Delivering place-based interventions and system change to tackle child poverty: learning and reflections from the child poverty pathfinders' evaluation

Key learning and reflections

The remainder of this paper reflects on challenges and learning around three main themes. Each theme relates to an overarching issue likely to arise when developing and delivering programmes structured around place-based, holistic support. These are:

• Using data effectively to inform and support delivery
• The value of, and challenges around, delivering holistic, person-centred, relationship-based support, and
• Designing ambitious programmes that aim to change how services work with families, rather than simply working within existing structures.

Using data effectively to inform and support delivery

Meeting people where they are: using data to inform place-based outreach

Effective place-based delivery requires data – both at the population level, to identify where programmes should be focusing their efforts to make the biggest difference, and at the individual level, to enable them to reach families directly. The Dundee and Glasgow pathfinders both used data in innovative ways, but also encountered challenges around data access and sharing that are likely to be relevant to other areas seeking to use data to support place-based outreach.

Both pathfinders used data to identify geographic areas with large numbers of children living in poverty in which to locate place-based support services. In Dundee, the decision to focus on Linlathen and Mid Craigie was based on analysis of Scottish Index of Multiple Deprivation Data for the city. In Glasgow, the pathfinder team were able to use Glasgow’s Child Poverty Dashboard, a tool developed to support and track the impact of the Council’s efforts to reduce child poverty, which brings data from multiple sources into an interactive tool that can be used by planners and others. This was used to identify specific wards with high numbers of families in particular circumstances to whom the pathfinder wanted to offer support.

Considering which routine data sources can help programmes identify specific geographic target areas should form part of early planning for any new place-based programmes. The Scottish Government may also wish to consider providing advice to local areas on which Scotland-wide data sources are the most appropriate for identifying how many families there are in specific groups (e.g. families with no earned income, families at risk of in-work poverty, areas with high proportions of minority ethnic populations, etc.), to avoid unnecessary duplication of effort and ensure greater consistency across local authorities.

Dundee also used data to target individual families with children and no income from employment and invite them to engage with support. Earlier iterations had involved writing to families to invite them into local venues for support. However, stakeholders reported that few families had engaged with this ‘in reach’ offer, appearing anxious about the purpose of their invitation. This led to a change of approach to proactive ‘outreach’, with key workers calling at people’s homes to introduce themselves and offer support. The large numbers of families from the Dundee target area (Linlathen and Mid Craigie) who had at least some level of contact with the pathfinder following this change of approach supports the effectiveness of this targeted outreach. Parents interviewed for the evaluation indicated that they would never have attended the pathfinder ‘drop-in’ service for help without this initial key worker visit.

However, the pathfinders encountered some challenges around the legal basis for being able to use individual-level data to support targeted outreach, particularly where identifying the groups of families they wanted to reach relied on data controlled by an external organisation (for example, DWP). In Glasgow, negotiation with data owners to be able to use data in this way had delayed the start of a number of place-based pilot projects. The pathfinders also encountered challenges around data sharing to support internal monitoring and external evaluation (see further discussion of this in the separate evaluation learning paper).

Here too, the Scottish Government, working alongside the UK and Scottish agencies who control the relevant datasets and with advice from the Information Commissioner, arguably has a role to play in providing guidance to local authorities on whether and when it is permissible to use different datasets to facilitate targeted outreach to families. At the time of writing, the Scottish Government were already taking forward work in this area under the Scalable Approach to Vulnerability Via Interoperability (SAVVI) project, led by the Improvement Service. This work should provide greater certainty to future projects seeking to use data to offer support to potentially vulnerable residents.

However, given the complexity and persistence of challenges around data sharing and use, it is also important that data issues are considered early on in the design of individual place-based projects. Where there are barriers, it is important to be clear on the likelihood of individual projects being able to overcome these within their delivery timetable, and to consider possible alternative approaches to both targeting support and measuring outcomes where the ideal data is not accessible (see separate learning paper on evaluation for more on different approaches to measuring outcomes). Clarity around likely limitations to data access will help manage expectations among project leaders and avoid spending time and energy on issues that individual projects may be unable to resolve in the short to medium-term.

