~£75k more council tax repaid compared to the control group
Reduced progression to court and enforcement action

My role

Service Design Lead

I led the work end to end, working within a multidisciplinary team that included data scientists, operational leads, behavioural specialists, and frontline officers.

My responsibilities included:

framing the problem and defining success measures
leading service design and facilitation
leading qualitative research and testing outreach models
shaping the data-led approach to identifying risk
designing the end-to-end service and operational workflows
contributing to the RCT design and interpretation of results
translating findings into recommendations for scale and business-as-usual delivery

The problem

The council’s existing council tax recovery model relied on reactive recovery after arrears had built up with enforcement as the primary escalation mechanism, and limited understanding of customers’ circumstances.

This resulted in:

low engagement and avoidance by customers
escalation into legal action and crisis
missed opportunities to support people earlier
high downstream costs

The council needed a way to identify financial risk earlier, intervene proportionately, and do so in a way that was ethical, transparent, and operationally viable.

Scoping workshop with key stakeholders

Theory of change

We believed that combining a data-led approach to identifying emerging financial risk with proactive outreach and support — rather than reactive reminders and enforcement — would fundamentally change outcomes. By making risk visible earlier and acting on it, the council could reach customers who would otherwise remain unseen and offer support while meaningful options were still available.

Discovery and framing

I facilitated workshops with frontline officers and managers to map the council tax journey, identify points of disengagement, and clarify regulatory and operational constraints.

Working with data colleagues, we matched council-wide datasets to define indicators of emerging financial risk, identify customers entering arrears in near real time, and segment them into different support pathways.

In parallel, I led qualitative research through interviews and focus groups with residents to understand attitudes to proactive contact, concerns around data use and consent, and what made outreach feel supportive rather than punitive. These insights shaped decisions around trusted channels, transparent messaging, and opt-in support.

Designing the intervention

Using storyboards and service prototypes, we tested different outreach and support models with customers and staff, including tone and framing, timing and channel of contact, consent models, and how customers were routed into appropriate support.

We designed a full end-to-end service covering:

data-driven cohort identification
multi-channel proactive outreach
relational, non-enforcement-led engagement
case management and follow-up
referrals into specialist debt and welfare support

The service was deliberately designed to operate alongside existing statutory recovery processes, not replace them.

Impact

Evaluation (Randomised Control Trial)

The pilot was evaluated through a randomised control trial, comparing outcomes for customers receiving proactive support against a control group receiving business-as-usual communications.

Over a four-month period, the intervention demonstrated:

26% engagement from proactively contacted customers
~£400k in additional benefits secured for customers who had previously been under-claiming
~£75k more council tax repaid than the control group
reduced progression to court and enforcement action

A significant proportion of engaged customers were not previously known to support services, showing that data-led identification surfaced previously hidden vulnerability.

“Client explained that this has lifted a weight off of her shoulders. She said that the support she has had has really helped as she was suffering with bad depression and was afraid to open any letters from the council but has now opened them all and has been trying to deal with them.“

- Frontline Officer

Direction

The pilot demonstrated that preventative, data-led intervention can materially improve outcomes for both customers and the organisation. Following the pilot, work began to integrate proactive debt support into business-as-usual operations.

Learnings

Data-led services require trust by design. Data ethics principles of transparency, consent, and clear communication were critical to engagement, particularly when working with vulnerable groups.
Rigorous evaluation enables confidence to change. Using a randomised control trial allowed innovative ways of working to be tested at low cost and small scale, while generating credible evidence. This helped build senior buy-in and influence decision-making.
Service design can bridge data, policy, and lived experience to deliver measurable change. Bringing together analytics, regulatory constraints, and frontline and customer insight made it possible to design an intervention that was ethical, operationally viable, and demonstrably effective.