Conference PaperPDF Available

Abstract and Figures

The digital transformation offers organizations new opportunities to expand their existing service portfolio in order to achieve competitive advantages. A popular way to create new customer value is the offer of analytics-based services (ABS)-services that apply analytical methods to data to empower customers to make better decisions and to solve complex problems. However, research still lacks to provide a profound conceptualization of this novel service type. Similarly, actionable insights on how to purposefully establish ABS in the market to enrich the service portfolio remain scarce. Our cluster analysis of 105 ABS offered by start-ups identifies four generic ABS archetypes and unveils their specific service objectives and pronounced characteristics. The findings contribute to a more profound theorizing process on ABS by providing a detailed characterization of different ABS types and a systematization regarding strategic opportunities to enrich service portfolios in practice.
Content may be subject to copyright.
15th International Conference on Wirtschaftsinformatik,
March 08-11, 2020, Potsdam, Germany
Pathways from Data to Value: Identifying Strategic
Archetypes of Analytics-Based Services
Fabian Hunke1, Stefan Seebacher1, Ronny Schüritz1, and Gerhard Satzger1
1 Karlsruhe Institute of Technology, Institute of Information Systems and Marketing,
Karlsruhe, Germany
{fabian.hunke,stefan.seebacher,ronny.schueritz,gerhard.satzger}
@kit.edu
Abstract. The digital transformation offers organizations new opportunities to
expand their existing service portfolio in order to achieve competitive
advantages. A popular way to create new customer value is the offer of analytics-
based services (ABS) services that apply analytical methods to data to empower
customers to make better decisions and to solve complex problems. However,
research still lacks to provide a profound conceptualization of this novel service
type. Similarly, actionable insights on how to purposefully establish ABS in the
market to enrich the service portfolio remain scarce. Our cluster analysis of 105
ABS offered by start-ups identifies four generic ABS archetypes and unveils their
specific service objectives and pronounced characteristics. The findings
contribute to a more profound theorizing process on ABS by providing a detailed
characterization of different ABS types and a systematization regarding strategic
opportunities to enrich service portfolios in practice.
Keywords: analytics-based services, archetypes, service portfolio, cluster
analysis.
1 Introduction
Rapid advances in information technology threaten existing product and service
portfolios, drive changes in established business strategies, but also open up new
opportunities for existing and new market participants [1, 2]. In these dynamic markets,
the ability to expand the existing service portfolio with new innovative services that
exploit technological advances has become a key focus of organizations [3, 4].
One approach is the use of data and analytics to enable new innovative services [5
7]. Analytics-based services (ABS) are a novel type of service which encompasses the
application of analytical methods (‘analytics’) to data. It aims to increase customer
value by supporting the customer to make better decisions, solve more complex
problems and ultimately reach their goals more effectively or efficiently [8]. For
example, BASF uniquely supports farmers with ABS to increase yields [9]: Using
satellite images of its customers' fields (data) and combining them with weather
simulations (analytics), it derives current vegetation indices to predict the risk of certain
plant diseases. Farmers proactively receive field-specific fertilizer recommendations
including the required dosage calculated for each field zone. Industry experts stress that
ABS hold great promise for companies to enrich their service portfolio by either
achieving or sustaining a competitive advantage [10] through a deeper integration in
the value creation process of the customer [11] or by exploiting entirely new markets
[12].
Despite the research focus that the topic has received in the academic literature [13],
it remains unclear how organizations can systematically exploit data and analytics to
expand their service portfolio [14]. Recent research suggests that a systematization of
opportunities provides insights into immature research topics. For instance, Remane et
al. [15] derive types of car sharing business models to shed light on the systematic
discovery of car sharing opportunities for organizations while Gimpel et al. [16] derive
types of FinTech start-ups to provide insights on how they leverage digital technologies
to offer innovative financial services. Thus, systematizing with the help of dedicated
service types seems a promising approach to provide organizations with actionable
insights that could pave the way for new value creation opportunities in the context of
ABS.
The objective of this study is to provide a systematization of ABS by revealing
different ABS-types. Thus, this paper addresses the following research question: What
are archetypes of analytics-based services that can be found in the market?
For that purpose, we apply archetype theory [17] to our study context to identify
archetypes of ABS, i.e. theoretical prototypes or modular service configurations that
provide a systematic perspective on the different possibilities to create customer value
with data and analytics [17, 18]. For the analysis, we build upon ABS offered by start-
ups, since start-ups are often referred to as pioneers by taking advantage of new
technology-enabled opportunities in their service offerings [19]. We use an existing
taxonomy of ABS providing their common components to classify a dataset of 105 real-
world ABS offered by start-ups. Afterwards, we perform a cluster analysis following
the established procedure suggested by Puni and Stewart [20] to identify different
clusters, i.e. groups of ABS-cases that share similar characteristics. We interpret the
resulting clusters to derive four different archetypes of ABS. Based on these results, we
consolidate the four ABS-archetypes into a strategic framework to discuss possible
pathways for organizations to establish ABS.
