Improving our travel and tourism statistics: September 2026 update

• This article outlines our progress in developing new methods and data underpinning our travel and tourism statistics since August 2025; this also accompanies the release of our updated 2024 estimates.

• Quarter 1 (Jan to Mar) and Quarter 2 (Apr to June) 2024 data were collected and produced under our previous International Passenger Survey (IPS) design, while Quarter 3 (July to Sept) and Quarter 4 (Oct to Dec) 2024 data were collected and produced under our new IPS design; this may lead to discontinuities between the two halves of 2024 and affects how the data should be used.

• We have investigated and corrected errors in our previously published 2024 estimates relating to destination airport and how we process package-holiday spend; we have since taken the opportunity to revisit estimates and implement other consistent methodology developments across Quarter 3 and Quarter 4 2024 data.

• Improvements include the way we account for freight cases and our approach to calibration, which should improve accuracy and consistency in our estimates.

• These methodological developments will also be implemented in the updated 2025 estimates, which we aim to publish in October 2026; this will provide users with six quarters of comparable data under the new IPS design.

• We will continue to develop and quality assure our new travel and tourism methods; we invite user feedback on our published estimates and developing methods.

Overview

This article provides an update on the data collection changes and methodological developments we implemented for Quarter 3 (July to Sept) 2024 onwards. We aim to publish another article in October 2026, alongside the planned release of 2025 data, with detailed plans for future developments.

The work to transform our travel and tourism statistics was first set out in our Improving our travel and tourism statistics: changes from July 2024 article, and we provided progress updates in our February 2025 article, our June 2025 article, and our August 2025 article.

These changes follow a statistical review that concluded in May 2022, with our Travel and tourism review: final report. This review explained why these changes are required and discussed how they would make our statistics more accurate and coherent.

This article accompanies the release of our updated and corrected Quarter 3 2024, Quarter 4 (Oct to Dec) 2024, and annual 2024 estimates. These 2024 estimates are published with corrections relating to destination airport and package-holiday costs, and improvements to increase consistency in methodology across estimates.

Compared with previous annual 2024 estimates, these corrected annual headline estimates of visits, spend, and nights have changed within 3%. The methodological changes outlined in this article also apply to the 2025 estimates that will be published in October 2026. We further detail our development work and progress in this article's sections 3, 4 and 5.

Correction to assignment of destination airport

In our August 2025 article, we discussed an issue relating to international airport codes being incorrectly assigned for some passengers, affecting Quarter 3 (July to Sept) 2024, Quarter 4 (Oct to Dec) 2024, and annual 2024 estimates. Upon further investigation, we discovered that this was because of a questionnaire processing error.

Air passengers are weighted, in part, by the airport to which they are travelling, meaning that the groups to be weighted were either over- or under-estimated. All cases in the same weighting groups as the passengers with the wrong airport codes were weighted incorrectly.

We have now corrected this and re-published our 2024 annual and quarterly datasets. We have improved our processing to prevent this error from happening again. The correction means that the weighting groups have changed, so any estimate could be affected. Estimates based on smaller sample sizes may be more likely to be affected. This correction also means that more detailed, country-based calibration groups can now be used on the Quarter 3 and Quarter 4 2024 data, as detailed further in this section.

Currency conversion for trip costs

Between July 2024 and June 2025, respondents who were unable to separate their total package costs between fares and accommodation, and who gave the figure in a foreign currency, did not have those costs converted correctly into British pounds.

This issue mostly affected overseas residents and has been corrected both in the data and within our processing system to prevent it from happening again. This change led to smaller decreases in spend estimates for the category "Holiday", and to more notable decreases in spend estimates for the category "Holiday - inclusive tour". The other corrections and improvements we made, which are detailed in this article, may also impact changes in spend.

Improvements to Dover and Eurotunnel freight under our new International Passenger Survey design

Under our new International Passenger Survey (IPS) design, there are differences in the way we collect freight data. At Dover seaport, we now sample on the quayside instead of onboard ferries. Health and safety concerns prevent interviewers from working between queuing high-sided vehicles, meaning we no longer sample these passengers (for example, lorry drivers).

In previously published estimates using the new IPS design, non-freight traffic at Dover (the only type sampled) was calibrated to total passenger numbers. However, freight passengers have different characteristics to non-freight passengers. This led to bias in the estimates. For Eurotunnel freight, we did not have a large enough sample size to calibrate freight cases separately to non-freight cases.

