What if preventing homelessness started earlier, in the community, before families ever needed to enter the formal homelessness support system? At the Centre for Homelessness Impact, we are working with the Greater London Authority, London Councils and four London boroughs to test whether embedding housing support in Family Hubs can help us do just that. By bringing housing expertise into services families already use and trust, we hope to learn whether we can spot housing instability earlier, connect families to the right support sooner, and ultimately prevent more families from experiencing homelessness.

England has a relatively progressive statutory framework for homelessness prevention compared to the rest of Europe. Since the introduction of the Homelessness Reduction Act 2017, local authorities have had a legal duty to support eligible households who are threatened with homelessness within 56 days. In practice, this means councils do not have to wait until a household has lost their home to intervene. There is, at least in principle, a window in which homelessness can be prevented rather than responded to. But the statutory 56-day threshold rests on two assumptions that are worth reconsidering if we are truly committed to upstream prevention: that people can identify when they are 56 days away from becoming homeless and that this period allows enough time to intervene.

For some routes into homelessness, there are relatively clear markers of homelessness risk. The Homelessness Code of Conduct for Local Authorities, Chapter 6, sets out circumstances in which someone who has received a valid notice requiring them to leave their home (i.e. an eviction notice) should be considered threatened with homelessness. An eviction notice provides something concrete: a date, a legal process and, crucially, an identifiable point at which the homelessness system can intervene.

Family life rarely comes with the same paperwork

If you are living with your parents, a sibling or a friend, there is unlikely to be a single moment when your living arrangement moves neatly from stable to unstable. Maybe the house is overcrowded. Money might become an issue. Arguments could happen more frequently. Someone may start saying that you need to move out. None of these things necessarily mean you will be experiencing homelessness within 56 days. However, by the time someone finally tells you that you need to leave, the opportunity to prevent the breakdown or support a planned move into alternative accommodation may already have passed.

When we look at the latest Ministry of Housing, Communities and Local Government (MHCLG) statutory homelessness data, there is a striking difference between the reasons families are identified as at risk of homelessness and the reasons they actually experience homelessness. 

In London, between January and March 2026:

• Among households threatened with homelessness and owed a prevention duty (6,670 households): 53% were at risk because of the loss of a private rented tenancy, while 22% were at risk because family or friends were no longer willing or able to accommodate them.
• Among households already experiencing homelessness and owed a relief duty (9,840 households): the pattern was almost reversed. Family or friends no longer willing or able to accommodate them was the most common reason, accounting for 29%, while the end of a private rented tenancy accounted for around 15%,  similar to the proportion of households where domestic abuse was the primary cause of homelessness.

There are several possible explanations for this difference, and the data cannot tell us for certain which is correct. It may, in part, reflect successful prevention. The loss of a private rented tenancy is often easier to identify in advance, giving local authorities an opportunity to intervene before homelessness occurs. If those interventions work, we would expect households losing a private tenancy to make up a smaller proportion of those who subsequently reach the relief stage. But another possibility is that some routes into homelessness are simply less visible to the prevention system in the first place. The breakdown of an arrangement with family or friends rarely comes with the same clear warning. A family may be experiencing increasing housing instability without recognising themselves as at risk of homelessness, much less knowing when they have crossed an invisible 56-day threshold. Yet signs of that instability may be visible elsewhere: financial stress, debt, overcrowding, relationship difficulties, frequent moves, or increasing reliance on other forms of family support. The issue, then, may not be an absence of warning signs. It may be where we are looking for them.

Where Family Hubs come in

Family Hubs are, at their core, community-based spaces. Building on early-help infrastructure such as Sure Start Children’s Centres, they bring together support for families with children and young people aged 0 to 19, or up to 25 for young people with Special Educational Needs and Disabilities (SEND). The exact model looks different across local authorities, but their value lies in their proximity to the families they serve.

Unlike homelessness services, Family Hubs do not only encounter families at the point when something has gone seriously wrong. Staff may see the same families repeatedly, sometimes across different services and at different points in their lives. They may know that a family is struggling financially, that a living arrangement is overcrowded, or that relationships at home have become increasingly difficult. None of these things necessarily means that a family is close to homelessness, but together they can tell us something about whether a housing situation is becoming less stable.

Family Hubs are not designed to be housing or homelessness services, and nor should they be. But being embedded within communities means they may be seeing signals of housing instability long before those signs become visible to the homelessness system. If the challenge is not that the warning signs are absent, but that we are looking for them too late, then perhaps we need to start looking in the places where families already are.

What we’re testing

In London boroughs of Camden, Brent, Westminster, and Kensington and Chelsea, we're testing what happens when housing advice and support is embedded directly within Family Hubs.

Family Hub staff and partners will be trained to recognise some of the early signs of housing instability, understand what housing support is available locally, and know when to bring in the embedded housing support worker. Rather than simply signposting families elsewhere, the model is built around active referrals, with families being connected directly by Family Hub staff and partners to the embedded worker who can help them understand their options.

That support will look different depending on what a family needs. It could mean helping someone manage rent arrears or maximise their income, supporting them to find private rented accommodation, explaining their tenancy rights, or negotiating with a landlord to prevent an eviction. For families staying with relatives or friends, it could mean mediation to try to prevent that arrangement from breaking down. And where a family needs statutory or more specialist support, the housing support worker can help connect them to it.

Through the programme and its evaluation, we want to understand whether this approach leads to families who are at risk of homelessness being identified earlier, improves their access to housing advice and other preventative support, and ultimately prevents families at risk of homelessness from crossing the line into crisis and homelessness. Just as importantly, we want to understand what it takes to make this approach work in practice, and whether embedding housing support within Family Hubs is an approach that could be adopted more widely.

Prevention before the 56 Days

None of this is an argument against the 56-day prevention duty. It is an argument for looking at what happens before it. Family Hubs give us an opportunity to test whether we can intervene earlier, by putting housing expertise in places where families already are and where those warning signs may already be visible. If we want to prevent more families from becoming homeless, the 56 days before they lose their home cannot be the only part of their journey that we look at.

• Emily-Rae Foreman is Data and Evaluation Lead at the Centre for Homelessness Impact 
• The image used in this blog is from CHI's free library of non-stigmatising images of people experiencing homelessness.

‍