1. Overview
The Office for National Statistics (ONS) has a rolling programme of research and development aimed at improving and maintaining its consumer price inflation statistics. The development plan will ensure that the statistics continue to meet user needs, make use of new and innovative methods and data sources, and follow international best practice. The delivery of quality consumer price statistics remains a core priority for the ONS, particularly following the findings from the Devereux Review of the Office for National Statistics performance and culture and the strategic direction set by the Economic Statistics Plan (ESP). The ESP priorities form the foundation of our transformation work, focusing on the delivery of core economic statistics. However, they are not exhaustive, and this article describes additional developments we aim to make to further strengthen our consumer price statistics.
Over the last few years, we have seen several transformational changes to our consumer prices statistics. This included incorporating new data sources, along with developing new methods and systems required to facilitate these changes. These changes are detailed in our Transformation of consumer price statistics article series. In particular, we have successfully implemented changes into headline statistics for several categories, starting with rail fares (launched in March 2023), followed by private rents (for Great Britain) and second-hand cars (launched in March 2024). In March 2025, we completed the transformation for private rents (for Northern Ireland). We also expanded the use of our new technology platform to include the processing of our traditional data sources, as well as improving the classification structure.
We aim to introduce groceries scanner data in our February 2026 consumer price inflation bulletin (published in March 2026), as described in our Introducing grocery scanner data into consumer price statistics methodology.
We are also working on reducing risk around the use of legacy technology (as discussed in the ESP), which is still used in some areas of the end-to-end processing system. This limits our ability to flexibly update our systems. We currently collect some data using legacy technology and use other legacy systems to manually process some data we have collected. Over the next 12 months, we will begin transitioning more of our data collection process from legacy technology to a cloud-based strategic platform.
While the work described in the ESP covers our highest priorities, we will also be progressing more specific work over the next 12 to 18 months, working through the development plan and ongoing continuous improvement. The workstreams on this plan capture our top priorities as currently understood, however, this is a non-exhaustive list and if circumstances change, there is scope to re-order work to allow for emerging high priorities as required.
Alongside the first annual refresh of our ESP in April 2026, we are planning to publish a consolidated prices development plan setting out our priorities for 2026 to 2027 covering:
consumer prices
house prices
private rents
business prices
deflators
2. Recent development of consumer price statistics
Recent work since our Consumer prices development plan: updated August 2024 article, includes:
March 2025: incorporated improved imputation methods, consumption segments and Northern Ireland private rental statistics into consumer price statistics, as shown in our Consumer price inflation, UK: February 2025 bulletin
March 2025: published an update to the shopping basket for the Consumer Prices Index including owner occupiers' housing costs (CPIH)
March 2025: expanded the coverage of gas and electricity items to include fixed-rate tariffs in addition to variable-rate tariffs
April 2025: published an indicative impact analysis for groceries scanner data for introduction in March 2026
We are not expecting to publish updates to our transformation and impact analysis articles once we have integrated scanner data, but we will keep this under review as other workstreams progress.
Nôl i'r tabl cynnwys3. Continuous improvement of consumer price statistics
Alongside the existing planned complex changes, we continually review the quality and relevance of our consumer price statistics. This work informs our priorities for future development, either as part of ongoing maintenance of our statistics or potential future development workstreams. In setting these priorities, we consider a range of factors, including:
stakeholder feedback
data quality and associated risks
efficiency gains
resources available
In March 2025, we set up an internal delivery board to prioritise continuous improvement work across consumer prices. As part of this, we have commissioned ongoing reviews ("deep dives") to identify areas for improvement in data collection and methodology of consumer price inflation.
The first review completed in summer 2025 and specifically focused on the price collection of those goods and services that contributed the most to the volatility in core inflation. The review also considered international best practices to inform recommendations. Some proposals, for example, expanding the sample coverage of video games, will be implemented as part of the regular annual basket update. Several other workstreams were identified as candidates for improvement, for example, enhancing the input data used for package holidays, but these are more complex, and require more detailed planning before they can be considered for prioritisation within the development plan. The review also proposed new processes for early identification of problems and for maintaining robustness of price statistics which are currently under development.
We will continue to commission these reviews as a standing part of continuous improvement to produce a list of recommendations for given basket categories. Those that require complex change will be pulled out as separate workstreams within this plan and prioritised accordingly.
As part of continuous improvement, we are also in the process of updating our Consumer Prices Technical Manual and Quality assurance of administrative data used in consumer price inflation statistics (QAAD), to be published in March 2026. These updates aim to ensure our documentation reflects current practices and supports transparency and confidence in our methods.
