Summary of progress against the detailed recommendations across all sessions covered at the Assembly

Below is a summary of progress against the detailed recommendations from the sessions covered at the inaugural Assembly. We have focused on the key points highlighted in the Independent Report (denoted in bold in Annex A). As outlined in the previous section, four sessions formed a key component of the high-level priorities to emerge from the Independent Report, namely Coherence; Data sourcing, quality and methods; User engagement; and Local and regional data. To avoid duplication, these sessions are not included below but users can explore progress against the detailed recommendations within these themes in Annex A.

5.1 Economic statistics and measuring progress

The Economic Statistics recommendations continue to be taken forward mainly through the ONS’s Economic Statistics Plan (ESP) and Survey Improvement and Enhancement Plan (SIEP), which are now being brought together into a single Surveys and Economic Statistics Improvement Plan. This reflects a move towards a more coherent end-to-end portfolio approach, bringing together ongoing delivery, improvement activity and enabling work in a more outcome-focused, disciplined and transparent way. Recent progress shows clearer momentum across priority areas, including the incorporation of grocery scanner data into consumer prices statistics to improve coverage and quality, and implementation of the major design improvements to the Transformed Labour Force Survey (TLFS), including data rotation, to strengthen labour market statistics.

Alongside these delivery gains, the ONS has continued to modernise core systems and reduce reliance on legacy processes, including progress in public sector processing and quality reviews. However, the most recent update (published on 15 July 2026) is clear that constrained specialist capacity means some of our ambitions need to be re-sequenced, with prioritisation focused on the most critical outputs and the wider foundations needed for longer-term improvement.

5.5 Health disparities

Recommendations across health and care statistics have been considered by the Health and Care Statistics Leadership Forum and Cross-UK Data and Statistics Group, led by the Department of Health and Social Care (DHSC). Some of the more specific recommendations (e.g. those related to data linkage) are being considered by the ONS where, similarly to economics statistics, the NSI has unique capabilities. There has been some good progress against recommendations relating to improving the quality of data for under-represented groups through better understanding of the causes of missing data, improved granularity and coherence of health data, and in understanding the impact of individual characteristics on health disparities (for example, as shown in the Health inequalities in health protection report 2025).

Work is ongoing to ensure improved communication with users on how health and care data are used and accessed. This has already included publishing case studies on this use, and what can be done to improve the trustworthiness of linked datasets, whilst recognising the important role surveys still play in this space, which is being covered by the cross-GSS Administrative Data Steering Group.

5.6 Public sector performance

Much of the work discussed at the Assembly (particularly in relation to Public Sector Productivity) is being delivered by the ONS as part of business as usual (BAU) workstreams. Recent progress includes quality adjustments being rolled out in line with the recommendations of the Public Service Productivity Review (2023), such as using data from the University of York to provide quality adjustments to health data within Public Service Productivity statistics. There has also been input from departments from across the GSS. For example, there is proactive engagement with HM Treasury, DHSC, and NHS England to improve NHS financial data being used in Public Sector Finances and Public Service Productivity statistics.

There are some recommendations requiring ongoing additional attention, particularly in relation to coherence across the four nations, as well as the harmonisation and standardisation of data collection processes to improve the quality of public sector metrics. That said, quality adjustments are being rolled out across all service areas in line with the recommendations of the Public Service Productivity Review (2023). Work will continue to build on the foundations already in place to most effectively measure public sector performance.

5.7 Data linkage

This is a highly cross-cutting theme where input from across the GSS will be needed to ensure continued progress across the recommendations from the Independent Report. As highlighted earlier, it is recognised that data linking and sharing between government departments is vital to support the production of high-quality statistics and the work of the GSS Data Linkage Champion Network will facilitate this.

The GSS recognises the importance of transparency to build and maintain public trust when linking data. Within the ONS specifically – where a significant proportion of cross-government data linkage takes place given the datasets and capabilities the organisation has access to – there is already significant use of privacy enhancing technologies, including in the Reference Data Management Framework (RDMF). Strict protocols also exist within analytical platforms to ensure there are shared opportunities to access certain datasets in the safest way possible. The creation of the NSO is allowing increased innovation in data linkage to support cross government requirements.

5.8 Net Zero

Prioritisation and business planning decisions across the GSS have affected work related to net zero and wider environmental statistics. This has contributed to the limited capacity within the GSS to take forward the net zero recommendations, including one of the key points from the Independent Report on creating a centralised dashboard. A UK Climate Change Statistics Portal, originally developed as part of the Integrated Data Service Programme (which has now closed), was deprioritised in 2023. Statistics producers are engaging with key users throughout prioritisation and are being transparent about outcomes, especially where outputs are affected.

The GSS Environment, Climate and Nature theme group has produced a first theme-wide workplan in November 2025, which encompasses net zero. The ONS also intends as in previous years to publish an Environment, Climate and Nature insights article, which would also include net zero, in July 2026, bringing GSS statistics together and highlighting trends to users.

5.9 The future of the census

Users should be aware of the decision taken by the UK Government in June 2025, on the recommendation of the UK Statistics Authority, to run a census in 2031. This followed extensive consultation processes and important contributions from the inaugural Assembly, many of which are included as recommendations in the Independent Report. Within this context, users will observe that a number of the recommendations under this theme are simply noted as they require no further action given the decision to run a census.

