This case study explores how Lithuania Travel leverages mobile positioning data to analyse travel patterns and indicators of international visitors in Lithuania. Launched in June 2022, the Mobile Data Project created three dashboards showcasing statistical indicators based on anonymised data from a mobile network operator. These indicators provide detailed insights into the mobility patterns of inbound travellers, complementing traditional accommodation statistics and surveys. The project aims to offer a more granular understanding of tourist behaviour, aiding decision-makers, tourism service providers, and event organisers. The data, updated thrice yearly, covers various geographical levels and is publicly accessible on the Lithuania Travel website.
Using mobile positioning data to provide information on travel patterns in Lithuania
Abstract
The Mobile Data Project, initiated by the public enterprise Lithuania Travel, is a big data initiative that provides information on travel patterns and indicators of international visitors in Lithuania. Inbound tourism indicators based on mobile positioning data were introduced in June 2022. As a result of the project, three dashboards were created, featuring statistical indicators that describe the mobility patterns of inbound travellers visiting Lithuania and are based on passive mobile positioning data.
The aim of the project is to have an alternative data source, in addition to the traditional accommodation statistics and surveys, to provide the missing aspects of knowledge about inbound tourist travels in Lithuania. Mobile data can be presented in more detail (geographical and timely granularity), thereby creating greater added value for the tourism sector.
The statistical indicators, which represent generalised and depersonalised travelling patterns of inbound visitors in Lithuania, are based on one mobile network operator’s anonymised data. During the pilot project, big data analytics was applied to the mobile network operator’s data to calculate, extrapolate and produce statistical indicators. By incorporating additional data, including official accommodation statistics, land use and road network data, and other relevant geographic information, a mathematical model was developed to calculate summarised indicators for all inbound visitors' journeys in Lithuania. The pilot project was carried out by the public authority Lithuania Travel in collaboration with the companies Tele2, Positium and PositIN. More information about each of the companies is provided further below.
Several dashboards are publicly available on the lithuania.travel website without login or software:
Heatmap, showing the movement patterns of the inbound travellers in Lithuania.
Dashboard for indicator comparison (the second dashboard on the page, currently in Lithuanian only, the English version is coming soon. The indicators are described in more detail further below).
Municipality (county) level indicators, such as number of unique visitors, number of stays, number of trips, number of distinct calendar days, number of same-day visitors, and others.
There are multiple target audiences, including local municipalities and national government (decision makers), tourism service providers and event / entertainment organisers. The different types of data can broadly be attributed to different stakeholders:
Indicators for municipalities, counties by month and country of origin: for decision makers and those who need a broader overview of the trends (local municipalities and national government or larger business chains planning investments).
Daily number of visitors: for event or festival organisers who need more detailed daily data.
Number of visitors at tourism objects (heatmap): for those, who make decisions based on tourist activity in the area (for example tourism service providers) or are involved in planning the development of tourist facilities or infrastructure renewal (local governments).
The indicators are provided in four levels of granularity: the whole of Lithuania, 10 counties, 60 municipalities and an adaptive data grid (various sizes, depending on the data characteristics) that is visualised as a heatmap.
The main needs that were expected to be met in the pilot project were as follows:
More detailed and accurate data about the number of tourists, including those that are not staying at the accommodation establishments (for example, visiting friends or relatives, in the unregistered short-term rentals, staying in the nature, etc.).
Information about the most visited locations and tourism objects.
Up-to-date and nearly real-time information, expected to be accessible earlier than the data from the Statistics Department, which, at that time, typically took several months for a reference period data to be published. This aspect was not realised in the current project due to the high costs involved to implement automatic data renewal.
No administrative burden to the tourism businesses and tourists.
There are no immediate plans for significant upgrades, but if a possibility of additional funding would allow it, it could be considered to make more frequent (or even automatic) updates of the data and potentially adding more indicators or filters to the data, for example the period of stay. In-depth analysis of the available aggregated data could provide answers and a deeper understanding of specific tourism phenomena, i.e., explaining the origin of country-specific behaviours and understanding the dynamics of event visitors, one-day visitors, or similar.
The implementation of the pilot project began in 2021. The first results were published on 1 June 2022. Until the end of 2024, the data was renewed annually at the end of the year, to have the most relevant summer season data as soon as possible. Starting at the beginning of 2025, the data will be renewed 3 times a year and shown on the dashboards for the continuous data period starting in April 2022. The project is expected to continue in the nearest foreseeable future. The data available covers two separate periods: the pilot for May 2021-October 2021, and a data publication for April 2022-September 2024 (and continuously adding additional 3-4 months through the year).
The owner of the project is the public entity Lithuania Travel, which plans and co-ordinates the project goals, scope, activities and funding. The indicators are calculated and analysed with methods developed by a private Estonian company Positium. To produce tourism statistics based on mobile positioning data, Positium uses proprietary software based on an open methodological framework called Positium Data Mediator. The input data are provided by a private mobile network operator Tele2, which provides the secure server for the data storage and calculation. Calculated indicators are displayed on the main dashboard, which is developed and updated by Positium and all necessary hardware and server space is provided by Tele2. Also, over the course of the project, two more dashboards were created by Lithuania Travel and a Lithuanian private company PositIN, using Power BI and ArcGIS software.
