Skip to main content

Evaluation aims and methodology

This paper reports on the findings of an evaluation of homeless interventions funded by three national programmes: Phase 2 approach to Homelessness (which was part of the Welsh Government’s response to the COVID-19 pandemic); Housing First; and the Youth Homelessness Innovation Fund. 

The aim of the evaluation was to understand whether, how and in which circumstances interventions funded by these three programmes have supported the implementation of Welsh Government’s homelessness strategy, i.e. prioritising rapid rehousing and prevention/ early intervention; promoting person-centred, trauma-informed, and joined-up working; and being informed by evidence and coproduction.   

A case study design was used, and fieldwork was conducted in a stratified sample of funded projects in 13 diverse settings across Wales. In total, 98 people were involved in interviews and focus groups, of whom 17 had lived experience. Professionals who took part were drawn from funded projects, local authorities, and a range of partner agencies. A review of project documentation, data and local and national policies was carried out, and nine Value for Money case vignettes have been produced.

Main findings

Overall, the evaluation found many examples of promising practice across the three funding programmes, and in different parts of the country. It has also identified examples of blockages, of projects not functioning as expected because other parts of the system are over-stretched, or because projects are not sufficiently embedded in policies, processes, and partnerships.

Rapid rehousing

The supply of suitable and affordable housing is essential to the functioning of a rapid rehousing system, yet this is challenging due to a lack of social housing and affordable private rented properties. Phase 2 funding has enabled some progress in increasing housing supply for those experiencing homelessness, e.g. through acquisitions or initiatives to increase private rented sector access, but the scale is small compared to demand. Barriers to new development included planning, environmental issues, and increasing costs in the building trade. Given the challenges around rehousing people, the term ‘rapid’ often did not feel relevant to interviewees. 

However, Phase 2 funding has enabled local authorities and their partners to develop and test other elements of a rapid rehousing system, e.g., multi-agency triage and assessment hubs, and to fund the transition to new models of housing and support. 

As well as having the right type and amount of provision at each stage of a rehousing system (i.e., prevention, emergency response and resettlement), a key finding is the need to invest in services which can support individuals into the right accommodation, housing-related and multi-agency support, and smooth transitions. Examples of this within funded projects include support workers who ‘stick’ to a person (rather than an accommodation setting) and provide continuity of relationship and advocacy to secure and set up tenancies; and projects where there is sufficient flexibility to step-up and step-down support, as people’s needs change. 

Prevention and early intervention

The focus of most funded projects is on reducing the risk of homelessness being repeated, rather than prevented in the first place; however, some also work further ‘upstream’, e.g. with young people at risk of homelessness, or people in prison. 

Having the flexibility and skills to work holistically with whatever ‘is on top’ for the individual, rather than focusing only on the presenting housing issue, was viewed by staff and service users as the most effective way to sustain prevention outcomes.

Joined-up working

The evaluation identified examples of joint working with health, criminal justice, social care and children’s services at both operational and commissioning levels. Promising examples from the case study sites included: multi-agency triage and assessment, a homelessness multi-disciplinary team, and a specialist homelessness nurse post. However, it takes time to build this level of joint working at project level and statutory services are over-stretched; subsequently some funded projects seemed to be operating in something of a ‘bubble’.

The awareness, appetite, and capacity of projects to challenge conditionality and advocate for the right to housing of those they support within wider housing allocations systems varied. Having direct formalised relationships with housing providers who understand and trust the model, ideally from the outset, seemed to help support providers secure settled tenancies for their service users. 

Enablers of effective partnership working included: partnerships which are built into project design, via Service Level Agreements or joint bids with cross-sectoral leadership and governance; secondments, joint training and co-location between agencies or departments; a balance between new funding to test and ‘springboard’ and ongoing integration (e.g. through Housing Support Grant); and a focus on designing services around the person. 

Person-centred and trauma-informed

The evaluation found examples of people’s needs and views being considered; experiences of relational and responsive support; of being linked to communities and interests to promote independence; and, in some elements of service provision in more embedded services, power being shared between service user and worker. 

Overall, there is a lack of clarity around what a ‘person-centred approach’ involves, and this approach is therefore not always being applied consistently across projects, or the wider pathways of which they form a part. There were examples of laws, funding criteria, policies, procedures, organisational culture, and a lack of resources in wider systems limiting or negating person-centred working by projects. 


