Reference metadata describe statistical concepts and methodologies used for the collection and generation of data. They provide information on data quality and, since they are strongly content-oriented, assist users in interpreting the data. Reference metadata, unlike structural metadata, can be decoupled from the data.
Eurostat, the statistical office of the European Union
1.2. Contact organisation unit
G2: European businesses
1.3. Contact name
Confidential because of GDPR
1.4. Contact person function
Confidential because of GDPR
1.5. Contact mail address
5, rue Alphonse Weicker
2920 Luxembourg, LUXEMBOURG
1.6. Contact email address
Confidential because of GDPR
1.7. Contact phone number
Confidential because of GDPR
1.8. Contact fax number
Confidential because of GDPR
2.1. Metadata last certified
2 October 2026
2.2. Metadata last posted
2 October 2026
2.3. Metadata last update
2 October 2026
3.1. Data description
This collection covers statistical data on short-stay accommodation offered via online platforms. The data relies on privately held data provided directly to Eurostat by major international online booking platforms (Airbnb, Booking and Expedia). The collection provides a comprehensive census of bookings made through these specific platforms, bridging an information gap regarding holiday rentals that are often outside the scope of traditional tourism registers.
The data includes monthly and annual statistics on the number of stays, the number of nights rented and the number of guest nights spent. It features breakdowns by geography (national, NUTS 1, NUTS 2, NUTS 3 and selected cities), the origin of the guest (domestic vs. international, country or world region) and accommodation characteristics (mode and size class).
The scope is limited to short-stay accommodation facilities that fall under NACE Rev.2 55.2 ("holiday and other short-stay accommodation") and are booked via the participating online platforms. It excludes hotels and similar accommodation (NACE Rev.2 55.1) as well as camping grounds, recreational vehicle parks and trailer parks (NACE Rev.2 55.3).
(Note: Following the general ESS implementation schedule, the transition to the updated NACE Rev. 2.1 classification applies to European statistics from reference year 2025 onwards.)
Number of stays: The number of times a facility offered via the platforms was occupied (i.e., individual bookings).
Number of nights (rented out): The number of nights a facility offered via the platforms was occupied.
Number of guest nights: The number of nights spent during a stay, taking into account the size of the travel party (e.g., a family of 4 staying 3 nights represents 1 stay, 3 nights rented and 12 guest nights).
Cities: Defined based on Local Administrative Units (LAU) where at least 50% of the population lives in an urban centre. The data specifically covers 'selected cities' (capital cities, cities with over 200,000 inhabitants and other cities that jointly account for 90% of the annual guest nights spent in cities in the country).
3.5. Statistical unit
The statistical unit is the accommodation facility for short-term accommodation offered on the online platforms.
3.6. Statistical population
The statistical population consists of all short-stay accommodations (NACE 55.2) located in the reporting countries that are actively listed and booked via the participating online platforms (Airbnb, Booking and Expedia).
3.7. Reference area
European Union Member States and EFTA countries.
3.8. Coverage - Time
Data are available from reference year 2018 onwards.
3.9. Base period
Not applicable.
Data are expressed in absolute values (number of stays, number of nights, number of guest nights).
The reference periods are the calendar month (for monthly datasets) and the calendar year (for annual datasets).
6.1. Institutional Mandate - legal acts and other agreements
The data collection is governed by landmark non-disclosure agreements (NDAs) signed in early 2020 between the European Commission (Eurostat) and the participating international platforms. These agreements establish the contractual framework for the secure transmission, processing and dissemination of privately held data for the sole purpose of compiling official statistics.
Note: Regulation (EU) 2024/1028 on data collection and sharing relating to short-term accommodation rental services applies from May 2026, complementing the broader statistical framework for this segment.
6.2. Institutional Mandate - data sharing
Aggregated and validated data are shared with the National Statistical Institutes (NSIs) of the Member States via secure bilateral agreements. This enables NSIs to conduct secondary validation, integrate the data into national dissemination programmes and reduce the reporting burden on local respondents.
7.1. Confidentiality - policy
Statistical confidentiality is strictly enforced in accordance with Regulation (EC) No 223/2009 on European statistics. The signed data sharing agreements ensure that no personal data (as defined by the GDPR) regarding individual guests or property owners is transmitted to Eurostat.