The project is funded from combined sources. The main source is the national budget allocated by the Ministry of the Economy and Innovation. The project has also received some project-based funding. It costs approximately EUR 100 000 (before tax) yearly, but the costs have varied significantly due to changes in the scope of activities in the different years, the number of months analysed and other factors.
The project’s focus is inbound travellers in Lithuania. Due to the characteristics of the data, an attempt to target solely the tourists would be limited and inaccurate, as the data does not provide information regarding the purpose of the visit. Indicators and their definitions are based on the International Recommendations for Tourism Statistics.
The mobile positioning data used in the project includes continuous data covering a period longer than 12 months. From this, two definitions could be applied:
Country of residence: the “home” country of a subscriber, where she or he has spent at least 6 months during the past 12-month period. If Lithuania is marked as an inbound visitor’s country of residence, they are removed from the data set.
Usual environment: the geographical area within which an individual conducts her or his regular routines. The usual environment is considered at the county level (10 counties) and is where a subscriber was active for at least 6 months during the past 12 months (present at least 4 days in a month). For example, if a person is attributed Vilnius County as her or his usual environment, only trips outside Vilnius County are included in the final data.
Inbound tourism statistics indicators are aggregated to county and municipality level, presented monthly or daily. For reference, monthly and daily inbound tourism statistics are also shown at the country level. The following indicators for inbound tourism are presented:
Number of unique visitors: number of unique visitors visiting a location over a period. One person visiting location A 3 times during 2 separate trips during one month, number of unique visitors = 1. This indicator helps estimate the actual number of people present during a chosen period.
Number of trips: number of trips in a location over a period. A trip is calculated from a visitor’s entry to the country until departure from the country and usually has more than one associated stay. A new trip starts if there have been no records made by the subscriber in the country for 48 hours. One person visiting location A 3 times during 2 separate trips over one month, number of trips = 2. Combined with the number of unique visitors, this indicator also enables the estimation how many repeat trips have been made (number of trips divided by the number of unique visitors).
Number of stays: number of stays present in a location over a period. A stay is a record, or several consecutive records algorithmically assumed to be made in one location. One person visiting location A 3 times for 2 separate trips over one month, number of stays = 3. Currently there is no limitation for the stay duration – even short stops are counted to enable the estimation of tourism potential. When combined with the number of unique visitors, this indicator allows the estimation of repeat visits (the number of stays divided by the number of unique visitors).
Number of one-day visitors: number of unique visitors who stay in a specific location without spending a night in this location. If a person conducts two trips in a month and both are one-day trips, this person will be counted once.
Number of first stays: number of first stays in a specific location during the trip in the period (point of the first appearance). A visitor can have only one first stay in a specific location during a single trip. This indicator helps to estimate from which parts of the country people tend to start their trips. Disclaimer: this indicator is not suited to estimate the actual border crossings.
Number of last stays: number of last stays in a specific location during the trip in the period (point of the last appearance). A visitor can have only one last stay in a specific location during a single trip. This indicator helps to estimate from which parts of the country people tend to end their trips. Disclaimer: this indicator is not suited to estimate the actual border crossings.
Number of distinct calendar days: number of distinct calendar days spent in a location over a period. When combined with the number of trips, it is possible to estimate the average number of days spent in a location during a trip (number of days divided by the number of trips).
Number of nights spent: number of nights spent over a period in a location. A spent night is assigned to a location where a person was detected to be present at 4 AM.
As the data represents a different sample and is influenced by distinct trends, such as border crossings, transit drivers, non-formal accommodation, and other types of travellers who are not classified as tourists, this project is published independently from existing national statistics to avoid user confusion. Work is done to compliment the data collection for the number of tourists in accommodation institutions. Once implemented, it will be considered whether to integrate the data, keeping in mind that it would require additional user education.
The need for more detailed and accurate data was known from discussions with stakeholders and sectoral data analysis. Additionally, changes have been made over the course of the project considering the aspects that have received most interest and feedback from stakeholders. For example, some additional ways to present data were chosen after a workshop with one of the municipalities. The workshop identified a need to conveniently compare the data over time and in different locations. Stakeholders in another seminar expressed the need to include the results in their annual report, but the municipality level map and the heatmap were not convenient for that purpose. Taking those requirements into account, an additional Power BI data dashboard was introduced.
The approach was to make the data as easy to read as possible to non-professional analysts. Mobile data has plenty of nuances and details that influence the results; therefore, it is presented as simply as possible. Some data was removed to avoid misinterpretation. For example, the number of unique visitors in adjacent areas should not be summed up, as there are some of the same people travelling between those areas. Therefore, the heatmap does not show the exact numbers, but is instead colour coded. The exact numbers can be seen in the Power BI data dashboard, which provides the numbers on a municipality, county and national level.