Our work contributes to the emerging discourse of ABS by specifying dedicated
types of ABS including their respective characteristics which lay the foundation for a
more profound conceptualization of this new type of service in future research.
Additionally, practitioners benefit from our discussion as we disclose possible
strategies for establishing ABS in the market and, therefore, respond to management
issues reported in the field of ABS [7, 14].
The paper is structured as follows. Section 2 provides an overview of the extant
literature our research builds upon. We subsequently lay out our detailed research
design in section 3, followed by the resulting findings in section 4. We discuss our
findings and implications, reflect limitations and propose opportunities for future
research in section 5, before we conclude our research.
2 Related Work
The aim of this paper is to identify ABS archetypes to provide a systematic
differentiation of possible business opportunities. Thus, two topics from the literature
are relevant to build our research upon. First, we present prior research that contributes
to the conceptualization of the use and exploitation of data and analytics in services.
Second, research investigating the classification and systematization of services in
general offers a promising basis for our research.
2.1 Analytics-Based Services
The continuously increasing amount of data, combined with advanced analytics
technologies are considered to have great potential to expand the organizations’
business portfolio by opening up opportunities to create new service offerings [10, 13,
21, 22].
Chen et al. emphasize two general practices to create new services, Data-as-a-
Service (DaaS) and Analytics-as-a-Service (AaaS) [23]. Firstly, DaaS focuses on
providing raw and aggregated content while, secondly, AaaS provides customers with
a wide range of customizable analytical methods that enables them to analyze large
amounts of data. These practices can be offered independently or in combination in
entirely new stand-alone data-driven service solutions [24].
Apart from that, data and analytics also enable the creation of added value in
combination with existing products or services. This can be achieved by providing
additional information-intensive services for the customer to enrich the value
proposition of existing services [25, 26]. Smart, connected products can be amplified
based on e.g. sensor data they generate. Smart services "wrap" meaningful product-
related insights around them based on that data to enrich the value that is offered by the
core product [2729].
In this still emerging field of research that investigates the use of data and analytics
for new service offerings, the terminology still varies (e.g., “smart services” [27], “data-
driven services” [24], “information-intensive services” [25], “data-wrapping” [28]).
The use of (sophisticated) analytics is acknowledged to play a key role in creating value
from data [30]. However, despite different conceptualizations proposed in the field, it
still needs to be further investigated how data and different kind of analytics can be
systematically combined to contribute to value in new service offerings. We refer to
services which encompass the application of analytics to data in order to create
customer value in new services either as a stand-alone solution or bundled with
existing products or services as analytics-based services.
2.2 Systematization of Analytics-Based Services
Systematization approaches, such as taxonomy development or archetype
identification, are an essential instrument for tapping into new fields of research. They
contribute to structuring pre-existing research, facilitate the positioning of new
contributions and, thus, support a profound theory-building process in a still
underdeveloped field of research [31, 32]. IS research has adopted this approach in
order to identify (managerially) useful generalizations and recommendations for
research and practice, e.g. to better understand and manage the different types of
relationships between companies and their customers afforded by services [33].
Identifying service archetypes representing generic, theoretical prototypes or
modular service configurations serves research with the description of key elements of
services and identifies possible strategies for realizing service-business opportunities
in practice [17, 32]. Allmendinger and Lombreglia suggest four archetypes of smart
services to provide a more systematic view on business opportunities to enrich existing
products [34]. Similar, several authors strive to systemize the manifold opportunities
that digitalization, respectively the data generated in its context, offers by proposing
archetypes to more systematically manage transformation processes in practice [35,
36].
Current systematization approaches still focus on very specific contexts or neglect
the key role of analytics for the service. Hartmann et al. [24] unveil completely new
ways to conduct business solely based on data as a key resource and introduce six types
of data-driven services. By differentiating the identified service types according to the
underlying data source, and according to the type of performed activity, they provide a
basis for understanding how companies can develop completely new data-driven
services that provide value. Rizk et al. [37] developed a systematization of data-driven
digital services to better understand their key elements consisting of data collection
mechanisms, data utilization, insights usage, and service interaction characteristics.
Applied to real-world use cases, they found that services that are based on data are
either used as encapsulated services in larger service systems, as data visualizers, or as
specialized recommenders. Thus, the emerging research field of ABS would benefit
from a more general systematization of ABS unveiling different ABS types to further
deepen the understanding of how data and analytics can be leveraged to create new
service offerings systematically.
3 Research Design
In this study we combine quantitative and qualitative research [38]. In the quantitative
phase, we use an existing taxonomy of ABS to classify a number of real-world ABS
cases from start-ups and group them using cluster analysis. In the qualitative phase, we
interpret the identified clusters to derive ABS archetypes. In this section, we focus on
the explanation of the data collection and analysis process for the quantitative analysis.
