Tourism statistics, indicators and big data: a perspective article Open Access

The role of tourism data in assisting tourism planners and decision-makers in evaluating performances, setting targets and anticipating future scenarios has been widely acknowledged by scholars (Smith, 1988; Latham and Edwards, 2003; Volo, 2018). This contribution aims at the following:

• Discussing the evolution of tourism data sources spanning from the traditional collection of tourism supply and demand statistics to the exploitation of big data; and

• Presenting and debating future perspective for producers and users of tourism data.

The need for tourism supply and demand statistics is well documented. Past literature indicates some of the usages of tourism data, such as:

• governments’ evaluation of tourism dimension and its significance to the national economy;

• destinations’ forecasts of tourism arrivals; and

• industry’s decision-makers use for strategic marketing purposes (Wöber, 2000; Massieu, 2001; Volo, 2004).

The collection of tourism statistics started, in most European countries, after the Second World War, and the first systematic collection of international travel statistics was the one published by the British Travel Association in 1949 (Lickorish, 1997).Tourism also attracted scientists’ attention enabling, 75 years ago, the foundation of the first international scientific journal Tourism Review, which remains an authoritative source (Laesser et al., 2019). In 1973, the European Travel Commission issued a program for tourism statistics, soon after the World Tourism Organization proposed its guidelines. To date, the United Nation World Tourism Organization (UNWTO) publications provide technical guidance in the collection and use of tourism statistics (UNWTO, 1994, 2008). Initially, statistics were used mostly with the focus of defining tourism (Frechtling, 1976; Smith, 1988; Pine, 1992; Hannigan, 1994). Provocative contributions (Shackleford, 1980; Lickorish, 1997; Volo and Giambalvo, 2008; Frechtling and Hara, 2016) have then raised a few issues and provided some initial guidance. Methodological aspects were discussed throughout the decades (Ritchie, 1975; Latham and Edwards, 2003; Liang et al., 2012; Aroca et al., 2017) and – as the collection of tourism statistics improved – academics, enterprises and governments have paid increasing attention to the creation and use of indicators. Demand and supply simple indicators have permeated tourism literature over the past 50 years. In the past few decades, composite tourism indicators have addressed more complex issues such as competitiveness, sustainability and innovation (Gooroochurn and Sugiyarto, 2005; Volo, 2005; Pulido and Sanchez, 2009; Castellani and Sala, 2010). However, many indicators still suffer from the following:

• the deficiencies in systematically collect the necessary elementary data; and

• the difficulties in accounting for the complex nature of tourism which requires suitable operationalization and measurements of the investigated constructs (Mazanec et al., 2007; Volo, 2015; Mendola and Volo, 2017).

Thus, despite the recent developments, the contribution of composite indicators to comprehensive theoretical frameworks and the actual usage – by tourism stakeholders and operators – of the derived rankings remain often unclear and undocumented (Volo, 2018).

In the midst of these challenges, novel forms of measurements of tourism activities are emerging (Volo, 2018). Big data – namely extremely large or complex data set – have recently entered the realm of data sources for various aspects of human activities that can be digitally traced. The debate on the characteristics and exploitation of big data in tourism is quite extensive (Önder et al., 2016; Li et al., 2018; Mariani et al., 2018). Indeed, unstructured and often unofficial, big data are valuable for scholars and industries alike, as they allow to trace individual tourist’s preferences, behaviors, points of interest, movements and even expenditures (Volo, 2018). Scholars are currently exploring big data sources to:

• exploit these powerful data as measurements of tourist flows in space and time (Hallo et al., 2012; Scaglione et al., 2016);

• predict tourism demand (Bangwayo-Skeete and Skeete, 2015; Song and Liu, 2017); and

• assess their validity as data sources (Mariani and Borghi, 2018; Mariani et al., 2019).

Along the same lines, a Eurostat task force has investigated the use of big data in complementing official statistics, including tourism statistics (Eurostat, 2014). The full potential of big data is under investigation, clearly the potential to induce real-time marketing actions (Buhalis and Sinarta, 2019) is appealing to tourism stakeholders, while scholars acknowledge the need for proper data mining and ad hoc algorithms to enable accurate use of these digital traces (Scaglione et al., 2016; Mariani et al., 2018).

Fast raising opportunities to use novel data sources impose few basic considerations for their successful integration as tourism data, as evident in Figure 1.

A basic system of tourism statistics is lacking in several countries – mostly the underdeveloped – thus significant work is needed, as these countries are receiving an increasing share of international tourism. The recent developments in the creation of composite indicator for tourism will continue but will need further strengthening and integration between traditional statistics and big data. Creating shared databases and replicable methods will allow scholars across countries to apply indicators to destinations of different magnitude enabling policymakers to easily access and soundly use these indicators. Meanwhile, tourists traces have become an invaluable source of data, and the emerging smart disruptive innovations in tourism will allow even more gathering of tourism-related big data and thus the need for reciprocity and fairness in data and information exchanges will be paramount (Buhalis et al., 2019). Legal and ethical exploitation of these novel data sources ought to be investigated, and pathways for mutual beneficial usage of scholars and businesses shall be designed. The challenges of data sharing, data extraction and data analytics have been explored albeit in an incomplete and fragmented way; thus, conceptual frameworks ought to be developed to ensure theory building and enhance customization and intelligent service supply (Mariani et al., 2018). Improved data analytics will enable using big data for not only tourism online marketing, design and recommendations but also demand prediction, precaution and emergency studies (Li et al., 2018). At the twilight of traditional measurements, tourism private and public stakeholders should foresee the enormous opportunity to combine, in real-time, information obtained by tourists’ digital traces with that of tourism companies’ databases and information systems. An exciting time to come for the tourism industry.

The slow development of tourism statistics, followed by the methodological scrutiny of those interested in indicators have left space to the disruption of big data. The challenge remains however on shifting the attention from a “big” to a “smart” usage of these data, adding layers of information, facilitating real-time usage and appropriate dissemination of trends. The thoughtful merging of tourists’ digital traces with industry data, the competences of data analysts and the theoretical strengths of tourism scholars will results in a redesign of the tourism data landscape. A considerate and integrated system of tourism data (statistics, indicators and big data) shall remain a priority for scholars and practitioners alike.