Summary of learning points

• Individual projects should consider what local-level data is available to support effective targeting of both areas and individuals at the outset.
• The Scottish Government should provide guidance and clarity around a) which data sources may be most appropriate for projects to use to support area-based targeting and b) options for using data to support direct outreach, including advice on the legal basis. The SAVVI project has been established to provide greater clarity in relation to (b).
• In the short to medium-term, there is likely to remain a need for early conversations among local and national partners to identify ongoing data sharing challenges, understand the implications, and consider what solutions are feasible within the timeframe of the project.

Who is missing from the data?

In addition to issues around access and use, when considering which datasets to use to enable the targeting of support it is also important to reflect on who may be excluded from them. Comments from support workers and other professionals in both Dundee and Glasgow identified families they felt were missing from the data used to support outreach (which typically identified those with specific benefit entitlements). This included families experiencing in-work poverty who are not claiming benefits (a group who may also be less likely to be available during the day, so may not be reached by door-to-door outreach) and asylum seekers with no right to state benefits. This raises questions about whether targeted engagement for a place-based approach should be informed by, but perhaps not limited to, those identified in the available data.

Summary of learning points

• Whatever data sources are used to inform the targeting of place-based projects, there is a need to think carefully about who is missing from this data – and whether or not it is appropriate to extend outreach to other groups, given the aims and scope of the project.

Balancing targeting and more open approaches

The evaluation highlighted debates around the balance between targeting of resources on specific areas or groups and taking a broader approach, where no one who arrives looking for support is turned away. On the one hand, it was argued that a more open approach avoids stigmatising low-income families by setting criteria on who is and is not eligible for support; enables projects to have an impact on wider poverty; and avoids other people who are not part of certain groups missing out on much needed support. On the other hand, it may lead to some dilution of the offer to those families in the greatest need, or at the very least stretch support workers more thinly than is perhaps desirable.

As discussed above, the benefits of offering very locally-targeted, place-based holistic support for engaging local families were clear in Dundee. However, the pathfinder drop-in sessions in Dundee (where people could call in and access support from a range of different local services, as well as the pathfinder team) were also very well used by people who were not from the target area.

As the pathfinders were ‘test’ projects, some of the pressures on resources stemming from high levels of demand from people outside the targeted areas may resolve over time if place-based approaches become the norm and are offered in a wider range of localities. However, the wider question of how to balance effective targeting of resources on key sub-groups (e.g. parents on low incomes) with avoiding unnecessary silos or barriers to accessing help may remain.

In any child poverty intervention, in particular, there may also be a need to consider how narrowly or broadly ‘families’ or ‘poverty’ are defined. For example, the Dundee pathfinder found that grandparents often played a crucial social and economic role in local children’s lives, including through childcare, but were not part of the initial ‘target’ group as they might not themselves be living with children under 16.

The approach adopted by the pathfinders could be seen as broadly aligned with ‘proportionate universalism’. As set out in the Marmot review, ‘proportionate universalism’ argues that “Focusing solely on the most disadvantaged will not reduce health inequalities sufficiently. To reduce the steepness of the social gradient in health, actions must be universal, but with a scale and intensity that is proportionate to the level of disadvantage.” Deciding how this balance between universal and more intensive investment of resources can best be achieved in practice, and what it means for the level of support offered to key target groups and the wider population, is a key issue for any place-based intervention.

Summary of learning points

• In any place-based intervention, there will be a need to consider how to balance geographic targeting of resources on areas of highest need, and managing the level of demand for that support, especially when approaches are being tested in specific areas and not offered universally across a local authority.
• Similarly, in any child poverty intervention, there may be a need to determine how tightly focused any support offer is, and whether those without children, or families who are on the ‘edges’ of poverty, should be able to access support (and at what level of intensity).
• While there is no straightforward formula for achieving the ‘right’ balance between targeted and more open approaches to service provision, it is important this is discussed at the outset of a programme or initiative and on an ongoing basis. This may involve partners considering how far a programme may need to adapt or move away from its original focus or approach in response to learning about local needs.

