Data-driven Initiatives of Destinations Supporting Sustainable Tourism

Data-driven Initiatives of Destinations

Supporting Sustainable Tourism
Completed Research

Stefan Neubig Julius Baecker

Outdooractive AG Technical University of Munich
[email protected] [email protected]
Nicolas Gehring Andreas Hein
Technical University of Munich Technical University of Munich
[email protected] [email protected]
Jörg Weking Helmut Krcmar
Technical University of Munich Technical University of Munich
[email protected] [email protected]

Abstract
Digitization has arrived at the heart of global tourism. However, due to challenges such as climate change
and overtourism, touristic destinations not only need to become more digital but also more sustainable.
Although ever-increasing quantities of touristic data provide a growing basis for innovative digital
solutions, destinations vary widely in driving sustainable, data-driven initiatives due to uncertain targets
and challenging data acquisition. Despite its rising importance, research still lacks a thorough
understanding of how destinations can use data to become more sustainable. We build on interviews with
ten destination management organizations (DMOs) to develop a taxonomy of their data-driven initiatives
for sustainable tourism to address this issue. Our findings contribute to research by bridging the gap
between available data sources and important targets of data-driven initiatives from the unique perspective
of touristic destinations. For practice, we provide a structured overview for positioning own ambitions of
DMOs.

Keywords
Sustainable Tourism, Taxonomy, Big Data, Destination Management Organizations (DMOs)

Introduction
Tourism has become increasingly globalized, resulting in increased tourist flows and the diversification of
destinations. At the same time, digitization provides an ever-increasing variety of data, often referred to as
big data, which is expected to greatly impact the tourism industry (Samara et al. 2020; Yallop and Seraphin
2020). The availability of big data offers novel opportunities to address challenges such as overtourism.
Caused by low-cost airlines, inexpensive accommodations, or overexposure due to social media,
overtourism is harmful to the environment and negatively affects tourism acceptance among local
inhabitants. Hence, overtourism is not only a growing concern of touristic destinations but also gained
attention within the research community (Capocchi et al. 2019).
Overtourism is also at the heart of the development of sustainable tourism (i.e., developing touristic
destinations while ensuring socio-economic, cultural, and environmental sustainability criteria), driven by
so-called destination management organizations (DMOs). DMOs are responsible, among other things, for
the marketing and infrastructural development of one or multiple incorporated destinations. Focusing on
sustainability, they ensure the long-term competitiveness of a destination, while taking care of the
preservation of its natural environment, its cultural heritage, as well as the needs of other involved
stakeholders (e.g., tourism-related enterprises). As a result, managing sustainable tourism is a complex

Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022 1

 Data-driven Initiatives of Destinations Supporting Sustainable Tourism

task, which includes (but is not limited to) the need of intelligent visitor guidance (e.g., promoting
alternative places to distribute visitors), service personalization, product and service optimization, as well
as infrastructure adjustments and mobility services. With the adaption of digital technology, however,
DMOs may be able to address these challenges, while, at the same time, decreasing costs, streamlining
operations, and improving productivity and efficiency (Ivanov 2019).
While previous research has mainly focused on literature and case studies to provide a better understanding
of the use of data in tourism (Li et al. 2018), this work aims at investigating data initiatives of DMOs (in the
following only referred to as data initiatives) that are currently being performed or planned and seek to
support the development of sustainable tourism. We focus our work on understanding the structure of such
initiatives, that is (i) which data is available, (ii) how it can be acquired, (iii) what insights can be derived
from it, and (iv) what use cases are involved in the sustainable development of a touristic destination. As
an illustrating example, a beach destination could exploit real-time parking lot data (available data),
collected by physical sensors (acquisition), to derive the number of beach visitors (insight). Afterwards, this
information could be used for a more effective visitor guidance, by informing arriving visitors early enough
and in an automized way, so that they can navigate to an alternative location. Therefore, to describe the
structure of such data initiatives, we aim at developing a conceptually grounded taxonomy. More precisely,
we seek to answer the following research question:
Which conceptually grounded and empirically validated characteristics structure data-driven
initiatives for sustainable tourism?
We follow a structured taxonomy development method based on qualitative interviews to reveal the
characteristics of data-driven initiatives for sustainable tourism. Thereby, we first draw on expert interviews
with ten DMOs from Germany to gain rich information on DMO initiatives. We then apply iterative cycles
for taxonomy development using qualitative coding techniques to develop the desired taxonomy for
structuring data-driven initiatives for sustainable tourism. The resulting taxonomy contributes to research
by bridging the gap between available and needed data sources and corresponding targets of data-driven
DMO initiatives. We increase the understanding on the current state and planned ambitions of such
initiatives. DMOs can use our findings for positioning own initiatives or planning new ambitions
systematically. Our illustrative case insights further support practical applicability.