The four UK nations are all planning their censuses for 2031, which addresses concerns that were raised following Scotland running their previous census in 2022 (a year after the England and Wales, and Northern Ireland 2021 censuses). The importance of small area data is recognised by the GSS (as is hopefully apparent from the update related to the granular statistics recommendations) and Census 2031 will be important in meeting this user need.

5.10 Equalities

Prioritised workstreams within the equalities space have seen good progress. The GSS Harmonisation team within the ONS is developing updated standards and guidance for ethnicity and have carried out a public consultation on the topic, the responses to which are currently being analysed. The initial findings from the consultation will be published in summer 2026, with an updated harmonised ethnicity standard expected by the end of 2026.

Collaborative working with teams working on Census 2031 across the ONS and the four nations will be important to ensuring further progress is observed across a number of recommendations. The Centre for Equalities and Inclusion, in the NSO, supports the increase and improvement in inclusive data across the GSS, taking forward the Inclusive Data Taskforce recommendations to do so. Building on a body of research providing insight into statistically under-represented groups, the following research is due to be published later in 2026: research to provide insights into the lived experience of the Roma community; and results from discovery research exploring children and young people’s perspectives on sharing their views and information with government.

5.11 Business Statistics

Within the ONS (who lead on this area), resources are currently focussed on the recovery of business surveys (as outlined in the SIEP), which has meant that some recommendations have not yet been taken forward. Efforts have focussed on improving the foundations from which business surveys are sampled, collected, and processed, which includes replacing the Inter-Departmental Business Register and legacy technology with the Statistical Business Register in a phased approach. There is a recognition that further research would be required to better understand what was meant by certain recommendations in order to properly respond.

There has been major progress in the classifications space with the ONS recently updating the Standard Industrial Classification (SIC) framework following a consultation process. The GSS recognises the challenge of measuring a changing economy using such a classification system, but this needs to be balanced against international comparability and consistency over time. The GSS will continue to consider if there is more that can be done in this area to meet statistical needs.

5.12 AI and Technology

The GSS recognises the potential for AI to act as a key enabler of future statistical production and dissemination, supporting improvements in efficiency, quality, and responsiveness. This potential has been actualised with the ONS being one of the first NSIs globally to deploy generative AI in the production of statistics, with the in-house development of a ClassifAI application now used to code labour market statistics by type of occupation and industry.

The recent creation of a GSS AI Steering Group also demonstrates the importance of this subject, whilst recognising the challenges AI poses. Early work is beginning to demonstrate use cases, and the ONS is starting to engage external technology partners to understand how AI tools may interact with official statistics.

The GSS AI Steering Group has been set up to:

• understand the current AI landscape within the GSS;
• identify and address gaps in AI use/guidance for statistics;
• establish mechanisms for sharing and updating best practice; and
• identify any actions needed to develop AI capability within the GSS.

Work is also being undertaken across the Analysis Function (AF), and as a starting point, an AI Hub, which brings together guidance from across the AF into a single location, has recently been launched. A core focus of the GSS group is ensuring AI use aligns with the Code of Practice, including transparency, quality and accountability. Consideration is also being given to validation of AI models, impacts on public trust, and risks from AI‑generated summaries or outputs.

Internationally, the Authority is playing a leading role in shaping how AI is used responsibly in official statistics. Through UN expert groups and partnerships with other national statistical offices, we are helping to develop shared approaches to quality, transparency, accountability and fairness in AI-enabled statistics, while bringing international learning back into the UK system.

The ONS’s Director General for Innovation and Emerging Technology is driving a focus on use of linked data, Machine Learning and AI to enable quality gains across the Official Statistics system.

5.13 Crime

This area is being led by the ONS with wider support from the GSS (predominantly the Home Office). Although all the Crime statistics recommendations to come out of the Assembly have been considered, teams have identified that, in some instances, further work would be required to implement these and there is not currently the resource to do this. There are also recommendations which are not statistical challenges, e.g. to change perceptions about the implications of reporting crimes on house prices.

That said, good progress has been made in some areas, such as through combining crime survey and administrative data. Within ONS crime statistics, administrative data and Crime Survey for England and Wales (CSEW) data have been triangulated. The Home Office has developed and published the Police Performance Framework. This takes a balanced scorecard approach to hold forces to account on key headline measures, including both police recorded crime and CSEW data, while monitoring a broad set of contextual measures to mitigate the risk of taking a narrow focus or creating perverse incentives.

5.14 Labour Market

Good progress has been observed across the full breadth of labour market-related recommendations to come out of the Assembly, with this work being led by the ONS. The importance of improvements to the quality of labour market statistics has meant the ONS is dedicating significant resource to ensuring these metrics are improved and quarterly updates covering both the current Labour Force Survey (LFS) and work on the new Transformed Labour Force Survey (TLFS) are published. These updates show users the progress that is being made whilst being transparent on challenges.

Importantly, although surveys will always play a fundamental role in labour market statistics, teams within the ONS are also exploring alternative data sources. The ONS has started a project to investigate how linking administrative data with labour market surveys could help deliver greater insight into the UK labour market picture. The importance of interpreting LFS or TLFS data alongside other labour market data is also part of the updated LFS narrative as improvements have been made – an important step forward for users of labour market statistics.