People and relationships at the centre

The value of a holistic, person-centred approach

In addition to meeting families where they are geographically, both of the child poverty pathfinders sought to ‘meet people where they are’ in their lives by offering holistic, relationship-based support which seeks to understand and address the range of different barriers and goals they may need help with. While quantitative evidence on longer-term outcomes was more limited at the time the phase 2 evaluation took place, qualitative evidence from both parents and staff strongly supported the potential benefits of engaging with people in this way, including those with negative previous experiences of support services. In particular, building consistent, trusting relationships with support workers was effective at supporting people to move from dealing with immediate ‘crisis’ issues (such as lack of food, being unable to pay for fuel, or being in debt) towards longer-term goals (such as employment, or becoming more financially stable).

This in turn was facilitated by:

• Support workers being empowered by management to offer support in a flexible, open-ended way
• Investment of substantial time and energy by the pathfinders into joint working with other services, helping widen the network of resources that support workers can easily link participants with
• The qualities and skills of the support workers themselves, including empathy, kindness, and (in Dundee especially) local knowledge from team members who had lived or worked in the area the intervention was based in.

It is essential that projects seeking to deliver this kind of support foster organisational and team cultures that empower staff to work in these ways, particularly if the aspiration is to move from a small, often exceptionally dedicated, pilot team to rolling out similar approaches across a wider area. This is important both for staff wellbeing and for the sustainability and scalability of projects, which may well fail if they rely long-term on ‘heroic’ levels of commitment from individual teams or staff members.

Both pathfinders placed considerable emphasis on promoting ‘culture change’ and shared values at a management/leadership level and across delivery teams. To support progress on this, the Glasgow pathfinder developed a ‘No Wrong Door’ network, which met regularly and aimed to support joint working and a common culture across a wide range of partner organisations working with families in the city. Co-location of services – as at the multi-agency drop-in in Dundee – is another possible model. Joint training – particularly training aimed at developing and articulating shared aims and values – may also help embed a consistent philosophy across decision-makers and delivery staff. Exploring how to codify and pass on the ‘tacit’ skills, values and expertise developed by both delivery staff and those trying to drive culture change from a leadership level is also something to consider. In addition to sharing learning through joint working, networks and training, this could include, for example, pilot teams developing practical tools or guidance for other services or programmes with similar aims.

Summary of learning points

• Holistic, relationship-based approaches have clear potential to engage families in poverty, including those who may initially be more reluctant to engage with services, and to support people to move from ‘crisis’ towards longer-term goals. Trusting, consistent relationships and flexible support are at the heart of this.
• Supporting this approach to delivery requires fostering organisational cultures that empower staff to work in this way, especially as programmes move from small-scale pilots to roll-out across wider services. Activities to facilitate this could include: investing in networks with a specific focus on common values and joint working; joint training; co-location; and sharing guidance from pilot projects to inform wider management and delivery.

Avoiding dependency

The very reasons that relationship-based support appears to be so effective also create some risks that need to be carefully managed. By leaning into the trusting relationship between support worker and family as the foundation for engaging families with support, there is a risk that families become over dependent on individual staff and less willing to approach other services directly. Feedback to the pathfinder evaluation team from both parents and delivery staff highlighted the potential for this dependency to arise.

If relationship-based approaches are to be sustainable, managing this is important both for families (since the same delivery staff will not be available to them forever) and for delivery teams (since relationship-based support is emotionally demanding work). There is a need to consider the extent to which projects based around relationship-based holistic support are equipping people to better navigate the system themselves in the longer run, as well as directly helping them to access other services in the short-term.