Related Work
As in many research fields, technologies like big data and artificial intelligence (AI) are seen as key driver
in the current and future of tourism. Especially the application of big data and AI technologies in different
fields have been addressed extensively in previous research (Ivanov 2019). Concerning overtourism and
associated visitor guidance, tourism demand forecasting is of great relevance. Song et al. (2019) review 211
publications between 1968 and 2018 to investigate how technologies for tourism demand forecasting
changed over time. They conclude that despite proposed forecasting models have become more complex
and also have improved largely with respect to generalization, there is still no single method that can be
applied for all kinds of situations, thus, research is still ongoing. As a sidenote, however, they also mention
that previous research mainly focusses on international tourism demand, while there is little research with
respect to domestic tourism. A different area of data use in tourism is sentiment analysis, that is, processing
free text by the means of natural language processing to extract sentiment scores. Alaei et al. (2019) review
existing approaches for sentiment analysis and emphasize its importance based on the grown volume of
data, which makes it impossible to review online feedback manually. Thelwall (2019) even considers
sentiment analysis a key component of tourism big data research and illustrates its value based on a range
of different examples from social media content. Another popular application of AI in tourism are chatbots
and smart assistants. Fensel et al. (2020) leverage integrated touristic data and describe how knowledge
graph technology can be used in order to create such technologies. Another direction of AI in tourism that
has been addressed are robots, however, most of the existing studies are purely conceptual (Ivanov 2019).
Besides investigating the application of concrete technologies, previous research has also addresses the
consequences of big data and AI in tourism at a more general level. Yallop and Seraphin (2020) review
opportunities and risks regarding the use of big data in tourism and hospitality organizations. After
investigating both academic and industry databases and reports, they conclude that big data can be
regarded as a beneficial technology, however, comes at the expense of certain risks, including ethical,

Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022 2

 Data-driven Initiatives of Destinations Supporting Sustainable Tourism

privacy, and security concerns. Among other things, they highlight the beneficial use of big data
technologies for revenue management and exemplify the use and combination of internal and external data
in this context. Samara et al. (2020) review the benefits and overall role of big data and artificial intelligence
in the tourism domain. In contrast to Yallop and Seraphin (2020), however, they do not focus on
disadvantages or risks, but on the ability of data-driven approaches to overcome challenges and highlight
the importance of adopting such technologies to stay competitive. Ivanov (2019) focusses on the impact of
robots, artificial intelligence and service automation and emphasizes the emergence of two large segments,
that is, high-tech tourism companies that are mainly offering robot-delivered services and high-touch
companies mainly relying on human interaction. Despite of the mentioned research, however, there are still
plenty of research gaps in the topic of understanding the impact of big data and AI on companies in the
tourism domain, especially when it comes to the topic of sustainability (Ivanov 2019).
Moreover, sharing data across the borders of touristic destinations (both domestically and across national
borders) has been identified as an important research direction. Open data initiatives have emerged and
address the need of exchanging data among different DMOs and service providers. Şimşek et al. (2019)
describe how they adapt the well-known schema.org ontology to create more expressive semantic data
models for the tourism domain. Besides, the AlpineBits Alliance (2019) is an initiative located in South
Tyrol that aims at creating a set of standards specifically targeting the alpine areas. Within the scope of the
BayernCloud project, a multi-domain reference architecture (Baecker et al. 2019) has been developed,
targeting the exchange of (open) data in its touristic fragment.
While research on how (big) data can be used to sustain the innovation process of tourism destinations has
in general received little consideration (Del Vecchio et al. 2018), this particularely holds for the on-site
perspective of individual DMOs. For this purpose, we introduce a taxonomy that considers the specifics of
data initiatives performed by individual DMOs taking into account the nature of available data, the on-site
data collection, the associated insights, and the specific targets, DMOs seek to achieve.

Research Methodology
This study follows a two-stage research approach. First, our data collection draws on qualitative interviews
with representatives from a total of ten DMOs. Second, we develop a taxonomy in iterative cycles based on
the guidelines proposed by Nickerson et al. (2013).