Nôl i'r tabl cynnwys4. Summary of our workstreams
Here we present a summary of our workstreams, by priority and complexity of implementation based on the data and methods work required, as well as any system build and development of new business processes. We have also added a time metric to these summaries to indicate when users are likely to see updates. Where not explicitly stated, it is anticipated that these projects will be multi-year deliveries, and we will provide updates on these in due course.
Grocery scanner data
Priority: high
Data: new data sources are being introduced
Systems: existing cloud-based systems can be used
Methods: existing methodological development can be used
Processes: new processes will need to be introduced
Timelines: this is expected to be delivered in March 2026
Removal of legacy platforms and systems
Priority: high
Data: existing data can be used; no major changes expected to input sources
Systems: significant system redevelopment required
Methods: existing methods remain largely unchanged; focus is on system infrastructure rather than methodological innovation
Processes: new processes will need to be introduced
COICOP 2018 implementation
Priority: high
Data: existing data can be used with some harmonisation required
Systems: some system redevelopment required but can take advantage of new systems already developed as part of the move away from legacy
Methods: high methodological development needed
Processes: new processes will need to be introduced
Seasonally adjusted Consumer Prices Index (CPI)
Priority: medium
Data: existing data can be used
Systems: can be integrated into existing systems
Methods: moderate methodological development needed
Processes: new processes will need to be introduced for publishing outputs
Timelines: this is expected to be delivered in the first half of 2026
Household Costs Indices (HCIs)
Priority: medium
Data: some further improvements could be made to the current data sources
Systems: existing system supports publication, however enhancements needed for accreditation
Methods: high methodological development needed
Processes: existing processes are in place for publication
Measures of accuracy for CPI
Priority: medium
Data: existing data sources required
Systems: analytical systems for variance estimation
Methods: high methodological development needed
Processes: new processes will need to be introduced
Expanding the use of administrative data
Priority: medium
Data: existing and new data sources required
Systems: existing cloud-based systems can be used
Methods: moderate methodological complexity
Processes: new processes will need to be introduced
5. Consumer price statistics development plan
High priority items
High priority items are the main element of our development plan and will usually be prioritised over the delivery of lower priority items (depending on complexity). However, inclusion in this plan does not guarantee delivery; final decisions will depend on prioritisation outcomes and any emerging priorities throughout the next 12 months (this list is non-exhaustive).
Grocery scanner data
We have now delivered many of the transformational changes set out for consumer price statistics. In April 2025, we produced indicative impact analysis for groceries scanner data, however, because of the complexity of data and methods and the importance of these statistics, further quality assurance is required to ensure they are of the highest quality.
We plan to parallel run scanner data for a year before integrating it into live production in March 2026, with the new data covering 50% of the market share. For the remaining 50% of the groceries market, we will continue to manually collect prices in-store and online.
As part of the groceries scanner data workstream, there has been ongoing research into reviewing our methods for the calculation of retailer weights and introducing a different classification of shop type based on market share. More information is available in our Introducing alternative data into consumer price statistics: aggregation and weights article. Further analysis has shown that rather than focusing specifically on shop type stratification and weights, it will be better to do a more holistic review of the stratification types used across the consumer prices hierarchy in conjunction with the choice of sample. We will reprioritise this workstream accordingly.
We currently publish much of the price microdata used to calculate consumer price inflation statistics, specifically our Consumer price inflation consumption segment indices and price quotes dataset. With scanner data, this will no longer be possible for groceries because of its size, structure, and data sharing agreements with retailers.
To meet user needs, we ran a User needs from consumer price inflation item indices and price quote microdata engagement questionnaire in late 2024. Responses suggest that new aggregate statistics could help meet current microdata use cases. We aim to share example aggregate outputs before March 2026.
Complexity assessment:
This workstream is high priority. A considerable amount of work has already been completed in delivering this transformation, so the complexity of the remaining work is lower as a result. The remaining work largely focuses on further quality assurance (including completing the parallel run).
Removal of legacy platforms and systems
While significant improvements have been made in recent years, consumer prices remain reliant on legacy technology for important parts of our production process. This limits our ability to flexibly update our systems. In particular, we currently collect some data in-house ("central" collection) using spreadsheets and other legacy systems. We also use other legacy systems to process some data we have manually collected. During 2026, the central data collection will begin transitioning from legacy technology to a cloud-based strategic platform. Feasibility work will also begin into remaining legacy improvements.
Complexity assessment:
This workstream is high priority with high complexity. While the data and methods are unlikely to change (and therefore it is expected to have low impact on headline numbers), the work required to develop new business processes and test the new system on the cloud-based strategic platform is complicated.