However, avoiding dependency is also, arguably, dependent on the efficacy of wider system change activities taking place alongside individual interventions. If the principles of relationship-based, person-centred, holistic support become embedded more widely across the system, then people’s experience should become more consistent whichever services they go to. Making the wider system easier to navigate, with more consistent person-centred support, should reduce the need for intensive, long-term support from support workers. Hence for sustainability, system change activities need to go hand-in-hand with the development of relationship-based holistic support.

That said, system change is a long-term project, while funding for intensive support teams may be shorter-term. Unless there is funding for the same delivery team to remain in a community long-term, there is a need to consider from the start what resources will remain in the community after the intervention (and funding) ends, to avoid a ‘cliff edge’ for both individuals and communities as the initial support is withdrawn.

Summary of learning points

• Relationship-based support arguably comes with an inevitable risk of dependency on the delivery staff with whom families have formed a trusting bond. How to manage this should be considered from the start.
• Equipping people to navigate the system independently should be an explicit aim of these kind of projects.
• At the same time, activities aimed at making the wider system more navigable and person-centred are also essential. Only if the wider system becomes easier to navigate and provides appropriate support when needed, will the need for long-term, intensive support to help people access the support they need reduce.
• However, particularly if funding for a specific intervention is likely to end before such system change is achieved, the ongoing resources that will be available to support families once dedicated funding ends should be considered from the start.

Moving away from the ‘as is’: designing ambitious system-change programmes

Change takes time, for individuals and systems

As discussed above, system change is a long-term project: the Glasgow pathfinder expect their system change ambitions to take 10 years or more to fully realise. Change for families also takes time, particularly where parents currently face deep barriers to accessing employment or achieving financial stability relating to confidence, or to their physical and mental health. A key early learning in Dundee was that a significant number of the parents they were hoping to support into work were much further from employment than originally anticipated, so initial expectations about the speed at which the pathfinder might be able to support large numbers into employment had to be revised. It is important that projects with similar ambitions are realistic about the time needed for change that is both transformative and sustainable, for both systems and for the families they plan to work with.

This should include clarifying what is meant by short, medium and long-term goals in the context of a particular project. For example, does "long-term" mean making a significant difference in a family's situation while their children are still under 16? If not, what time period is considered appropriate? It should also involve considering the sequencing or staging of outcomes, including what ‘softer’ or intermediate outcomes need to be captured – and over what timescales – to give funders and other partners confidence that a project is progressing towards its ultimate aims. For families, this might be about capturing changes in confidence or wider wellbeing, and feelings about how financially stable their situation is compared to when they first engaged with support, alongside ‘hard’ outcomes around jobs and income. For systems, it may be about capturing early progress in ‘culture change’ or strengthening of relationships and referral mechanisms between delivery organisations, as evidence of the potential for activities to support more radical changes in the future.

Both of the pathfinders emphasised the importance of taking a test and learn approach to project development, where different delivery approaches are tested and then adapted as projects learn about what does and does not work for the families they are aiming to help. Taking this approach is likely to require some flexibility around agreed timings of outcomes (as well as raising questions about the best approach to evaluating the evolving impacts – discussed in the separate learning paper on learning for evaluation). However, there is a strong case for building in explicit review points to such ‘test and learn’ projects, where the evidence to date on what works is formally considered in order to inform decisions about next steps – for example, whether to embed and scale, or revise and retest.

Summary of learning points

• Changing systems and achieving sustainable change for families, especially those facing the most barriers, takes time; it is important to understand and clarify timelines – including what is meant by short, medium and long-term goals – from the start.
• When the ultimate outcomes are only likely to be realised over the long-term, it is important to consider what intermediate outcomes and indicators of progress (for both families and systems) are relevant and what data is required to evidence these and over what timescale.
• If adopting a ‘test and learn’ approach, there is a need to build in review points, where the evidence on what works and the implications for future delivery (and evaluation) plans are discussed and agreed.

Getting the right people round the table (and being clear on roles)

Partnership working is essential to the success of place-based and system change interventions; both the Dundee and Glasgow pathfinders were multi-agency endeavours from the outset. Considering who needs to be round the table and how they can best work together should be an essential part of similar projects.