To help us overcome this, we have created pseudo records in the data for Dover freight and Eurotunnel freight. These represent passengers who can no longer be sampled under the new method, or those that we do not have a large enough sample for. This should reduce the bias from not sampling enough freight traffic.

These pseudo records are based on several assumptions, including that:

• the proportion of freight traffic at Dover can be estimated using the total number of passengers and of road haulage vehicles passing through the port each quarter, as provided on the Port of Dover website (currently covering 2024)

• the split between UK and overseas residents' visits at Eurotunnel under the old IPS design provides a reasonable estimate for freight cases now

• spend and nights of Eurotunnel freight cases in the last year of the old IPS design provide a reasonable estimate for freight cases now

We aim to continue to monitor and develop this, including seeking alternative data on freight passengers from other government department sources, and evaluating our approach to sampling Dover freight.

Additionally, there are some limitations to this approach, caused by detailed data, such as nationality and final-destination information, not being available in the pseudo records. Therefore, freight data pseudo records can be included only within broad-categories outputs, such as:

• Eurotunnel freight data within the "Europe total - business", "Europe total - all purposes", "World total - business", and "World total - all purposes" categories

• Dover freight data within the "World total - business" and "World total - all purposes" categories

• Dover freight data in some annual sea estimates and Eurotunnel freight data in some annual tunnel estimates

As a result, estimates may not sum to equal the world totals and Europe totals.

Improvements to calibration

Calibration is the last stage of weighting and involves adjusting sample weights so that the responding sample aligns with what we already know about the target population. For example, correcting if too many people from one group or too few from another took part in the survey. This process corrects any imbalances in the responding sample and usually improves the precision of estimates.

In the IPS, we use passenger totals from administrative sources from the UK Government's Department for Transport and the Civil Aviation Authority, calibrating them so that the weights of sampled IPS cases sum to those totals within specified groups. In our previously published estimates for Quarter 3 2024 onwards, we used a one-partition approach to calibration, with a single set of calibration groups defined by departure port and general destination area.

Following improvements to the way flight destinations are assigned, and user consultation, we have moved to a two-partition calibration approach, with one based on departure airport and a second based on destination-country of flight. This should reduce uncertainty and improve precision in our country-level estimates.

As a result, our calibration method is consistent across Quarter 3 and Quarter 4 2024, and will also be consistent with Quarter 1 (Jan to Mar) to Quarter 4 2025, when these are published in October 2026. This calibration method differs to that used in Quarter 1 and Quarter 2 2024.

However, we note that there is some variation in the detail used in the calibration groups, particularly for Quarter 3 2024, where sample sizes were smaller compared with later quarters. This required some differences in the country groupings used.

Minor improvements to the Great Britain Tourism Survey modelling of spend while abroad

We collect and model post-departure spending of residents of Great Britain (GB) while abroad using the Great Britain Tourism Survey (GBTS). The IPS can no longer capture this information under the new method because respondents are not interviewed on arrival to the UK.

For these latest 2024 estimates, we have realigned the country groupings used in these models to improve consistency with our published datasets. The most notable change is the separation of North America into Canada and the US, with other North American destinations grouped separately. This will improve the level of detail in our estimates, while having only a limited impact on the North America estimates and the overall spend estimates. This same approach will be applied to the 2025 estimates when we publish these, in October 2026.

Cognitive testing work is currently underway, and we continue to investigate and develop our methods for measuring post-departure spending. Earlier research found that some respondents misreport their post-departure spending and often include pre-departure accommodation costs. As an interim solution, we removed reported post-departure spend on accommodation from Quarter 3 2024 onwards, prior to fitting the GBTS model. This means our estimates of GB residents' overseas spend rely on assumptions and inaccurate data collection, potentially causing bias in estimates that is difficult to measure.

In the future, we will use findings from our cognitive testing work to redesign the post-departure spending survey questions on GBTS and build new models on this data. Our long-term aim is to publish improved estimates along with adjustments to the published series from Quarter 3 2024.

The GBTS modelled data is also used to estimate standard errors, which create our published confidence interval data for GB residents' spend abroad. In our quarterly and annual datasets, we have paused the publication of confidence intervals for GB residents' spend while abroad, pending further method development.

These estimates are being shared as official statistics in development. However, we are confident that these developing estimates are relevant and provide useful insights into travel and tourism in 2024.