Details on the derived archetypes are provided in the results section.
3.1 Data Collection
This paper aims to derive archetypes of ABSnew, innovative services that provide
customers with new value based on data and analytics. For this purpose, we base our
analyses exclusively on ABS offered by start-ups. Start-ups tend to be the first to exploit
the opportunities of new technologies for their business [19] and, unlike large
organizations, start-ups often offer a single, clearly defined service. Therefore, building
on ABS from start-ups as a unit of analysis seems to be adequate for our investigation.
To identify ABS cases, we drew upon AngelList's database, an online platform that
enables start-ups to raise money and investors to invest in attractive business concepts.
For this purpose, start-ups can advertise their projects via profiles on the platform and,
thus, publish information about their company and their proposal. In addition, the
companies self-categorize in the database by specifying their thematic focal points with
the help of keywords.
In the first step of our collection process, we examined the available keywords and
selected those that could be thematically linked to ABS. We then collected all start-up
cases that were related to the identified keywords, namely “analytics”, “machine
learning”, “artificial intelligence”, “data mining”, “big data”, “deep learning”, “internet
of things”. By collecting all cases in the first step, we aimed to eliminate researchers’
selection bias [39]. Second, we reviewed the identified cases to see whether they
actually described ABS. To this end, we examined whether data and analytics were
described in the service description as a key aspect to create value for B2B or B2C
customers. After removing duplicates, this filtering resulted in a final set of 2472 cases
identified as ABS.
In order to obtain a manageable dataset for the subsequent coding phase, we
randomly selected 15 use cases for each keyword, which resulted in a final set of 105
ABS cases consisting of the descriptions provided by the start-ups on AngelList. To
expand this database with deeper and insightful information about the selected ABS
as we found that the level of detail of these descriptions varied enormously additional
information about the ABS was collected in each case from the start-up's own website.
This included detailed information about the functioning of their ABS, but also
information about the start-up’s evolution, insights and beliefs that they had developed
over time.
3.2 Coding Mechanism
Our resulting dataset consisted of textual descriptions of real-word ABS. Literature
provides a rich collection of possible coding mechanisms for textual case analysis. A
provisional coding approach may be used in case there exists a conceptual framework
to serve as an underlying basis of a research inquiry [39, 40]. For that purpose, we built
on research previously conducted to conceptualize the nature of ABS which introduced
a taxonomy identifying commonly shared characteristics of this service type [8].
Taxonomies are a well-established instrument to describe and analyze new phenomena
using a unified classification schema [31]. Thus, the ABS-conceptualization defined in
this taxonomy served as a codebook for coding each of the 105 ABS use cases. It
consists of six dimensions (cf. Figure 1); each is represented by a distinct set of generic
characteristics. A detailed description of its six dimensions and their respective
characteristics which are perceived as key to conceptually describing ABS can be found
in [8].
The coding of the ABS use cases was performed by a single author. To ensure
validity of the conducted coding, a random 10% sample of our dataset was individually
coded by a second author in an ex-post quality check. A resulting inter-coder agreement
of 88.3% as percentage agreement and 73.1% as Cohen’s Kappa [41] suggests an
adequate coding quality. Following Landis and Koch [42], a Kappa between 61% and
80% indicates a “substantial” strength of the agreement among the coders.
Dimension
Characteristics
Data Generator
Customer
Non-customer
Process
Objects
Data Origin
Internal
External
Data Target
Non-customer
Process
Objects
Environment
Analytics Type
Descriptive
Diagnostic
Predictive
Prescriptive
Portfolio Integration
Stand-alone solution
Wrapped around product
Wrapped around service
Service User Role
Recipient
Provider
Interactor
Figure 1. Taxonomy of ABS serving as a codebook.
3.3 Cluster Analysis
Archetypes represent basic patterns from which individual copies can emerge [17].
Cluster analysis is a promising approach for identifying such archetypes. It is a
statistical technique to group similar objects according to their properties aiming to
achieve high homogeneity within each cluster and high heterogeneity between objects
of different clusters [43].
We followed the two-step procedure suggested by Punj and Stewart [20]. In the first
step, agglomerative hierarchical clustering was performed using Ward's method [44].
Agglomerative hierarchical clustering algorithms do not require a predefined number
of clusters, but generate solutions for all possible cluster numbers by gradually merging
the two nearest clusters in each step [43]. In order to determine the distance between
the individual objects, we dichotomized our data set, i.e. each taxonomy-characteristic
was represented with “1”, if the characteristic was identified in the respective use case
by the previous coding and a “0”, if not. To measure the distance, we used the simple
matching coefficient [45], since its interpretation for binary variables well fits our
context. Since cluster analysis does not provide guidance in determining the number of
clusters, this preliminary analysis step helped us to obtain a first approximation of a
solution examining the results in a dendrogram. It allowed us to determine a candidate
number of clusters and provided the opportunity to detect outliers for which cluster
analysis is sensitive. In our case, four or five clusters were perceived to provide the
most comprehensive insights.