7.2. Confidentiality - data treatment
Data are processed in a secure IT environment. To protect the commercial interests of the platforms and to guarantee a level playing field, Eurostat only disseminates aggregated figures across the platforms; no data pertaining to an individual platform's market share or volume is published. Furthermore, a secondary confidentiality algorithm is applied to highly granular geographical aggregates (e.g., NUTS 3 regions) to prevent the accidental disclosure of occupancy information for businesses operating in areas with a very small number of bookings.
8.1. Release calendar
Data are released according to a schedule based on the reference period:
Quarterly data: Released mostly one quarter after the end of the reference quarter. The exception is data at the NUTS 3 regional level, which is released two quarters after the end of the reference quarter to respect statistical disclosure control rules.
Annual data: Released once annually, two quarters after the end of the reference year.
In line with the Community legal framework and the European Statistics Code of Practice Eurostat disseminates European statistics on Eurostat's website (see item 10 - 'Accessibility and clarity') respecting professional independence and in an objective, professional and transparent manner in which all users are treated equitably. The detailed arrangements are governed by the Eurostat protocol on impartial access to Eurostat data for users.
Data are disseminated on a quarterly basis (for the monthly occupancy datasets) and on an annual basis (for detailed datasets featuring structural breakdowns). See item 8.1 - 'Release calendar'.
10.1. Dissemination format - News release
News articles and highlights are published on Eurostat's website coinciding with major data updates.
10.2. Dissemination format - Publications
Dedicated Statistics Explained articles are regularly updated:
The data are disseminated in the online database under the Tourism (tour) domain, specifically within the "Accommodation offered via online platforms (tour_ce)" sub-tree. It comprises 2 monthly and 6 annual datasets.
10.4. Dissemination format - microdata access
Not available. Eurostat does not receive or hold any microdata. The platforms transmit data already aggregated at the level of Local Administrative Units (LAU); booking-level data remain with the platforms.
10.5. Dissemination format - other
Not applicable.
10.6. Documentation on methodology
A dedicated Methodological Note for statistics on short-stay accommodation offered via online platforms is available. It details the scope, governance, data transmissions, processing and validation steps.
10.7. Quality management - documentation
The metadata collected from the platforms documenting their internal methodologies (e.g., definition of stays, LAU matching, origin of guests) are reviewed to ensure compliance with agreed-upon definitions.
11.1. Quality assurance
Data undergo a multi-layered quality assurance process. Format, completeness and consistency checks are executed automatically via Python scripts. Additionally, National Statistical Institutes perform a secondary validation layer using local market knowledge before the data is published.
11.2. Quality management - assessment
The data is of high quality as it represents a full census of the bookings made via the partner platforms rather than a sample survey. Known methodological limitations (such as very minor LAU matching issues due to user-reported addresses or discrepancies in the allocation of multi-month stays) have been assessed and their statistical impact is considered negligible (under 0.1%).
12.1. Relevance - User Needs
The short-term rental segment of the tourist accommodation market has grown rapidly but has traditionally been under-covered by official statistics due to the small size of the operators (often below the 10-bed reporting threshold).
This data collection fulfils a critical need for policymakers, researchers and the tourism industry to monitor the volume, geographical spread and seasonality of platform-based tourism, aiding in urban planning and housing policy.
12.2. Relevance - User Satisfaction
User satisfaction is monitored via direct feedback to the functional mailbox and through interactions with the Working Group on Tourism Statistics.
12.3. Completeness
The data offers a highly complete picture of the market segment handled by the partner platforms. However, users should be aware that it does not represent the entire short-term rental market (as direct bookings or bookings via non-partner platforms are excluded).
13.1. Accuracy - overall
The overall accuracy is high. Because the data is collected directly from the platforms' databases, it represents a complete census of their transactions. Consequently, there is no sampling error. In terms of occupancy data, double-counting is highly unlikely, as a specific transaction for a given stay is recorded by only one platform.
13.2. Sampling error
Not applicable, as this is a census of the participating platforms' data.
13.3. Non-sampling error
Minor non-sampling errors may include:
Coverage error: A negligible fraction of bookings (estimated at <0.1%) may fail to match with Local Administrative Units (LAU) due to inaccurate address information provided by the hosts ("offshore" bookings).
Measurement error: Differences in internal platform methodologies (e.g., whether a stay spanning two calendar months is attributed to the month of arrival or departure) can result in minimal chronological shifts, though the impact on total volume is negligible.