Annual 2024 estimates are unusual in that they represent a composite year where the new International Passenger Survey (IPS) design was implemented from Quarter 3 (July to Sept) 2024. This means that Quarter 1 (Jan to Mar) and Quarter 2 (Apr to June) 2024 estimates are based on a different methodology to Quarter 3 and Quarter 4 (Oct to Dec) 2024. While interpreting these latest 2024 estimates (and subsequent 2025 estimates, planned to be published in October 2026), users can:

• compare quarters over time that are based on the new IPS design (that is, from Quarter 3 2024 onwards); when the 2025 estimates are published, in October 2026, this will provide six quarters of comparable data

• compare annual change between Quarter 3 2024 and Quarter 3 2025, and between Quarter 4 2024 and Quarter 4 2025 (when published in October 2026)

• use confidence intervals to help understand the uncertainty around the estimates, using caution when interpreting estimates where confidence intervals are not currently provided

Users should not:

• directly compare estimates produced from the old IPS design with estimates from the new design, as we are continuing to investigate discontinuities between the old and new methods and will update users as we progress with this work; we will publish an update to discontinuity investigations in October 2026

• compare single 2024 annual estimates against earlier single year or later single year estimates, noting that there may also be a discontinuity between the two halves, based on different IPS methodologies

Users should:

• note our concerns about the quality of data collected on the overseas spend of residents of Great Britain and our ongoing work to improve them, which will result in revisions, and use the estimates published in this article with caution

Remaining data issues and areas for investigation

Some data issues and areas for further investigation remain in these 2024 estimates. The following continue to affect the estimates in this article:

• incorrectly reporting some overseas pre-departure accommodation spend by residents of Great Britain (GB) as post-departure accommodation spend; more information on this is available in Section 3: Corrections and updates to Quarter 3 2024 estimates onwards

• missing "Fly, Cruise stay on board" spend information for GB residents; this is part of our ongoing development work related to post-departure spend captured on the Great Britain Tourism Survey (GBTS)

• missing confidence interval information for GB residents' spend abroad, pending further GBTS method development

• with six quarters of estimates on the new design to be made available in October 2026, we are assessing options for measuring discontinuities from the previous time series and options for adjusting the back series; for now, no adjustments for potential discontinuities have been applied

• following the incorporation of Northern Ireland (NI) data into our headline estimates, we are continuing to explore the feasibility of incorporating NI data into more of our travel and tourism estimates in the future

• we are working with the Civil Aviation Authority (CAA) to harmonise the International Passenger Survey (IPS) and CAA departing passenger survey data collection methods; we continue to monitor and improve these through response rates and the quality of responses, and to review, improve and quality assure methods used on the IPS

Where possible, we will provide updates on these issues and areas in our October 2026 update to this article series.

Time series analysis

Users should be aware that under our new IPS design and methods, some estimates may appear to differ from previous travel patterns, which could reflect our methods changes, real-world changes in travel patterns, or sampling variability in sample surveys. We continue to investigate these differences, and to invite user feedback (see Section 7: Provide feedback).

When we publish the 2025 estimates in October 2026, we plan to provide more detail on the use of time series analysis to investigate the comparability of our new estimates between the old and new IPS design. At that point, we will have six quarters of comparable data produced from the new IPS design. In time, we then aim to use this time series analysis to make any necessary adjustments to the data series. This should improve the usability and comparability of the data.

Discontinuity

Changes in time series resulting from a sudden change in the average level of the series or a complete shift in periodic effects, such as seasonal patterns.

Overseas residents

An overseas resident is a person who permanently resides outside the UK.

Pseudo records

Pseudo records allow us to include estimates for missing groups by adding equivalent artificial records to the sample file where we have gaps.

Visits

The figures for visits relate to the number of completed visits, not the number of visitors. Anyone entering or leaving more than once in the same period would be counted on each visit.

Visits by overseas residents to Great Britain or UK

A visit by an overseas resident is a visit for a period of less than 12 months. Great Britain or UK citizens residing overseas for 12 months or more coming home on leave are included in this category.

Visits outside the UK by residents of Great Britain or the UK

Visits for a period of less than 12 months by people permanently residing in Great Britain or the UK, who may be of foreign nationality.

Weighting

A method used to make the responding sample of passengers representative of all passenger trips. The final weight includes components that represent the sample design, non-response and calibration to known passenger totals, to improve consistency.

If you have any feedback or suggestions on the work we are doing to develop our travel and tourism statistics, please contact us at travel.and.tourism@ons.gov.uk.

If you would like to receive travel and tourism updates, subscribe to our GovDelivery newsletter.