In the second step, we performed an iterative partitioning clustering using the k-
medoids algorithm. The k-medoids algorithm groups objects into a predefined number
of clusters (k) by minimizing the distance between each individual object and its
corresponding cluster-representative object (medoid) for all objects in a cluster [43].
Since it uses objects from the dataset as medoids, it was selected over the more common
k-means algorithm making the results more meaningful for an archetypal interpretation.
Subsequently, we validated the cluster solution using the silhouette coefficient [46]. A
solution with four distinct cluster turned out to be the stronger solution with a silhouette
value of 0.41 indicating a weak, yet existing clustering structure in our dataset [47]. For
social science data such as ours, this is a typical result, as such data rarely exhibits
strong natural groups [32].
4 Four Archetypes of Analytics-Based Services
The cluster analysis identified four different clusters, consisting of 24 to 28 cases. Each
cluster exhibits different centers along the characteristics of the ABS taxonomy. To
derive generic archetypes from our cluster solution, we followed Hembrick's [32]
recommendation and inspected the characteristics frequency distributions of each
cluster to identify the most pronounced characteristics that could serve as archetype
boundaries. The resulting cross table provides an overview of the frequency distribution
of characteristics for each archetype (cf. Table 1). In the following, we provide a
detailed description for each archetype by highlighting typical characteristics and
providing an illustrative example.
Archetype A: Making data useful to customers. The first cluster characterizes ABS
that aim to make existing data sources useful to their customers. While these customers
oftentimes are aware of or might have access to large amounts of data in their business
context, data in its raw form exhibits a limited applicability to them. Thus, this ABS-
type processes data in ways so that customers can access it and use it more easily in
their daily activities. Typical applications in this cluster constitute aggregated reports,
dashboards, or APIs which provide customers with the opportunity to make better
decisions based on data.
Typical for this type of ABS is the use of process data (93%), e.g., business KPIs or
production data, which is generated by the ABS customers (external, 100%). Other
examples for an external origin are data collected from publicly available sources. The
data is processed using descriptive analytics (100%) to aggregate or visualize insightful
information targeting these processes (96%). Thus, the customer predominantly
engages with the ABS by providing the service-relevant data (68%) in case the data
originates from publicly available sources, the customer merely receives the aggregated
information without any further engagement with the ABS (29%).
One example for a start-up offering this ABS-type is Rollbar. Rollbar provides a
real-time error reporting system and continuous deployment monitoring for software
development teams. By integrating Rollbar into the customer’s local development
architecture, it collects all tickets that are created for detected errors. This data which,
potentially, is available independently from Rollbaris aggregated and visualized in a
real-time dashboard. Additionally, it automatically links subsequent bug reports to the
respective ticket and calculates the correlation of errors with previous occurrences.
Providing aggregated information team leaders can build on, the service allows to
capture errors earlier during development and, thus, creates value for them by
improving their software delivery processes across the entire development lifecycle.
Table 1. Results of cross table analysis.
Dimension
Characteristic
Archetype
A
B
C
D
Number of cases per cluster
28
26
27
24
Data Generator
Customer
14%
65%
37%
0%
Non-Customer
0%
4%
4%
0%
Processes
93%
69%
93%
17%
Objects
14%
23%
22%
100%
Data Origin
Internal
0%
12%
19%
92%
External
100%
96%
85%
8%
Data Target
Customer
4%
96%
59%
17%
Non-Customer
0%
12%
4%
0%
Processes
96%
27%
41%
33%
Objects
14%
0%
11%
63%
Environment
4%
0%
0%
33%
Analytics Type
Descriptive
100%
35%
7%
21%
Diagnostic
0%
65%
4%
13%
Predictive
0%
0%
67%
58%
Prescriptive
0%
0%
22%
8%
Portfolio Integration
Stand-alone
100%
96%
96%
4%
Wrapped around product
0%
0%
0%
96%
Wrapped around service
0%
4%
4%
0%
Service User Role
Recipient
29%
8%
4%
17%
Provider
68%
88%
0%
4%
Interactor
4%
4%
96%
79%
Archetype B: Delivering Data-Based Insights. The second cluster describes ABS that
aim to create new value by delivering meaningful insights to their customers based on
data. In contrast to the previous archetype which aims at enabling customers to use data
on their own behalf, this archetype also “digests” the data for the customer. It uses more
sophisticated, diagnostic analytics to deliver actionable insights customers can apply,
e.g. to make more informed decision. Typical applications in this cluster comprise
targeted benchmarks or meaningful alerts.