Data Collection
For data collection, we conducted a total of ten qualitative interviews with German DMOs to gain rich
information on data-driven initiatives of touristic destinations (see Table 1). We selected our interview
partners to cover a wide range of touristic focuses (Hiking (H), Biking (B), Relaxing (R), Swimming (S),
Winter sports (WS), Sightseeing (SI)) and environmental properties (Alpine (A), Lakes (L), Nature (N),
Historical buildings (HB), Sea and Beach (SB)) distributed all over Germany. Our focus on qualitative case
data enables the generation of novel theories in this little-examined field by providing deep insights into
the topic under investigation (Eisenhardt 1989). Therefore, we applied semi-structured interviews to
increase the explorative character of the interviews in our qualitative data collection (Myers and Newman
2007). The interviews focused on data initiatives of the DMOs that are currently running as well as
initiatives that are planned, but also expert assessments of what should be done in the future. To avoid
active listening (McCracken 1988), we phrased the questions in an unobtrusive and non-directive manner.
Furthermore, we conducted all interviews in the respondent’s language to avoid ambiguity of language
(Myers and Newman 2007).
We divided all interviews into five parts: (1) a personal introduction of the interviewer and organizational
matters; (2) personal introduction of the interviewee including the current position in the DMO; (3)
questions on the status quo on the use of data and analysis in the DMO; (4) questions on the use cases of
data and analysis in the DMO; (5) questions on further potentials relating to the use of data and where there
is a lack of data and how the yet non-existing information could be used. The open questions centered on
understanding how DMOs already use data and analysis to assist DMO initiatives, what further potential
lies in the application of data and analysis, and which challenges concern a lack of data and information for
DMOs. Alongside the interviews, publicly available information from the web presences of the DMOs
provided additional insights and supported data triangulation (Yin 2003).

Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022 3

 Data-driven Initiatives of Destinations Supporting Sustainable Tourism

Position Touristic Focus of DMO Interview

ID
H B R S WS SI A L N HB SB Duration
IP-1 Deputy Managing Director X X X X X X X X X X 46 min
IP-2 Project Manager X X 48 min
Divisional Management “sustainable
IP-3 X X X X X 32 min
recreation and tourism”
Digital Marketing - Systems,
IP-4 X X X X X X X X X X 62 min
Campaigns
Authorized officer, Head of Strategic
IP-5 X X X X X X X X X X 48 min
Development
Product Management Information and
IP-6 Communication Technology / Online X X X X X X X X X 61 min
Marketing
Deputy manager at a Tourist
IP-7 X X X X X X X 55 min
Information
Managing Director, responsible for
IP-8 finances, personnel, training, projects, X X X X X X 47 min
public relations
IP-9 Tourism Director X X X X X X X X 45 min
Deputy Board Member Online
IP-10 X X X X X X X X 73 min
Marketing

Table 1. Overview of Interview Partners (IP)

Data Analysis
For data analysis, we applied the taxonomy development approach proposed by Nickerson et al. (2013)
based on standards for qualitative research (Mayring and Fenzl 2014; Miles and Huberman 1994).
Following Nickerson et al. (2013), we describe a taxonomy 𝑇 as a set of 𝑛 dimensions 𝐷𝑖 (1 ≤ 𝑖 ≤ 𝑛), each
of which consisting of 𝑚𝑖 ≥ 2 characteristics 𝐶𝑖𝑗 (2 ≤ 𝑗 ≤ 𝑚𝑖 ). To acknowledge the complex nature of data-
driven initiatives (Hunke et al. 2019), however, we allow the selection of multiple characteristics
𝐶𝑖𝑗 , 𝐶𝑖𝑘 (𝑗 ≠ 𝑘) within a single dimension 𝐷𝑖 , since data-related dimensions tend to occur in combined forms
in practice (Baecker et al. 2021). In addition, this modification ensures to keep the taxonomy both concise
and comprehensive, which are two key attributes of useful taxonomies (Nickerson et al. 2013).
For taxonomy development, we first specified a meta characteristic that serves “as the basis for the choice
of characteristics in the taxonomy” (Nickerson et al. 2013). We, therefore, chose as our meta-characteristic
the “data requirements and analysis objectives of existing and prospective data-driven initiatives of tourism
destinations”. Second, we agreed on the eight objective and five subjective ending conditions for the
taxonomy development process suggested by Nickerson et al. (2013). Based on the interview transcripts,
we applied empirical-to-conceptual cycles with our DMO cases. We conducted a structured analysis for
qualitative data following our chosen meta-characteristic to code the case information into common
characteristics (Nickerson et al. 2013). Open and axial coding supported the analysis of our transcripts (see
Table 2) to identify the characteristics and dimensions of the taxonomy (Strauss and Corbin 1990). We
iteratively classified the instances to the current state of the taxonomy and added further dimensions and
characteristics until all cases were included. Finally, we checked the ending conditions by Nickerson et al.
(2013). We concluded on 15 dimensions and 65 characteristics, which we further aggregated into four meta-
dimensions for readability.
Interview statement Open Code Axial Code
“[It is] in quite a lot of destination really just the most urgent issue: How can I Visitor Guidance Use Case Targets
direct visitors? How can I steer visitors? And that's where we are really at the
very beginning in terms of the data we need for this”. (IP-4)

Table 2. Illustration of the Coding Process

Results
The resulting taxonomy of our work is illustrated in Table 1. Each line 𝑖 corresponds to a single dimension
𝐷𝑖 , while each column 𝑗 describes the respective characteristic 𝐶𝑖𝑗 . Four meta dimensions (Available Data,
Data Acquisition, Gained Insights, and Use Cases) have been added for better readability.

Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022 4

 Data-driven Initiatives of Destinations Supporting Sustainable Tourism

Dimension Characteristics

Volatility Static Data Dynamic Data

Available Data

Origin Internal External Public Open

Touristic User Destination Count & Nature &
Categories Customer Social Data
Entities Interaction Status Movement Environment
Format Unstructured Semi-Structured Structured

Automatization Manual Semi-Automatic Automatic

Data Acquisition

Collection Space Virtual Space Physical Space

Access
Collection Object Feedback Subscription Booking Service Sensor Tracker
Control
Cost Factor Free Personnel Technology Data

Perspective Online Offline

Gained Insights

Unit Individual Group Location Time

Type Qualitative Quantitative

(Movement)
Reference Visitor Motivation Impact Mood Occupancy Distribution
Pattern

Approach Informing Sharing Monitoring Forecasting Benchmarking

Stakeholder Visitor Local Inhabitant Service Provider DMO

Use Cases

Visitor Information Tourism Acceptance Product & Service Optimization

Target Visitor Guidance Improved Marketing Infrastructure Adjustment

Efficiency Gain Personalization Collective Growth

Table 1. Taxonomy to classify data initiatives of DMOs

In the following, we briefly summarize each (meta) dimension separately. For the sake of a better
understanding, we underpin selected dimensions with exemplary quotes from the conducted interviews,
which we translated into English.

Available Data
𝑫𝟏 : Volatility: Data types relating to tourism initiatives of DMOs can roughly be characterized as static or
dynamic. Static data refers to data that is unlikely to change often (e.g., existing points of interest), while
dynamic data refers to data that is subject to frequent changes (e.g., the number of people at a certain place).
𝑫𝟐 : Origin: During our interviews, we identified four levels of data origin. Internal data refers to data that
is available to the DMO performing the classified initiative only, that is, data that is owned by this DMO
(e.g., web analytics data from its website). In contrast, external data is obtained from other organizations,
including service providers or other DMOs (e.g., by voluntary data sharing). This data origin is important
for two reasons: first, DMOs may not have direct access to data, which resides within the scope of larger
apps. IP-9 notes that especially outdoor and fitness tracking apps are great data suppliers. IP-4 adds,
The visitor will not have our app installed. We’re not fooling ourselves, we’re too small.
A second yet evenly important reason for using external data is that most DMOs do not have the necessary
expertise in its maintenance. IP-1 states that data should always be maintained by its specific experts and
underlines the need for creating and maintaining data points at their origin region. Besides the more generic
characteristics of internal and external data, we distinguish further between data from public authorities

Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022 5

 Data-driven Initiatives of Destinations Supporting Sustainable Tourism

and open data, which we consider a special case of external data. While data of public authorities (e.g.,
official booking numbers of a destination) can be obtained by DMOs directly, it is not available for other
organizations or individuals. In contrast, open data refers to data published under an open license, that is
it can be obtained and used by everyone.
𝑫𝟑 : Categories: Besides volatility and origin, we also distinguish between different touristic data
categories. The sum of all data that is available within this dimension can be seen as the digital
representation of a destination. Characteristics of this dimension (i.e., categories) refer to different groups
of data types, namely, touristic entities (e.g., events, points of interest), user interaction (e.g., web tracking),
destination status (e.g., infrastructure, construction sites), count and movement (e.g., individual or
aggregated visitor positions, frequented ways), customer (e.g., reviews, ratings), nature and environment
(e.g., forestry data, water temperature), and social data (e.g., public holidays).
𝑫𝟒 : Format: A rather technical yet important distinction of data is given by its format. More precisely, data
can be unstructured (e.g., textbooks), semi-structured (e.g., data on websites), or structured (e.g., data
originating from APIs or databases). Depending on its format, data must be processed differently. Together
with the origin (𝐷2), this dimension indicates the overall variety of the available data.

