In February 2026, the Government published the Autumn 2025 Rough Sleeping Snapshot, recording 4,793 people sleeping rough in England on a single night, a 3% increase on the previous year and the highest figure since comparable data collection began.

While rough sleeping in London saw a slight year-on-year decrease, 43% of those counted were in London and the South East, underlining the continued concentration of housing pressure in and around the capital. Alongside this, separate statutory homelessness data continues to show record numbers of households in temporary accommodation across England, with more than 100,000 households currently placed in TA, and the highest number on record of children growing up in temporary accommodation. 

These figures matter not only because of what they represent in human terms, but because of what they reveal about the structure of our homelessness system.

Through Greater Change’s delivery across London boroughs and beyond, we see the dynamics that sit beneath these headline numbers.

When Small Barriers Become Street Homelessness

Amelia*, a woman of pensionable age, was living in her car for a prolonged period because she could not afford a rental deposit. That someone in their fifties or sixties can spend their retirement years in this position should give pause to anyone who believes our safety net is functioning as intended.

When her support worker identified a rare and affordable private rental property, it presented a genuine route out of homelessness. However, the landlord required a deposit and first month’s rent immediately. Having previously experienced delays in Housing Benefit payments, the landlord was unwilling to proceed without financial protection upfront. Delays within the welfare system had effectively shifted financial risk onto the next tenant.

Amelia was therefore required to produce thousands of pounds immediately or lose the property. Without rapid intervention, she would have remained homeless, not because housing was unavailable, but because the system could not move quickly enough.

Greater Change provided £1,500 within 24 hours. A modest, flexible intervention prevented prolonged rough sleeping.

Her case is not an anomaly. It illustrates a structural vulnerability in how prevention is currently delivered.

A System Oriented Around Thresholds

The February 2026 snapshot should prompt reflection on system design.

Over the past two decades, homelessness policy has become increasingly statutory and threshold-based. Access to meaningful support is often unlocked only once individuals demonstrate that their situation meets specific legal criteria under the Homelessness Reduction Act framework.

These safeguards are important. However, in practice they can mean that assistance is triggered only once a crisis has escalated.

The consequence is that opportunities for early intervention are frequently missed. Small financial gaps, such as deposits, arrears, or essential goods, are not resolved quickly enough. Situations that could have been stabilised become more complex and entrenched. By the time support is provided, the human and financial costs are significantly higher.

Rough sleeping is among the most traumatic and expensive outcomes the system can produce. Estimates suggest that one person sleeping rough can generate over £20,000 per year in additional public costs through emergency healthcare, temporary accommodation, policing and other crisis services.

Yet the financial barriers that precipitate homelessness are often comparatively modest. In many cases, they fall within the range of £500–£2,000.

In Amelia’s case, £1,500 prevented a trajectory that could have cost tens of thousands annually.

What the Snapshot Tells Us

The 2026 publication does not simply present a number. It provides a lens through which to examine whether prevention is functioning effectively.

If rough sleeping remains at record levels despite significant public expenditure on homelessness services, it suggests that resources are disproportionately concentrated at the point of crisis rather than upstream.

Greater Change’s personalised budget model is one attempt to address this imbalance. Working with trusted frontline charities and local authorities, we provide targeted, client-designed grants to remove specific financial barriers to housing stability.

Our outcomes show that 86% of those supported sustain or move into stable accommodation 12 months later. Among individuals with previous criminal justice involvement, 92.8% had not reoffended after one year. The average cost of intervention is approximately £1,400 per person, compared with estimated public costs of over £41,000 per year associated with homelessness.

These findings are currently being examined further through a government-commissioned Randomised Controlled Trial delivered in partnership with the Centre for Homelessness Impact and King’s College London. However, the core economic principle is clear: earlier intervention reduces both human harm and fiscal cost.

From Measurement to Redesign

The February 2026 snapshot should not be treated solely as a performance indicator. It should be understood as feedback on system architecture.

Street homelessness is rarely the result of a single event. It is typically the culmination of delayed intervention, limited financial resilience, and insufficiently flexible prevention tools.

If we continue to operate a system that waits for a visible crisis before acting, rough sleeping figures will remain resistant to sustained reduction, regardless of the scale of downstream spending.

Embedding personalised budgets within mainstream homelessness prevention pathways, and equipping frontline workers with the discretion and resources to act quickly, represents one practical step toward redesigning the system around prevention.

The central question raised by the February 2026 snapshot is not simply how many people were sleeping rough on one night. It is how many instances of homelessness could have been prevented earlier, and at far lower cost, had systems been designed for speed, flexibility and trust.

‍