ABS in this cluster use data that is generated by dedicated business processes (69%)
or data that is generated elsewhere by the customer (69%) thus it originates externally
(96%). The data is predominantly analyzed to identify individual insights about
customers (96%) using diagnostic analytics (65%), i.e. it is not only identified whether
something happened, but also delivers insights why something happened. Similar to the
previous archetype, the customer mainly engages in the ABS by providing the relevant
data (88%).
14% 0% 93% 14% 0% 100% 4% 0% 96% 14% 4% 100% 0% 0% 0% 100% 0% 0% 29% 68% 4%
65% 4% 69% 23% 12% 96% 96% 12% 27% 0% 0% 35% 65% 0% 0% 96% 0% 4% 8% 88% 4%
37% 4% 93% 22% 19% 85% 59% 4% 41% 11% 0% 7% 4% 67% 22% 96% 0% 4% 4% 0% 96%
0%
0%
17%
100%
92%
8%
17%
0%
33%
63%
33%
21%
13% 58% 8% 4% 96% 0% 17% 4% 79%
Data Generator
Data Orig.
Data Target Analytics Type
Portf. Integration Serv
. User
Role
Customer
Non-customer
Process
Object
Internal
External
Customer
Non-customer
Process
Object
Environment
Descriptive
Diagnostic
Predictive
Prescriptive
Stand-alone
Wrapped around product
Wrapped around service
Recipient
Provider
Interactor
Archetypes
1
2
3
4
20% – 50% 51% – 80% 81% – 100%
UBiome provides a health care ABS that allows customers to understand their
microbiome with the ultimate goal to improve their lifestyle. UBiome provides a self-
sample kit, which their customers use at home. The sample is sent to UBiome, where it
is analyzed using advanced statistical techniques. A personalized diagnosis report is
prepared comparing the results with a healthy reference range and individual insights
are provided on which the customer can improve his everyday life, e.g., via healthier
nutrition.
Archetype C: Providing Data-Based Recommendations. The third cluster describes
ABS that aim to provide customers with meaningful, contextual recommendations for
action to solve problems. Business value is really unlocked from data when critical
insights are gained followed by immediate actions applying that new knowledge. To
this end, this type of ABS processes data in a way that allows to predict possible
outcomes and to make recommendations to the customer inspiring immediate action
and influence the customer's decisions.
Typical for this type of ABS is the use of process data (93%) which is occasionally
enriched with customer-generated data (37%). The data mainly originates externally
(85%). However, in this cluster start-ups also start using own, internal data (19%) such
as self-built machine learning models or relevant self-collected data. This ABS-type is
primarily intended to derive customer-specific, i.e. individually tailored insights (59%).
Distinctive for this cluster also is the use of advanced analytics using predictive (67%)
or prescriptive methods (22%) to derive the required insights. The results suggest that
the recommendations provided by the ABS are highly tailored to the customers’
individual needs. To achieve this, this type of ABS requires a deep integration and high
interaction with the customer (96%).
One example for a start-up offering this ABS-type is Proximus. Proximus uses
advanced machine learning models to analyze consumer behavior in brick and mortar
stores. They identify popular product dependencies among consumers and predict
future revenue streams on a daily basis to recommend better store layouts or product
bundles. Highly engaging with their customers via an online platform, Proximus’
recommendations inspire immediate actions for decision-makers that lead to higher
sales on a single-store level.
Archetype D: Enabling Novel Ways to Conduct Business. The fourth cluster
describes ABS that aim to create new value by enabling truly new ways to conduct
business for their customers. For that purpose, this ABS-type creates, in contrast to the
previous archetypes, completely new data sources that contain customer-specific
information. Thereby, this type of ABS creates new opportunities to identify
meaningful insights and to make purposeful recommendations for their customers.
Typical approaches for this archetype are sensor based IoT objects that are integrated
into the customer's workflow to deliver new data, insights, and ultimately provide
ground for improved ways to conduct business.
Typical for this type of ABS is the use data generated by objects, e.g., sensor data
(100%). Thus, these services draw upon their own, internal data (92%). The data is
analyzed with regard to the object’s own condition and state (63%) or to the customer’s
general business processes (33%). These insights are derived predominantly using
predictive analytics (58%). Similar to archetype C, these services use customer-related
data sources resulting from deep customer interaction (79%).
Skycatch is an example for a start-up offering this type of ABS. Skycatch provides
its customers, construction management firms, with a self-developed drone, which
allows them to digitize large construction sites using 3D-mapping technology. Skycatch
seamlessly integrates this data, which is continuously updated, e.g., on a daily basis,
into the customers own data models using a cohesive data suit. This integration allows
their customers to renew existing workflows, reaching previously unattainable gains
and efficiencies, e.g., by controlling contractors’ billing for removing dirt through
calculating the dirt volume based on 3D image.