The organisations that place-based partnerships need to engage will vary depending on their precise aims. A key learning from both Glasgow and Dundee has been around the importance of strengthening links with health and housing support services, as many of the families they were supporting faced significant barriers in these areas which needed addressing before they could realistically engage with employment-related support. Both projects also worked closely with third sector partners, recognising their key role in supporting families. The scope for public sector organisations to learn from the third sector and to increase their reach to families by benefiting from the trust placed in third sector partners was also noted. Having all these partners around the table – rather than simply establishing referral routes with them – arguably enables projects to identify new opportunities to strengthen joint working, whether through new referral routes, co-location, direct input to delivery from staff, or other mechanisms.

Both pathfinders also highlighted the importance of families themselves shaping design and delivery. In Dundee, early conversations with families about their needs reshaped the pathfinder’s understanding of the barriers families were facing, and of the sorts of support they might need to move them towards a more sustainable financial and social position. In Glasgow, feedback from Glasgow Helps clients during the pandemic evidenced the potential for a service based around a holistic support worker approach to engage those being missed by other services, informing both the principles of the pathfinder and the development of further interventions. Glasgow has made service user involvement a key component of its service design activities for subsequent stages of the pathfinder. The Scottish Government have asked FFPs to consider how to embed community voice in planning and decision-making on a systematic basis from the outset, which may generate further creative ideas about how to integrate community voice in decisions about spending, service design and delivery.

As well as considering which partners should be involved at the start, as projects develop it is important to continue to reflect on whether the key partners should remain the same. Ongoing reflection on roles and responsibilities is also key, particularly when leadership is shared between multiple organisations. Both these points are particularly important where a project is evolving over time and moving from a ‘test and learn’ phase to a more settled phase of delivery – are the same partners required, in the same roles, throughout? There is a risk of some ‘drift’ unless all partners are clear on roles and who is leading on each element of strategy and delivery. A related learning point is that when programmes involve large organisations, like councils and national bodies, it is important that buy-in is achieved across the organisation (including from elected members). If programme communications and engagement focus only on particular teams or departments, this may hamper both programme delivery (for example, if actions are identified that require input from other departments who are unaware of the programme) and the effectiveness of wider system change activities.

Finally, if the goal of a project is whole system change, then there is also a need to consider where accountability sits so that, ideally, all organisations involved in ‘the system’ that the project aims to change are working towards the same goals. In Glasgow, the pathfinder addressed this via the Community Planning Process, working to ensure the Local Outcome Improvement Plan reflected a shared vision, outcome and goals for tackling and eradicating child poverty, supported by the development of a shared Performance Framework with 1, 3 and 10-year indicators. This could be a model for other local areas, depending on the level and nature of the changes being sought.

Summary of learning points

• Multi-agency partnership is essential to tackling child poverty. Projects should consider carefully who needs to be involved at both a strategic and delivery level from the outset, including considering creative mechanisms for embedding community voice in decision-making.
• It is also important to reflect as projects develop on whether the core partners remain appropriate, whether roles and responsibilities remain clear, and how to facilitate clear lines of accountability for delivering on the project’s aims across partners.
• Relevant partners will vary depending on the projects’ aims and the families they work with. The pathfinders’ experience indicated that housing, health and the third sector are often important partners. Involving wider partners ‘around the table’, rather than simply as referral partners, might help identify new mechanisms for supporting joint working and more effective delivery to meet clients’ needs.

Building in capacity to drive innovation and system change

‘System change’ is likely to require significant departures from ‘business as usual’. It is important for projects that want to achieve this to consider how they create sufficient capacity and space for innovative thinking.

The commitment to a multi-agency approach in the Glasgow pathfinder went beyond inviting people from different organisations to join a working group or strategic board. Instead, responsibility for driving the pathfinder sat with a ‘Multi-Agency Change Team’, which initially included around 30 staff, seconded or with time bought out for the pathfinder, from a range of different organisations and with a wide range of professional skills. Having this dedicated collaborative resource was seen as absolutely critical to driving the level of change the Glasgow pathfinder is working towards.

