Jason Lowther
For many researchers and analysts working in local government, evaluation can sometimes feel like yet another additional reporting requirement attached to programme delivery. Yet a series of recent evaluations (including homelessness, local growth and skills, and community ownership) demonstrates why evaluation is increasingly central to effective public policy. These studies also highlight several methodological lessons that should influence how local authorities design, commission and use evaluation in the future.
The first lesson is that evaluation is no longer primarily about proving impact. Increasingly, it is about understanding systems. Recent evaluations of homelessness and rough sleeping have moved beyond individual interventions to examine how housing, welfare, health, justice and local government interact. Rather than asking simply whether a programme worked, they explored why outcomes emerge within a complex environment. The findings suggest that fragmented funding, conflicting incentives and a persistent focus on crisis response often undermine preventative approaches, regardless of the quality of local services.
This reflects a broader shift towards complexity-aware evaluation. As councils have known for years, many local challenges, including homelessness, economic development and community regeneration, are influenced by multiple organisations and external factors. Traditional experimental approaches are often difficult to apply in such settings. Consequently, recent evaluations have increasingly used systems mapping, qualitative fieldwork, contribution analysis and mixed-method designs to understand causal pathways rather than attempting to isolate a single intervention effect. The emphasis is increasingly on understanding contribution rather than claiming attribution. We may need to re-consider having systematic reviews of RCTs atop our methodological pyramids?
A second lesson concerns the value of process evaluation. Across multiple policy areas, some of the most useful findings relate not to outcomes but to implementation. Evaluations of local growth funds, skills programmes and place-based regeneration initiatives consistently show that local flexibility enables adaptation to local circumstances. Conversely, they also reveal how compressed delivery timescales can constrain innovation, limit stakeholder engagement and encourage delivery of readily implementable projects rather than longer-term transformation.
Third, the evaluations expose the continuing challenge of measuring long-term outcomes. In local growth programmes, evaluators highlight the difficulty of attributing changes in productivity, employment or economic resilience to specific interventions when impacts emerge over many years and are shaped by wider economic conditions. Similarly, evaluations of community ownership initiatives have already generated evidence of strengthened community participation, local pride and volunteering, but economic sustainability and growth impacts remain less certain because many projects remain at an early stage of development.
This raises an important methodological point. Public sector evaluation often demands evidence before outcomes have had sufficient time to emerge. Analysts should therefore distinguish carefully between outputs, intermediate outcomes and longer-term impacts. Developing robust theories of change and identifying realistic milestones along a causal pathway may provide more meaningful evidence than premature attempts to demonstrate definitive impact. In many cases, demonstrating credible progress towards expected outcomes may be a more realistic and useful objective than proving final impacts.
A fourth lesson concerns the importance of context. Recent place-based evaluations show that outcomes depend heavily on local conditions, governance maturity, partnership networks and existing organisational capacity. Similarly, community ownership projects appear most successful where community organisations already possess strong leadership, governance and project management capabilities. These findings reinforce a recurring conclusion from evaluation research: “what works” is often contingent rather than universal.
For managers commissioning evaluations, this means moving beyond simple questions of effectiveness. The more useful questions are often: what works, for whom, in which circumstances, and through what mechanisms? Evaluations that address these questions may generate practical learning that is far more valuable than a single overall success measure.
Perhaps most importantly, these evaluations demonstrate that good evaluation supports organisational learning rather than accountability alone. While demonstrating value for money remains important, evaluation also provides a mechanism for understanding complexity, adapting delivery, strengthening partnerships and improving future decision-making. In an era of fiscal constraint, rising demand and increasing policy complexity, this learning function may be its greatest contribution.
Taken together, these recent evaluations point towards a more mature understanding of evidence in local public services. They show that evaluation is at its most valuable when it helps organisations understand systems, implementation, context and causation. For local government researchers and analysts, the challenge is not simply to generate more evidence, but to generate evidence that helps decision-makers navigate complexity. In an environment where public services face growing pressures and finite resources, that capability is becoming an essential component of effective public policy and organisational improvement.
Discussion of the available evaluations mentioned are available here.
This article was first published in the September 2026 edition of the LARIA (Local Area Research and Intelligence Association) newsletter.
Jason is Director of the Institute of Local Government Studies at the University of Birmingham. Email [email protected]