5 Discussion and Implications
Using data and analytics in service offerings as a means to create new customer value
has recently become a much-regarded strategy by organizations [3, 10] and is being
actively explored by academics [7, 13]. Yet, little is known about how organizations
can leverage data and analytics to systematically expand their service portfolio. To this
end, this research identifies four archetypes of ABS that shed light on the objectives
market-pioneers pursue when offering such services creating novel value for their
customers. That is, 1) making data useful to customers, 2) delivering data-based
insights, 3) providing data-based recommendations, and 4) enabling novel ways to
conduct business. Each archetype is described by a set of distinct characteristics that
unveils the unique interplay of data, analytics, and customer integration of each type.
In order to contribute to systemizing the field of ABS and to develop an
understanding of how these four archetypes relate to each other, we developed a
strategy positioning map on two key dimensions. Rust and Huang [48], amongst others,
describe how services are able to evolve over time. Service offerings expand from static
“selling services” to interactive “co-creating services” as the relationship between the
service provider and the customer becomes stronger over time. Building on that, we
define a transactionalrelationalaxis to capture the degree to which a continuing,
stronger relationship is maintained and the ABS is deeper embedded in the customer’s
own systems and working habits. In addition, we noted that our results suggested a
tendency towards a steady transition from external to internal data sources that were
used in the different archetypes. We define a second axis, commonunique, which
captures the degree of uniqueness of the data used in the ABS.
Figure 2 shows the position of the four archetypes using these two dimensions in the
strategy positioning map. Archetype A is based on common data and a rather static
customer interaction, as the service-relevant data is also easily applicable for potential
competitors and the actual value is created downstream by the customer. In
archetype B, the interaction with the customer is more pronounced since the service is
more tailored to the individual customer and the service provider more strongly
contributes in the value creation process [49]. This results in the application of more
demanding analytics in the service. Archetype C is characterized by a strong,
continuous and interactive customer relationship, as the service is deeply embedded in
the customer's own processes. While customer-provided data could, potentially, also be
applied by competitors, type C integrates own data and experiences to create individual,
more meaningful recommendations increasing the uniqueness of the underlying
database for the service. Archetype D resembles similar relational characteristics
compared to type C. Yet, archetype D strongly builds on new, self-generated data
sources resulting in a high uniqueness of the data. Both, archetype C and D rely on
sophisticated, predictive analytics to reach the intended service objective.
Figure 2. Strategy positioning map of ABS archetypes.
Our analysis did not reveal a single ABS creating unique data that would build on a
transactional customer relationship to make that data simply more accessible to
customers (hatched area in Figure 2) suggesting that this area should be avoided since
the customer value seems to remain limited. Instead, we noticed a number of ABS in
our sample that had previously taken an “evolutionary” pathway along both axes by
intensifying customer interaction and advancing analytics sophistication over time.
Shoppermotion, for instance, offers in-store analytics for the retail and consumer
industry. Using beacons attached to shopping cards transmitting their store location,
Shoppermotion is able to provide stores with highly individualized, real-time
recommendations to leverage in-store consumer behavior (type D). Before using
sensor-based, self-generated data, Shoppermotion relied solely on customer-generated
in-store cash system data to provide recommendations (type C). In the very beginning,
the company simply aimed to reveal insights from past-sales data (type B).
5.1 Implications for Research and Practice
This study offers a number of theoretical implications that contribute to a deeper
understanding of ABS. Using an exploratory and descriptive analysis, this research
identifies four archetypes of ABS and introduces a framework to identify possible
Archetype A:
Making data useful
to customers
Archetype B:
Delivering data-
based insights
Archetype C:
Providing
recommendations
Archetype D:
Enabling novel
ways for business
Common data Unique data
Relational
Transactional
strategies to expand service portfolios with ABSa systematization of ABS that was
missing so far. The findings provide a possible reference point for further studies
aiming to theorize how analytics applied to data can be of real use for customer-facing
business practices; thereby providing a basis to tap into this, so far, blind spot in IS
literature [50].
This study also offers a number of managerial implications that can be particularly
helpful for organizations that are already taking advantage of ABS to create new
customer value or are planning to do so. First, our quantitative analysis revealed typical
approaches for using ABS to expand the service portfolio. This overview may help to
establish a more informed and systematic development of strategies to use data and
analytics in service offerings, in general. Second, each identified type of ABS is
described with the purpose of revealing commonalities and key components. This might
provide a valuable orientation for organizations when investing in ABS initiatives, e.g.
as possible blueprints for developing new services or guidelines for transforming
existing ABS into more sophisticated ones.