The Dundee pathfinder was smaller in scale, but also established a multi-agency partnership of local (Dundee City Council) and national (Scottish Government, DWP and Social Security Scotland) bodies to provide leadership and drive the pathfinder forward. This partnership facilitated some innovations in delivery – such as the multi-agency key worker team, with staff drawn from DCC, DWP and Social Security Scotland and allowed time and resource for the pathfinder to experiment with different delivery approaches. However, questions were raised over whether the national partners involved could have made more of the opportunity to influence change within their own organisations, particularly where organisational policies and practices posed barriers to positive outcomes for families. In terms of learning for future projects, where national partners are involved, there is a need to agree clear expectations and mechanisms for how learning from local projects will be fed back into those national organisations in order to support achievement of outcomes.

The level of resourcing or funding available will differ for different projects. However, radically transforming the support available for families in poverty is likely to be a long-term project, requiring ongoing capacity to drive implementation and support innovation. If this ‘change work’ is a ‘bolt on’ to people’s existing jobs, they are arguably less likely to have the capacity to deliver the desired outcomes.

Summary of learning points

• Projects that aim to test new approaches to joining up family support services in order to tackle child poverty, or which aspire to other forms of system change, need to consider in advance how they will create sufficient capacity and space for innovation, particularly where this is conceived as a long-term endeavour.
• ‘Bolting on’ innovation thinking and system change activity to people’s existing job is arguably less likely to be effective due to capacity constraints.

Communicating (and re-communicating) vision

Another challenge when adopting a flexible or ‘test and learn’ approach to project development is that, as things evolve, partners’ understanding of the overarching goals and vision can drift. At the same time, communicating a clear vision and plan for ‘whole system change’ across the large number of local and national partners who are likely to be involved in delivering it is a challenge, as it involves numerous interlinked activities and outcomes and can easily risk seeming amorphous and impenetrable to those not involved with it on a daily basis.

For projects that aim for system-wide change, maintaining a close focus throughout on the vision the project is trying to achieve is key. Ensuring that all key partners remain on the same page in this regard is also critical. Avoiding jargon and thinking carefully about terms, particularly where the same terms might mean different things to different stakeholders, may help support the clear, ongoing communication of aims.

Summary of learning points

• Communicating a clear, consistent vision when delivering evolving or long-term system change projects can be challenging.
• Consistent communication and avoiding jargon and potentially obscure terms are important in keeping all partners on the same page.

Considering spheres of influence

The ultimate aim of both the Glasgow and Dundee pathfinders, as well as the wider Fairer Futures Partnerships programme, is to move families out of poverty. However, the factors impacting on whether a family is in poverty are multiple and complex, and not all of them are within the scope of initiatives like the pathfinders to influence or change. As well as data sharing challenges already mentioned, issues around national benefit rules and their application, and the capacity of wider services to support families (especially health) were felt to be barriers to achieving outcomes in both areas that were difficult for the pathfinders to change. These are likely to be challenges that other projects will also have to grapple with.

In seeking to reshape services to better meet the needs of families in poverty, it is important that projects are ambitious, and that mechanisms for influencing all parts of the system are built in where possible. However, there is also a need to consider the sphere of influence of local projects and to avoid disparate local projects all spending time on issues that, like data sharing, might arguably be better led at a national level. The involvement of the Scottish Government and other national bodies as partners in funding and developing these local initiatives provides a strong opportunity both to identify common challenges across child poverty projects, and to consider who is best placed to work on resolving them, particularly where this requires collaboration between or within these national organisations.

Summary of learning points

• Multiple and complex factors impact on child poverty – not all of these will be within the sphere of influence of individual place-based partnerships.
• As a key partner in the pathfinders and FFPs, the Scottish Government has a strong opportunity to both identify common challenges and consider the level at which they should be resolved, especially where these relate to national (Scotland or UK-wide) organisations, rules or procedures.