14.1. Timeliness
The platforms transmit the monthly occupancy data no later than 60 days after the end of the reference quarter. This delay is tied to the financial reporting obligations of the platforms; as publicly listed companies, they are restricted from sharing operational data prior to their official quarterly earnings releases. Following validation, Eurostat disseminates the data according to its release calendar. See item 8.1 - 'Release calendar'.
14.2. Punctuality
Data are generally delivered and published according to the agreed-upon schedules outlined in the data-sharing agreements signed with the platforms and their amendments.
15.1. Comparability - geographical
The data is highly comparable across countries, as it is based on a uniform data structure and common definitions applied by global platforms. Geolocation mapping to LAUs is performed by each platform prior to transmission, using the same official GISCO LAU shapefiles provided by Eurostat. Aggregation from LAUs to NUTS regions and cities is carried out by Eurostat.
Minor discrepancies exist regarding the attribution of guests residing in overseas territories, but this does not affect overarching comparability.
15.2. Comparability - over time
Data are generally comparable over time starting from 2018. No structural break in the time series was observed when Tripadvisor exited this market segment at the end of 2024. The data from reference year 2025 onwards reflects the activity of 3 platforms (Airbnb, Booking, Expedia) rather than 4.
15.3. Coherence - cross domain
Data from the tour_ce domain cannot be added to the traditional tourism statistics (occupancy of tourist accommodation establishments, tour_occ) collected under Regulation (EU) No 692/2011 concerning European statistics on tourism. Depending on national survey frames and thresholds, there is an overlap between the short-term rentals captured by the platforms and those captured by national registers.
15.4. Coherence - internal
Validation checks ensure internal consistency. Cross-checks are run to ensure that capacity files (CAP) match the occupancy files (OCCL/OCCV) and that period-to-period developments are logically consistent.
The data collection imposes zero burden on individual hosts and enterprises. By utilizing privately held digital data provided centrally by the platforms, the European Statistical System gains census-level data while significantly reducing the response burden that would otherwise fall on thousands of small service providers and National Statistical Institutes.
Reported errors are assessed for seriousness to determine whether they should trigger a correction of already disseminated data.
When platforms submit revised datasets (e.g., following updates to their internal geographical mapping), Eurostat processes the corrections and updates the public database accordingly, communicating major revisions to users.
Reported errors are corrected in the disseminated data as soon as the correct data have been validated.
Data are only published once they are deemed to be sufficiently complete for all data providers contributing to the aggregate.
Whenever new data are provided and validated, the already disseminated data are updated.
18.1. Source data
The source data consists of aggregated datasets securely transmitted directly by major international online platforms (Airbnb, Booking and Expedia).
18.2. Frequency of data collection
Data files pertaining to occupancy and capacity are collected quarterly.
18.3. Data collection
The platforms encrypt the data files and transmit them securely via EDAMIS (Eurostat’s standard data transmission tool). Eurostat stores the files in a secure environment.
18.4. Data validation
Eurostat applies a list of automated format, completeness and consistency checks. This includes comparing Local Administrative Unit (LAU) codes against reference lists, validating subtotals across different files, verifying that the number of guest nights is logically coherent with the number of nights rented and checking period-to-period development. Merged datasets are subsequently transmitted to the NSIs for a secondary, market-specific validation layer.
18.5. Data compilation
Valid files from the individual platforms are aggregated by Eurostat to create single master files. This aggregation process is crucial for statistical disclosure control; it ensures that the outputs do not reveal the individual market share or proprietary data of any single platform. The compiled data is then aggregated geographically (NUTS and LAU levels) and structurally (by size class, guest origin, etc.) for dissemination.
18.6. Adjustment
No specific statistical adjustments (e.g., seasonal adjustment) are currently applied to the data.
To ensure the correct interpretation and use of the statistics on short-stay accommodation offered via online platforms, users should take note of the following important methodological and operational clarifications:
Distinction from official tourism statistics and non-additivity The released occupancy datasets are conceptually and legally distinct from the traditional official tourism statistics collected under Regulation (EU) No 692/2011. The data refer exclusively to accommodations booked through the partner online platforms. Because there are unquantifiable overlaps between properties listed on these platforms and those reporting to traditional national tourism registers, these figures cannot be added to or combined with official tourism statistics (such as hotel stays or traditional holiday rentals) to calculate total market size.