5.2 Limitations and Future Research
Our research certainly comes with some limitations. First, our analysis solely builds on
start-up use cases. While we argue that start-ups are a purposeful source to identify
ABS offerings, this decision limits the generalizability of our results regarding larger
organizations. Second, our data collection approach is limited to the AngelList database
and, therefore, this choice might influence the resultsgeneralizability. We were only
able to consider ABS by start-ups that tout for investors on this platform, increasing the
chance to miss innovative ABS elsewhere. Third, the data sample size in this study was
limited to 105 use cases due to the significant amount of manual work required to code
each case. While this reflects the exploratory nature of our research, this decision
potentially limits the ability to identify more nuanced differences between use cases
using cluster analysis.
These limitations at the same time leave the potential for future research. First, the
analyses should be conducted again using a larger sample of cases, ideally including
ABS use cases from larger organizations to increase the data sample’s diversity.
Second, future research could deeper investigate causal effects of ABS-enriched service
portfolios. For instance, it would be interesting to investigate the organization’s
different business capabilities required depending on the archetype they intend to offer.
As we already pointed out, the skillset regarding analytics capabilities seems to vary
between the archetypes, making it a fruitful topic to start with. Third, we see the
monetization of ABS as a promising field for future research. Interestingly, we found
that organizations that had evolved, e.g., from archetype A to archetype B, usually kept
their initial service offering and used it as a “basic service”, often in combination with
a freemium revenue model to attract possible customers. Thus, future research might
investigate how revenue models look like for different ABS archetypes.
6 Conclusion
With the digital transformation gaining momentum, organizations increasingly explore
how to expand their existing service portfolio using ABS. Yet, conceptual knowledge
on this novel service-type remains limited and IS literature misses to provide actionable
insights how such a portfolio enrichment can be achieved in a systematic way. Building
on a data sample of 105 ABS use cases that are offered by start-ups in the market and
using an established clustering procedure, we derive four distinct types of ABS
archetypes that provide initial empirically grounded evidence how ABS create new
customer value. These archetypes suggest that organizations may enrich their existing
service portfolio using ABS by either 1) making data useful to customers, 2) delivering
data-based insights, 3) providing data-based recommendations, or 4) enabling novel
ways to conduct business.
The findings in this research shed light on how ABS create customer value based on
data and analytics using an archetypal systematization. We believe, this systematization
contributes to conceptualizing ABS in this young research field and helps researchers
to channel future ABS-related theorizing attempts. Additionally, we believe
practitioners benefit from our discussion on how the identified archetypes might be
strategically distinguished. We introduce a strategy positioning map to propose
different opportunities to enrich the service portfolio providing a helpful starting point
for ABS initiatives in practice.
References
1. Huang, M.H., Rust, R.T.: Technology-driven service strategy. J. Acad. Mark. Sci. 45, 906
924 (2017).
2. Legner, C., Eymann, T., Hess, T., Matt, C., Böhmann, T., Drews, P., Mädche, A., Urbach,
N., Ahlemann, F.: Digitalization: Opportunity and Challenge for the Business and
Information Systems Engineering Community. Bus. Inf. Syst. Eng. 59, 301308 (2017).
3. Davenport, T.H., Harris, J.G.: Competing on Analytics: The New Science of Winning.
Harvard Business Review Press, Boston (2017).
4. Vargo, S.L., Lusch, R.F.: Why “service”? J. Acad. Mark. Sci. 36, 2538 (2008).
5. Troilo, G., De Luca, L.M., Guenzi, P.: Linking Data-Rich Environments with Service
Innovation in Incumbent Firms: A Conceptual Framework and Research Propositions. J.
Prod. Innov. Manag. 34, 617639 (2017).
6. Lehrer, C., Wieneke, A., vom Brocke, J., Jung, R., Seidel, S.: How Big Data Analytics
Enables Service Innovation: Materiality, Affordance, and the Individualization of Service.
J. Manag. Inf. Syst. 35, 424460 (2018).
7. Hunke, F., Engel, C.: Utilizing Data and Analytics to Advance Service Towards Enabling
Organizations to Successfully Ride the Next Wave of Servitization. In: Satzger, G., Patricio,
L., Zaki, M., Kühl, N., and Hottum, P. (eds.) Exploring Service Science, 9th International
Conference, IESS 2018. pp. 219231. Springer, Cham (2018).
8. Hunke, F., Engel, C., Schüritz, R., Ebel, P.: Understanding the Anatomy of Analytics-Based
Services a Taxonomy To Conceptualize the Use of Data and Analytics in Services. ECIS
2019 Proc. 115 (2019).
9. BASF Digital Farming: FIELD MANAGER: Simply Smarter Crop Protection,
https://www.xarvio.com/en/Field-Manager.
10. Demirkan, H., Bess, C., Spohrer, J., Rayes, A., Allen, D., Moghaddam, Y.: Innovations with
smart service systems: Analytics, big data, cognitive assistance, and the internet of
everything. Commun. Assoc. Inf. Syst. 37, 733752 (2015).