Data Acquisition

𝑫𝟓 : Automatization: In this dimension, initiatives are classified with respect to the effort of their data
collection part. Manual data collection processes are completely performed by humans and can be time-
consuming and tedious (e.g., counting visitors manually at a specific place). In contrast, semi-automatic
approaches refer to data collection processes that involve both human interaction and some degree of
automatization (e.g., a sensor that collects data automatically but requires manual synchronization to a
computer). On the other hand, automatic approaches do not require any kind of human interaction (e.g.,
sensors that are fully connected to a cloud). While automatization certainly reduces the effort of data
collection, IP-2 admits that gathering data about guests has mainly been done based on surveys and online
questionnaires. Other DMO representatives add that even data that already exists digitally is collected
manually (e.g., approximates of visitor counts from webcams, car traffic from online maps applications).
𝑫𝟔 : Collection Space: Data can be collected in both the virtual and the physical space. The virtual space
includes data collection methods within digital services such as website analytics. Accordingly, physical
space methods comprise all data collection methods that collect data in the physical world. As a result of
our interviews, data collection methods in the virtual space seem to be more widespread nowadays, even
for destinations that also use data collection methods in the physical space. As an example, IP-10 notes that,
despite using quite progressive methods to collect data within the physical space, they still consider their
web analytic data as the linchpin of their data acquisition. It is important to note, however, that collection
methods in the virtual space according to our definition are not limited to collect data about virtual objects.
One prominent example is given by digital platforms of the outdoor and fitness domains that allow to collect
both data about virtual customer behavior (e.g., web analytics) as well as data about physical movement
patterns (e.g., customer-recorded tracks).
𝑫𝟕 : Collection Object: While the collection space (represented by 𝐷6) refers to the space where data is
collected, the collection object refers to the object that performs the actual data collection. Examples that
have already been mentioned include (online and offline) services or sensors. Other collection objects
include bookings that usually contain a lot of demographic data.
So I know exactly who the guest is. Family of four with two children, including their ages,
all of this data is known. 4-star hotel, half-pension, for example. (IP-4)
Besides, other collection object characteristics include visitor feedback (e.g., based on surveys after a stay
(IP-8)), subscriptions (e.g., apps but also newsletters (IP-10)), trackers, which involves both online (e.g., on
websites) and offline trackers (e.g., WiFi and Bluetooth beacon trackers (IP-6)), as well as access controls.
One prominent example for visitor data that can be captured by the means of access controls is a guest card,
which allows creating movement profiles (IP-9).
𝑫𝟖 : Cost Factor: Costs of data acquisition can involve several types. Although some destinations report
that volunteers help to collect data (e.g., about infrastructural conditions (IP-9)), the acquisition of data
usually involves further costs. Even when relying on internal or open data only, costs usually include

Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022 6

 Data-driven Initiatives of Destinations Supporting Sustainable Tourism

technology costs (e.g., purchasing and maintaining hardware) and salary expenses, as collected data must
be integrated into the existing systems, which can involve serious effort (see automatization (𝐷5)). Using
external data, however, can lead to significant data fees.

Gained Insights

𝑫𝟗 : Perspective: While the collection space dimension (𝐷6) explains where a data collection activity is
performed (virtual or physical space), this dimension refers to whether the generated insights from the
collected data describe online or offline states. To clarify, data can be collected in the virtual space (e.g.,
outdoor and fitness apps) but still refer to tours that were conducted by visitors offline. In the same fashion,
surveys performed within the physical space can ask questions regarding online behavior (e.g., the use of
outdoor and fitness apps).
𝑫𝟏𝟎 : Unit: Gained insights may refer to individuals, groups of people, locations, or time. This is important
to differentiate as we can directly conclude the kind of granularity under investigation. Insights referring to
individuals contain personal information and thus can be used for personalization use cases. On the other
hand, insights about groups can help to optimize services and products ( 𝐷15 ), as it rather captures
knowledge about aggregated guest needs. Two other important characteristics of this dimension are
location and time. As an example, the occupancy of locations usually depends on the time (e.g., day of week).
𝑫𝟏𝟏 : Type: Besides the perspective and unit, insights can be further distinguished between qualitative and
quantitative insights. While qualitative insights may contain information such as the motivation of visitors,
examples for quantitative insights include occupancies of certain locations and (statistical) distribution of
visitors. One conclusion of our interviews is that qualitative data seems harder to track than quantitative
data, especially, when it comes to day tourists (although this group is generally harder to track than
overnight guests (IP-7)). IP-10 notes,
As far as day tourists are concerned, no we actually have no way to get qualitative data.
𝑫𝟏𝟐 : Reference: The reference dimension captures possible objects and properties gained insights may
relate to. Insights may refer to visitors (both individuals and groups), their (online and offline) movement
patterns, (statistical) distributions or occupancies, as we already outlined in several examples above.
However, depending on what data is acquired, obtained insights can also relate to the motivations or intents
of visitors. IP-3 underlines the importance of visitor’s motivations, asking
Why is one chosen? Are you elected because you are located inland or is it a conscious vacation decision?
Other characteristics to classify insights include mood (e.g., mood of visitors or local inhabitants) and
impact (i.e., getting insights about the impact of previously performed actions, such as marketing
campaigns).