One recommendation would be to keep the project flexible and adaptable. As the data are new to stakeholders, questions, new interpretations and new needs arise as they learn how to use it. The possibility to add and adjust the outcomes of the project makes it possible to adapt to stakeholder needs.
Communication included training sessions (up to 2 hours), where the dashboards were introduced, and stakeholder representatives were asked to find relevant data themselves; data workshop with a group of main local stakeholders (for example from one municipality or one type of tourism services), with key representatives who would be working with data; and presentations of various content, including general information and the concrete data and results that could be relevant, in stakeholder events.
Benefits
The main benefits come from the high level of detail of the data: daily data provides a better understanding of the habits of the inbound visitors based on the country of origin. For example, the different effect of holidays, weekends, arrival of cruise ships or popular events by country of origin can be seen. Geographical granularity makes it possible to understand the travel patterns and most visited objects and locations, most used roads by visitor’s country of origin. Further, it is possible to use the data not only on national level, but to analyse and compare the travel patterns of inbound visitors within each municipality. The data are especially important in regions where traditional data has been insufficient until now.
Data on inbound visitors' movement patterns highlights the most relevant locations and enables the identification of typological visitor types. For example, visitors from certain countries concentrate in major cities, travel less, and are likely interested in urban and cultural tourism. Other visitors are inclined to travel in nature – their travel trajectories cover the entire country and are particularly dispersed.
The data covers not only the patterns of those who choose accommodation institutions, but of all the inbound visitors. Therefore, the data can be used not only for tourism planning, but also considering potential tourists and all the services for the inbound visitors. Further, the data includes information on one-day visitors and their trends which is usually challenging to obtain through traditional methods.
The mobile positioning data allows for faster processing and production of statistical indicators, supplementary and new indicators in previously unavailable magnitude, improved temporal and spatial coverage and accuracy of the data, with minor or no burden on the tourists or tourism service providers. Cost-efficiency is especially prominent when used in large quantities and for extensive data needs Possibilities exist for more detailed data analysis by using additional data (or big data) sources, such as dates of cruises arrivals or weather information.
Challenges
The data are not always accurate due to limited granularity in very specific geographical locations, for example in specific points of interest, meaning that it should primarily be used for larger areas. The location of cell towers influences the data, leading to a slight skew in the visitor activity data. As estimates are used, there is a margin of error. It is essential to consider these factors when drawing conclusions from the data.
The total cost is high in comparison to other sources, even though the cost per data point is relatively low.
No socio-demographic characteristics or indications of the purpose of the visit are available, which would be instrumental in identifying the target travellers.
The sheer volume of information poses a challenge, with limited resources available for thorough analysis. The data requires manual analysis by skilled analysts and tourism specialists, as automation alone cannot address all issues. Users also need an understanding of mobile data principles to be able to interpret the results. Some users require training to understand how to interpret the data.
Lessons learned
The project should allocate sufficient time to educate stakeholders throughout its duration. This project offers a substantial volume of data, demanding dedicated resources for comprehensive analysis. Despite simplifying the data as much as possible, there remains a significant number of questions and instances of misunderstanding. Limited comprehension of the data may lead to a lack of trust in the new information and the dismissal of results.
The data includes not only tourists, but all the inbound visitors, which allows for conclusions about the tourism situation. For instance, many transit visitors from neighbouring countries travel through Lithuania, primarily on main roads, rarely exploring nearby attractions. Despite this potential, tourism has not fully redirected these flows to adjacent infrastructure.
For further information please contact:
Jogilė Miežienė, Head of Analytics and Research Subdepartment, Lithuania Travel, jogile@lithuania.travel
This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area.
© OECD 2025
Attribution 4.0 International (CC BY 4.0)
This work is made available under the Creative Commons Attribution 4.0 International licence. By using this work, you accept to be bound by the terms of this licence (https://creativecommons.org/licenses/by/4.0/).
Attribution – you must cite the work.
Translations – you must cite the original work, identify changes to the original and add the following text: In the event of any discrepancy between the original work and the translation, only the text of original work should be considered valid.
Adaptations – you must cite the original work and add the following text: This is an adaptation of an original work by the OECD. The opinions expressed and arguments employed in this adaptation should not be reported as representing the official views of the OECD or of its Member countries.
Third-party material – the licence does not apply to third-party material in the work. If using such material, you are responsible for obtaining permission from the third party and for any claims of infringement.
You must not use the OECD logo, visual identity or cover image without express permission or suggest the OECD endorses your use of the work.
Any dispute arising under this licence shall be settled by arbitration in accordance with the Permanent Court of Arbitration (PCA) Arbitration Rules 2012. The seat of arbitration shall be Paris (France). The number of arbitrators shall be one.
This document was produced with the financial support of the European Union. The views expressed herein can in no way be taken to reflect the official opinion of the European Union.