From Huddersfield to Birmingham and on to Belfast, Whitehall’s Labour Market Evaluation and Pilots Fund is testing what actually helps people into good work. Announced at the Spring Budget 2023, the fund pays for departments to run trials and proper evaluations, with HM Treasury and the Cabinet Office’s Evaluation Task Force running three bidding rounds. Crucially, the activity is spread well beyond London, with clear Northern interest.

Kirklees is the standout local case. The Better Outcomes Partnership, an outcomes-based housing support programme, swaps rigid service specs for flexible, person-centred help. An August 2025 impact evaluation found participants were, on average, 3% more likely to stay in work or start a new job within six months than matched comparators. They were also 6% to 10% more likely to sustain non‑receipt of the Universal Credit Housing Element for up to 18 months, and the model came out more effective and less costly per outcome achieved.

In the West Midlands, a Rent Simplification and Support proof of concept aimed to help 18–24‑year‑olds in supported housing step into work. Participants paid a slice of rent from earnings while a top‑up covered at least 87% for up to six months. Take‑up was low and feedback mixed: some said the support wasn’t generous or long enough, and work progression was often short‑lived. Even so, it surfaced the real frictions young people face when benefits, housing and entry‑level jobs collide.

Northern Ireland’s JobStart 50+ pilot backed 180 economically inactive people into six‑month roles at 25 hours a week, reimbursed at National Living Wage rates with employer NI and pension costs covered, plus £1,000 per placement for onboarding and training. Roles were spread across private, public and third sectors. An independent evaluation by Ulster University (Epic Futures) has been completed to inform next steps.

Transport remains a quiet deal‑breaker for jobseekers. The ONS Data Science Campus has built code and a dataset to estimate job accessibility at fine spatial scales, showing stark differences by mode and job type. That gives councils and combined authorities a way to track access to employment over time and to test whether local transport schemes are widening job reach.

Childcare access is the other piece of the puzzle. Ofsted and ONS have mapped “deserts” and “oases” across 180,000 neighbourhoods using a model that factors in local transport and the number of children. An interactive tool and commentary show how accessibility has shifted since March 2020. Employers can now plan shifts and flexibility around the provision that actually exists in their patch.

On affordability, HMRC commissioned Ipsos to survey around 3,600 parents and interview 60 in depth on Tax‑Free Childcare. The scheme nudged hours for some but sat alongside other drivers like childcare choices, work patterns and household finances. Results are due in November 2025 and will sit alongside the expanded free hours offer set out at Spring Budget 2023.

HMRC has also done the hard yards on data linkage, joining up Tax‑Free Childcare, Child Benefit, PAYE and self‑assessment data. That allows analysis of mothers of one‑ to four‑year‑olds before and after using the scheme, and a counterfactual built from non‑users to examine income and timing of return to work from pregnancy to age two. The work supports the Department for Education’s evaluation of the early years expansion.

Health and work are inseparable. ONS analysis of endometriosis, using linked census, hospital and tax records, estimated about 2% of women of reproductive age in the dataset had a diagnosis between 2011 and 2021, with an average age at diagnosis of 35. A February 2025 follow‑up found a statistically significant average drop in monthly earnings one to five years after diagnosis and a 2.7% fall in the probability of being a paid employee over the same period.

There’s better news after bariatric surgery. ONS reported a sustained increase in monthly earnings from six months to five years post‑surgery, largely because more people entered work. Further analysis, including the NHS Diabetes Prevention Programme, is due in December 2025 and will probe who benefits most and how access varies.

Mental health support makes a difference too. NHS Talking Therapies, accessed by nearly 1.2 million people in 2021/22, is linked to a 1.5 percentage point increase in the likelihood of being a paid employee seven years after starting treatment, with a peak average pay rise of £17 a month two years after completing therapy. Impacts were strongest for people not working but seeking work and for 25–34‑year‑olds. Further work will look at benefits, healthcare use and tax paid.

Adverse pregnancy events carry lasting financial costs. ONS found average total earnings losses ranging from £3,511 after a missed miscarriage to £13,581 after a stillbirth over the periods studied. Employment rates typically return to pre‑event levels within one to two years, depending on the event. For families juggling rent and rising childcare costs, these sums are not small change.

Musculoskeletal conditions are another major driver of inactivity. ONS, working with the universities of Leicester and Southampton and NHS England, has linked population‑level data to track how MSK diagnoses and orthopaedic surgery affect employment, pay and hours by socio‑economic group. Findings are expected in 2025 and will be essential reading for sectors with high manual work.

On the tech side, DSIT’s Flexible AI Upskilling Fund launched in April 2024 to co‑fund AI training for SMEs in professional services. Demand was far lower than hoped: 327 successful applications against ambitions to support around 2,000 businesses and 4,300 staff. Process findings point to low awareness, a need for guidance on choosing courses, and ongoing cost barriers for smaller firms-though matched funding did encourage investment. The impact evaluation will report in 2027.

Recruitment trials brought reality checks. A Behavioural Insights Team randomised test on Indeed added a ‘flexible schedule’ badge to sponsored ads. Views ticked up 0.5%, but applications did not rise significantly. A second trial sent official letters about flexible working rights and offered personalised career support to recent parents. Six months on, there was no significant lift in re‑entry rates, though the support sessions were popular and valued. Reports are due this winter.

DBT also explored whether short, bespoke HR support helps smaller firms recruit and retain people at risk of long‑term inactivity. Low take‑up meant a pivot to process evaluation, but the message was clear: one to two days’ support helps tidy policies, improve hiring practice and keep up with employment law, yet it barely touches deeper workforce challenges that need sustained help. Publication is due this winter.

Jobs Plus, a long‑running US model now being piloted here, aims to improve employment outcomes in social housing communities where disadvantage stacks up. Ten pilot sites went live in summer 2024, with an implementation and process evaluation due in summer 2025 and the first impact results in spring 2026. Given the high share of disabled residents and lone parents in social housing, this is one to watch across the North.

Inside government, two July 2025 Cabinet Office reviews matter for productivity. One finds a strong link between skills and productivity-especially in high‑skilled sectors like public administration, health and social work-with better management and workplace innovation strengthening the effect. Another sets out what works in learning design, with blended, interactive methods such as peer coaching typically outperforming single‑method training.

A further evidence review looks at how programmes for public sector staff with chronic conditions are implemented. Success rests on supportive national policy, delivery teams with the right attitudes, and management capable of co‑ordinating services-proof that the way support is run day‑to‑day can make or break outcomes.

The Cabinet Office also piloted a Digital Excellence Programme for Senior Civil Servants. Early signals suggest positive shifts in behaviours and attitudes across data, digital and AI confidence, though not on a new self‑reported productivity measure. A full report was expected in September 2025. Meanwhile, DWP is re‑analysing past trials to gauge how quasi‑experimental methods stack up against randomised control trials-dry work that should sharpen how future programmes are judged and funded.