Use Cases
𝑫𝟏𝟑 : Approach: Use cases as identified during our interviews involve different actions that our taxonomy
characterizes in the approach dimension. Informing refers to the action of passing information onto a
stakeholder (𝐷14 ), sharing refers to the action of sharing data among different stakeholders. In the data
origin dimension (𝐷2), we already highlighted the importance of sharing data. Besides the aforementioned
reasons of availability and data expertise, one important use case is reducing the maintenance effort that
may originate from doubled work, such as the redundant creation of points of interest. As I8 notes,

The vision is basically that you only have to create a POI once. if I now want to make use of this dataset
on my website, then I take it when it is free there. (I8)

This points out the importance of open data that is approached by different cross-organizational initiatives
and was introduced in the Background section. However, also smaller cross-organizational initiatives
turned out to be helpful. IP-1 describes a content network in their region consisting of 50 co-located
websites that share data with each other. Besides informing and sharing, monitoring refers to the action of
describing a destination’s current state (e.g., with respect to the occupancy of places), while forecasting

Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022 7

 Data-driven Initiatives of Destinations Supporting Sustainable Tourism

refers to predicting a destination’s future state. Finally, benchmarking refers to the comparison of different
stakeholders (e.g., the performance of two DMOs).
𝑫𝟏𝟒 : Stakeholder: Next to the approach dimension, the stakeholder dimension explains what
stakeholders are involved in the planned use cases. According to our interviews, stakeholders involve
visitors, local inhabitants, touristic service providers, and other DMOs.
𝑫𝟏𝟓 : Target: Following the results of the conducted interviews, we identified nine targets. By informing
visitors (visitor information), guests can be sensitized to the specific conditions of a certain region, such as
the animals that live there and the behaviors that are necessary to protect them. Tourism acceptance is an
important touristic challenge that appeared as a consequence of overtourism and refers to the mood of local
inhabitants. Visitor guidance aims at distributing guests with respect to time and location and is a possible
counter measure to overtourism and a way to improve both tourism acceptance and environmental
protection. One way of implementing visitor guidance is recommending suitable alternatives that fits a
guest’s needs, thus, personalization is another important target. On a more general level, both product and
service optimization, as well as improved marketing measures are further targets and can even support
visitor guidance by targeting marketing campaigns towards less crowded places. Infrastructure adjustments
(e.g., strengthening public transport) can help reduce car traffic in favor of more sustainable transportation
means. Efficiency gain (e.g., automatization of data collection) frees personnel and allows to focus more on
the sustainability-related tasks described above. Finally, collective growth (e.g., cross-DMO benchmarking)
fosters learning processes and allows collaboratively shaping a sustainable future together.

Discussion
One compelling result of our interviews is that many DMOs, while being convinced about the usefulness of
the use of data to support the development of their destination, are still struggling in one or multiple stages
of their data initiatives. Challenges range from knowledge about which data to collect and how, its
processing to derive appropriate insights, as well as concrete use cases that really help them improving their
destination in a sustainable way. Moreover, data initiatives as discussed within this paper are quite
complex. They differ from traditional business intelligence processes in that they need to address particular
sustainability issues with a wide range of conflicting objectives (e.g., revenue as well as environmental
protection) and stakeholders (e.g., visitors, local inhabitants as well as touristic enterprises). Furthermore,
they must be considered on a cross-destination basis, as many of the underlying challenges can only be
solved through cooperation (e.g., overtourism). The proposed taxonomy can be used to classify data
initiatives of DMOs and is helpful at any stage, including planning, performing, and reflecting in
retrospective, as it incorporates the challenges mentioned above and sheds light on the data types,
acquisition methods, insights and use cases that may help to guide a destination’s development in a more
sustainable direction.
One important finding is the relevance of economic factors, even when targeting other sustainability goals
as main targets, which is also reflected by the target dimension (𝐷15 ). On the one side, economic success
can be seen as a precondition as data purchases and data collection technology are considerably expensive,
as mentioned by IP-9 and IP-10, respectively. If these investments did not pay off economically, probably
few DMOs would take this step. On the other side, however, both may even go hand in hand. As an
illustrating example, IP-10 describes how they use humans to manually count visitors and send the results
via text messaging to administrators who, in turn, add this data to the destination’s website. On the other
side, IP-10 also reports that they are using advanced technologies like sensors and image recognition
technology to collect visitor numbers. They are also planning to continue their automatization efforts, which
is likely to result in both a larger and better data base that can be used for improved visitor guidance as well
as reduced personnel costs and a gain in efficiency. Furthermore, the more stable a data base is the better
DMOs can address other economic goals, including improved marketing and service optimization.