The evaluation found pockets of good practice in involving service users in service design (e.g. project name and décor), in staff recruitment, in volunteering, and in contributing to strategic initiatives. 

However, the overall conclusion is that co-production principles are not being consistently applied across funded services and across wider homelessness pathways, despite stated intentions in funding applications and project documents. Barriers include: a lack of clarity around what ‘co-production’ means, why it matters, and what is expected or required, what ‘good’ looks like, and the ethics and practicalities of implementing this. 

Using evidence 

Some projects have developed their own innovative approaches to measurement, and it is positive to note the breadth of non-housing related wellbeing outcomes being captured, albeit by varying methods and measures.

The funding programmes pre-date Welsh Government publication of the Ending Homelessness and the Housing Support Grant outcomes frameworks. Consistent and robust measurement processes are generally not in place across funded projects, with more of a focus on outputs than outcomes and learning. The range of approaches has meant that it is not possible to aggregate performance and impact data at programme level. 

Conclusions and recommendations

Overall, the evaluation identified positive outcomes and impact at individual project level, with examples of service users experiencing a person-centred service, and where positive housing outcomes were achieved, despite widespread reports of a challenging environment where suitable accommodation options are scarce. The most fundamental barrier to projects applying and working toward the Welsh Government national vision was the lack of join up at a local system level. 

The Value for Money assessment identified potential for significant cost avoidance for health, criminal justice, adult social care and children’s services, as well as local authority homelessness services from a rehousing approach which combines stable and suitable housing with the right support.  

The evaluation team makes the following recommendations, based on these findings.

Explore ways to embed a whole system approach at local level 

When applying for national funding, applicants should demonstrate an understanding of how their project fits into the local (vision for a) rapid rehousing system, where the opportunities and bottlenecks lie within this and how they will respond to these. Projects developed by partners working across the pathway, or providing services which support individuals to move more seamlessly through pathways should be prioritised for funding. As the Housing First model matures in some parts of Wales, there is a need to plan for and develop models to respond to individuals’ ongoing and newly emerging health and care needs as they age in place. 

Building in early intervention and prevention

Funded projects should be encouraged to build in a preventative element within their models, wherever this is appropriate. This might, for example, include a floating support offer attached to an accommodation-based model, the flexibility to take some referrals from outside of the statutory homelessness route, or to offer Housing First to people in prison, leaving care or whose homelessness is ‘hidden’. 

Measuring local and national level impact 

The Ending Homelessness Outcomes framework sets out a structured approach to defining and creating measurable indicators and covers the key strategic themes of this evaluation. Many funded projects are also drawing on Housing Support Grant for additional or follow-on funding, so the outcomes framework for that funding stream will also be relevant for them. It will be important to test how indicators from both frameworks work together in different models, contexts, and conditions. Additional bespoke, and self-reported indicators and qualitative data can also generate learning about how models work for diverse groups and what matters most to individuals.  

Projects should consistently record their impact on flow into and through the local rapid rehousing system, e.g. where homelessness is prevented, suitable resettlement achieved, and tenancies sustained following resettlement. 

Welsh Government should hold regular progress meetings to monitor and support funded projects to capture quality evidence, good practice and insights, keeping track of how these are picking up outcomes and impact as per national priorities. 

Building in clarification and oversight across concepts which form key national priorities 

Steps should be taken to ensure there is a consistent and accurate understanding of how key concepts such as ‘rapid rehousing’, ‘coproduction’, ‘person-centred’ and ‘trauma-informed’ should be operationalised. This should happen through: Welsh Government oversight of nationally funded projects; work with the sector to develop succinct practical guidance which sets out minimum standards and best practice; and structures for peer learning and reflection, both horizontally (i.e. with other projects and areas) and vertically (i.e., between Welsh Government, local authorities, and project delivery partners).

Contact details

Report authors: Imogen Blood, Sarah Alden, Shelly Dulson, Anita Birchall, Nicholas Pleace, Sarah Chalmers-Page

Views expressed in this report are those of the researchers and not necessarily those of the Welsh Government.

For further information please contact:
Becca McPherson

Social research number: 29/2024
Digital ISBN 978-1-83577-961-3

GSR logo