Statistical purpose and enforcement limitations The data are collected and processed exclusively for statistical purposes under strict confidentiality guidelines (in line with Regulation (EC) No 223/2009 and the data-sharing agreements signed with the platforms). To protect the confidentiality of individual statistical units, Eurostat does not receive or hold any microdata. The data published are aggregated and, therefore, cannot be used for administrative, regulatory or enforcement purposes (e.g., identifying individual citizens, property owners or specific bookings).
Geographical allocation (LAU mapping) The assignment of a booking to a specific Local Administrative Unit (LAU) or city is based on location information entered by hosts. Because Eurostat does not receive microdata, this geographical matching is performed directly by the platforms themselves prior to transmission, using official LAU shapefiles provided by Eurostat.
This collection covers statistical data on short-stay accommodation offered via online platforms. The data relies on privately held data provided directly to Eurostat by major international online booking platforms (Airbnb, Booking and Expedia). The collection provides a comprehensive census of bookings made through these specific platforms, bridging an information gap regarding holiday rentals that are often outside the scope of traditional tourism registers.
The data includes monthly and annual statistics on the number of stays, the number of nights rented and the number of guest nights spent. It features breakdowns by geography (national, NUTS 1, NUTS 2, NUTS 3 and selected cities), the origin of the guest (domestic vs. international, country or world region) and accommodation characteristics (mode and size class).
Number of stays: The number of times a facility offered via the platforms was occupied (i.e., individual bookings).
Number of nights (rented out): The number of nights a facility offered via the platforms was occupied.
Number of guest nights: The number of nights spent during a stay, taking into account the size of the travel party (e.g., a family of 4 staying 3 nights represents 1 stay, 3 nights rented and 12 guest nights).
Cities: Defined based on Local Administrative Units (LAU) where at least 50% of the population lives in an urban centre. The data specifically covers 'selected cities' (capital cities, cities with over 200,000 inhabitants and other cities that jointly account for 90% of the annual guest nights spent in cities in the country).
The statistical unit is the accommodation facility for short-term accommodation offered on the online platforms.
The statistical population consists of all short-stay accommodations (NACE 55.2) located in the reporting countries that are actively listed and booked via the participating online platforms (Airbnb, Booking and Expedia).
European Union Member States and EFTA countries.
The reference periods are the calendar month (for monthly datasets) and the calendar year (for annual datasets).
The overall accuracy is high. Because the data is collected directly from the platforms' databases, it represents a complete census of their transactions. Consequently, there is no sampling error. In terms of occupancy data, double-counting is highly unlikely, as a specific transaction for a given stay is recorded by only one platform.
Data are expressed in absolute values (number of stays, number of nights, number of guest nights).
Valid files from the individual platforms are aggregated by Eurostat to create single master files. This aggregation process is crucial for statistical disclosure control; it ensures that the outputs do not reveal the individual market share or proprietary data of any single platform. The compiled data is then aggregated geographically (NUTS and LAU levels) and structurally (by size class, guest origin, etc.) for dissemination.
The source data consists of aggregated datasets securely transmitted directly by major international online platforms (Airbnb, Booking and Expedia).
Data are disseminated on a quarterly basis (for the monthly occupancy datasets) and on an annual basis (for detailed datasets featuring structural breakdowns). See item 8.1 - 'Release calendar'.
The platforms transmit the monthly occupancy data no later than 60 days after the end of the reference quarter. This delay is tied to the financial reporting obligations of the platforms; as publicly listed companies, they are restricted from sharing operational data prior to their official quarterly earnings releases. Following validation, Eurostat disseminates the data according to its release calendar. See item 8.1 - 'Release calendar'.
The data is highly comparable across countries, as it is based on a uniform data structure and common definitions applied by global platforms. Geolocation mapping to LAUs is performed by each platform prior to transmission, using the same official GISCO LAU shapefiles provided by Eurostat. Aggregation from LAUs to NUTS regions and cities is carried out by Eurostat.
Minor discrepancies exist regarding the attribution of guests residing in overseas territories, but this does not affect overarching comparability.
Data are generally comparable over time starting from 2018. No structural break in the time series was observed when Tripadvisor exited this market segment at the end of 2024. The data from reference year 2025 onwards reflects the activity of 3 platforms (Airbnb, Booking, Expedia) rather than 4.