Contribution
Investigating the data-driven initiatives of DMOs for promoting sustainable tourism has implications for
theory and practice. Our developed taxonomy contributes to research by strengthening the important
perspective of DMOs within the scientific discourse on the development of sustainable tourism. We thereby
provide details on available data, data acquisition, analytical insights, and use cases of data-driven DMO

Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022 8

 Data-driven Initiatives of Destinations Supporting Sustainable Tourism

initiatives which create a novel view in the digital development of touristic destinations. Thus, our taxonomy
paves the way for addressing important future directions in the context of sustainable tourism development
of DMOs.
For practice, our findings provide valuable orientation for DMOs. First, the taxonomy helps DMOs to
position the status quo of their data-driven initiatives in the light of other destinations. It further inspires
potential use cases on which to increase focus. Finally, the taxonomy supports practical exchange by
providing a structured overview of relevant characteristics as well as a consistent terminology.

Limitations
Like other studies, our work has some limitations. As of the nature of taxonomies, no taxonomy is ever
complete or perfect. Although we applied rigorous ending conditions as proposed by Nickerson et al. (2013),
it is still possible that more dimensions or characteristics pop up when including additional DMOs.
Furthermore, our study focuses on DMOs from Germany. While we address a variety of touristic focuses
(such as hiking, swimming, sightseeing, or winter sports) and environmental properties (e.g., alpine, lakes,
nature, beach), future research may add the perspective of destinations from other countries and cultures.

Avenues for Future Research

During our study, we also identified fruitful avenues for future research. First, as a consequence of our
limitation to the German tourism area, we welcome any contribution to our taxonomy that adds the
perspectives of other countries and cultures to create evidence on the transfer of the taxonomy to other
touristic destinations. Second, the conducted interviews shed also light on aspects that are worth
investigating, including challenges that are involved in data initiatives. Many DMOs reported that there
exist serious problems with accessing data that is principally available.
I'll say the data is there. But they are not yet accessible to us (IP-1).
These data are available. Getting at it is certainly difficult. (IP-2).
This can be due to the costs of proprietary data or because the needed data is not sold by its owner (e.g.,
traffic from online maps). The same applies to data that could potentially be collected but requires
expensive hardware that DMOs are reluctant to buy. We consider reasoning over such associated challenges
an important direction for future research. Third, as of the nature of taxonomies, they are limited to specific
classification tasks and thus are unable to express certain relationships within the described dimensions
and characteristics, such as – in our case – the relation between data types and ways to acquire that data,
or the relation between these data types and the targets, which may help DMOs to decide which data to
collect. Making this decision is a serious challenge for DMOs as we found during our interviews. Therefore,
we conclude that future research should address this challenge and propose our taxonomy as a basis for
quantitative analysis and classification of data initiatives of DMOs to detect interrelationships as mentioned
above. Such an investigation could result in an expressive ontology (Guarino et al. 2009) which uses our
taxonomy at its core (Nickerson et al. 2013).

Conclusion
In this study, we developed a taxonomy that identifies and classifies data-driven initiatives of DMOs
targeting the development of sustainable tourism. Based on interviews with ten DMOs, we derived a total
of 15 dimensions and 65 characteristics that increase the understanding of the current state and future
ambitions of DMOs in this global field of action. Our taxonomy offers a generalized structure incorporating
a variety of different DMO cases and expert assessments on existing and prospective data-driven initiatives
of tourism destinations. We, therefore, see fruitful avenues for DMOs in the development of data-driven
initiatives to advance sustainable tourism.

Acknowledgements
This work is a result of a cooperation between Outdooractive AG and the Technical University of Munich.
The publishing of this study was supported by the AIR public research project (67KI21005B), funded by the
federal ministry for the environment, nature conversation, nuclear safety, and consumer protection.

Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022 9

 Data-driven Initiatives of Destinations Supporting Sustainable Tourism

REFERENCES
Alaei, A. R., Becken, S., and Stantic, B. 2019. “Sentiment Analysis in Tourism: Capitalizing on Big Data,”
Journal of Travel Research (58:2), pp. 175–191.
AlpineBits Alliance. 2019. “AlpineBits. Open Data Exchange in the Alpine Tourism.”
(https://www.alpinebits.org, accessed August 30, 2019).
Baecker, J., Böttcher, T. P., and Weking, J. 2021. “How Companies Create Value From Data–A Taxonomy
on Data, Approaches, and Resulting Business Value,” Twenty-Ninth European Conference on
Information Systems (ECIS 2021).
Baecker, J., Engert, M., Müller, J., Neubig, S., Schaffer, N., Kunz, I., Schneider, A., Krcmar, H., and Eckert,
C. 2019. “BayernCloud Gesamtreferenzarchitektur.”
Capocchi, A., Vallone, C., Pierotti, M., and Amaduzzi, A. 2019. “Overtourism: A Literature Review to Assess
Implications and Future Perspectives,” Sustainability (11:12), p. 3303.
Del Vecchio, P., Mele, G., Ndou, V., and Secundo, G. 2018. “Open Innovation and Social Big Data for
Sustainability: Evidence from the Tourism Industry,” Sustainability (10:9), Multidisciplinary Digital
Publishing Institute, p. 3215.
Eisenhardt, K. M. 1989. “Building Theories from Case Study Research,” Academy of Management Review,
pp. 532–550.
Fensel, D., Şimşek, U., Angele, K., Huaman, E., Kärle, E., Panasiuk, O., Toma, I., Umbrich, J., and Wahler,
A. 2020. “Why We Need Knowledge Graphs: Applications,” in Knowledge Graphs: Methodology, Tools
and Selected Use Cases, D. Fensel, U. Şimşek, K. Angele, E. Huaman, E. Kärle, O. Panasiuk, I. Toma, J.
Umbrich, and A. Wahler (eds.), Cham: Springer International Publishing, pp. 95–112.
Guarino, N., Oberle, D., and Staab, S. 2009. “What Is an Ontology?,” in Handbook on Ontologies, S. Staab
and R. Studer (eds.), Berlin, Heidelberg: Springer Berlin Heidelberg, pp. 1–17.
Hunke, F., Engel, C., Schüritz, R., and Ebel, P. 2019. “Understanding the Anatomy of Analytics-Based
Services – A Taxonomy to Conceptualize the Use of Data and Analytics in Services,” in ECIS 2019
Proceedings . 27th European Conference on Information Systems (ECIS), Stockholm & Uppsala,
Sweden, June 8-14, 2019. Research Papers, Stockholm/Uppsala: AIS eLibrary (AISeL), Paper:-25.
Ivanov, S. 2019. “Ultimate Transformation: How Will Automation Technologies Disrupt the Travel,
Tourism and Hospitality Industries?,” Zeitschrift Für Tourismuswissenschaft (11:1), pp. 25–43.
Li, J., Xu, L., Tang, L., Wang, S., and Li, L. 2018. “Big Data in Tourism Research: A Literature Review,”
Tourism Management (68), pp. 301–323.
Mayring, P., and Fenzl, T. 2014. “Qualitative Inhaltsanalyse,” in Handbuch Methoden Der Empirischen
Sozialforschung, N. Baur and J. Blasius (eds.), Wiesbaden: Springer Fachmedien Wiesbaden, pp. 543–
556.
McCracken, G. 1988. The Long Interview, Sage.
Miles, M. B., and Huberman, A. M. 1994. Qualitative Data Analysis: An Expanded Sourcebook, sage.
Myers, M. D., and Newman, M. 2007. “The Qualitative Interview in IS Research: Examining the Craft,”
Information and Organization, pp. 2–26.
Nickerson, R. C., Varshney, U., and Muntermann, J. 2013. “A Method for Taxonomy Development and Its
Application in Information Systems,” European Journal of Information Systems (22:3), pp. 336–359.
Samara, D., Magnisalis, I., and Peristeras, V. 2020. “Artificial Intelligence and Big Data in Tourism: A
Systematic Literature Review,” Journal of Hospitality and Tourism Technology (11:2), Emerald
Publishing Limited, pp. 343–367.
Şimşek, U., Angele, K., Kärle, E., Panasiuk, O., and Fensel, D. 2019. A Formal Approach for Customization
of Schema.Org Based on SHACL, pp. 1–16.
Song, H., Qiu, R. T. R., and Park, J. 2019. “A Review of Research on Tourism Demand Forecasting:
Launching the Annals of Tourism Research Curated Collection on Tourism Demand Forecasting,”
Annals of Tourism Research (75), pp. 338–362.
Strauss, A., and Corbin, J. 1990. Basics of Qualitative Research, Sage publications.
Thelwall, M. 2019. “Sentiment Analysis for Tourism,” in Big Data and Innovation in Tourism, Travel, and
Hospitality, M. Sigala, R. Rahimi, and M. Thelwall (eds.), Singapore: Springer Singapore, pp. 87–104.
Yallop, A., and Seraphin, H. 2020. “Big Data and Analytics in Tourism and Hospitality: Opportunities and
Risks,” Journal of Tourism Futures (6:3), Emerald Publishing Limited, pp. 257–262.
Yin, R. 2003. “Case Study Research: Design and Methods,” Sage Publications, Inc, p. 11.

Twenty-eighth Americas Conference on Information Systems, Minneapolis, 2022 10