Classification of individual consumption by purpose (COICOP) 2018 implementation
The COICOP provides the foundational structure for the Consumer Prices Index including owner occupiers' housing costs (CPIH) and the Consumer Prices Index (CPI). It also underpins the construction of Household Final Consumption Expenditure (HHFCE) estimates within the System of National Accounts, which are used to weight the CPIH and CPI baskets. In 2018, the United Nations finalised revisions to the COICOP framework, shown in the United Nations COICOP publication (PDF 1286KB), prompting the need for UK consumer price statistics to be restructured accordingly.
We are currently in a discovery phase for consumer prices, working to understand the main requirements for COICOP implementation. This includes close collaboration with other outputs, particularly the Living Costs and Food Survey (LCF) and HHFCE, to ensure a coordinated and aligned approach to implementation.
Complexity assessment:
This workstream is high priority with high complexity, because of its cross-cutting nature and the need for harmonisation across multiple statistical domains for the first time. The effort involved is expected to span several years and will require both methodological development and system-level updates.
Medium priority items
Medium priority items form an important part of the development plan. However, with our current focus on high priority items, we are only able to progress medium priority items if the work does not impact on the progress of the high priority items.
Seasonally adjusted Consumer Prices Index (CPI)
We commissioned a feasibility study on the most appropriate methodology to use in the production of seasonally adjusted consumer prices inflation by the National Institute of Economic and Social Research (NIESR).
The project has now been successfully handed over to us, where development work to implement the adjustments are underway. We aim to produce a monthly seasonally adjusted set of indices that can be published as supplementary tables from the first half of 2026.
Complexity assessment:
This workstream is medium priority with low to moderate complexity in its current phase. The feasibility study has provided a clear methodological foundation, and the implementation is expected to be relatively contained. However, integration into existing systems and processes, and ensuring consistency with headline outputs will require careful planning and testing.
Developing the Household Costs Indices (HCIs)
The HCIs, which measures price change as experienced by different household groups, was first suggested by Astin and Leyland as Household Inflation Indies. This was highlighted in in the Johnson Review 2015, UK Consumer Price Statistics: A Review - UK Statistics Authority. We first published preliminary experimental HCIs in 2017.
We now publish monthly HCIs data in our quarterly Household Costs Indices for UK household groups bulletin. The HCIs are currently classed as "Official statistics in development" so future work will focus on seeking official statistic accreditation.
Complexity assessment:
This workstream is medium priority with moderate complexity. While the existing infrastructure supports regular publication, achieving accreditation will require enhancements to methodology and data sources, as well as additional quality assurance, and stakeholder engagement.
Developing measures of accuracy for consumer price statistics
As a result of the complex survey design, calculating standard errors for the level and growth of consumer prices is challenging.
In 2021, an academic review on producing standard errors for consumer price indices was published, Estimating Sampling Errors in Consumer Price Indices (Wiley Online Library). We plan to publish further work in collaboration with the University of Southampton, including:
interim estimates of sampling variance for consumer price inflation statistics
an academic paper on these estimates
extending variance estimation to new data sources such as rail fares, second-hand cars, grocery scanner data and rental microdata
Complexity assessment
This workstream is medium priority with high complexity. The estimation of standard errors for consumer prices is methodologically challenging because of the dual sampling dimensions (item selection and outlet selection), which complicate variance calculations. Extending variance estimation to new data sources adds further technical demands. The effort is expected to be substantial and sustained, requiring specialist input, robust data infrastructure, and ongoing methodological development.
Expanding the use of administrative data
We have successfully incorporated administrative data in several categories across consumer prices. This workstream looks at where we can continue to expand the use of these data sources. There are a few examples, including:
using our new data source for used cars to expand the coverage of electric and hybrid second hand vehicles
using our existing retailer scanner data to cover additional categories (for example, household goods)
incorporating additional retailer data into the calculation of our grocery indices
These developments are still under consideration alongside our other priorities, so there are currently no associated timelines.
Complexity assessment
This workstream is medium priority with medium complexity. In some cases, additional data sources may be integrated. Existing systems and methods have been designed to integrate these changes without much additional work, although there may be some methodological considerations for specific categories (for example, the product definition for electric used cars) and existing processes may need updating.
6. Provide feedback
We will review this content annually.
If you have any feedback or suggestions on our work on consumer price statistics, please contact us at cpi@ons.gov.uk.
Nôl i'r tabl cynnwys8. Cite this article
Office for National Statistics (ONS), released 18 September 2025, ONS website, article, Consumer prices development plan: updated September 2025