11. Saarijärvi, H., Grönroos, C., Kuusela, H.: Reverse use of customer data: Implications for
service-based business models. J. Serv. Mark. 28, 529537 (2014).
12. Davenport, T.H., Lucker, J.: Running on data. Deloitte Rev. 515 (2015).
13. Ostrom, A.L., Parasuraman, A., Bowen, D.E., Patrício, L., Voss, C.A.: Service Research
Priorities in a Rapidly Changing Context. J. Serv. Res. 18, 127159 (2015).
14. Lim, C.-H., Kim, M.-J., Kim, K.-H., Kim, K.-J., Maglio, P.P.: Using data to advance service:
managerial issues and theoretical implications from action research. J. Serv. Theory Pract.
28, 99128 (2018).
15. Remane, G., Hanelt, A., Nickerson, R.C., Tesch, J.F., Kolbe, L.M.: A taxonomy of
carsharing business models. ICIS 2016 Proc. 119 (2016).
16. Gimpel, H., Rau, D., Röglinger, M.: Understanding FinTech start-ups a taxonomy of
consumer-oriented service offerings. Electron. Mark. 28, 245264 (2018).
17. Schilling, R.D., Haki, M.K., Aier, S.: Introducing Archetype Theory to Information Systems
Research: A Literature Review and Call for Future Research. Proc. 13th Int. Conf.
Wirtschaftsinformatik. 574588 (2017).
18. Taran, Y., Nielsen, C., Thomsen, P., Montemari, M., Paolone, F.: Business Model
Archetypes: a Mapping Tool for Fostering Innovation. R&D Manag. Conf. 885902 (2015).
19. Criscuoloa, P., Nicolaoub, N., Salter, A.: The elixir (or burden) of youth? Exploring
differences in innovation between start-ups and established firms. Res. Policy. 41, 319333
(2012).
20. Punj, G., Stewart, D.W.: Cluster Analysis in Marketing Research: Review and Suggestions
for Application. J. Mark. Res. 20, 134148 (1983).
21. Müller, O., Fay, M., vom Brocke, J.: The Effect of Big Data and Analytics on Firm
Performance: An Econometric Analysis Considering Industry Characteristics. J. Manag. Inf.
Syst. 35, 488509 (2018).
22. Schüritz, R., Seebacher, S., Satzger, G., Schwarz, L.: Datatization as the Next Frontier of
Servitization: Understanding the Challenges for Transforming Organizations. ICIS 2017
Proc. 10981118 (2017).
23. Chen, Y., Kreulen, J., Campbell, M., Abrams, C.: Analytics Ecosystem Transformation: A
Force for Business Model Innovation. In: 2011 Annual SRII Global Conference. pp. 1120
(2011).
24. Hartmann, P., Zaki, M., Feldmann, N., Neely, A.: Capturing value from big data a
taxonomy of data-driven business models used by start-up firms. Int. J. Oper. Prod. Manag.
36, 13821406 (2016).
25. Lim, C.-H., Kim, K.H., Kim, M.J., Heo, J.Y., Kim, K.J., Maglio, P.P.: From data to value:
A nine-factor framework for data-based value creation in information-intensive services. Int.
J. Inf. Manage. 39, 121135 (2018).
26. Martin, D., Kühl, N.: Holistic System-Analytics as an Alternative to Isolated Sensor
Technology: A Condition Monitoring Use Case. Proc. 52nd Hawaii Int. Conf. Syst. Sci.
10051012 (2019).
27. Beverungen, D., Müller, O., Matzner, M., Mendling, J., Vom Brocke, J.: Conceptualizing
smart service systems. Electron. Mark. 29, 718 (2019).
28. Woerner, S.L., Wixom, B.H.: Big data: Extending the business strategy toolbox. J. Inf.
Technol. 30, 6062 (2015).
29. Wuenderlich, N. V., Heinonen, K., Ostrom, A.L., Patricio, L., Sousa, R., Voss, C.,
Lemmink, J.G.A.M.: “Futurizing” smart service: implications for service researchers and
managers. J. Serv. Mark. 29, 442447 (2015).
30. Ackoff, R.L.: From data to wisdom. J. Appl. Syst. Anal. 16, 39 (1989).
31. Nickerson, R.C., Varshney, U., Muntermann, J.: A method for taxonomy development and
its application in information systems. Eur. J. Inf. Syst. 22, 336359 (2013).
32. Hambrick, D.C.: Taxonomic approaches to studying strategy: some conceptual and
methodological issues. J. Manage. 10, 2741 (1984).
33. Lovelock, C.H.: Classifying Services to Gain Strategic Marketing Insights. J. Mark. 47, 9–
20 (1983).
34. Allmendinger, G., Lombreglia, R.: Four Strategies for the Age of Smart Services. Harv. Bus.
Rev. 83, 131145 (2005).
35. Weking, J., Stöcker, M., Kowalkiewicz, M., Böhm, M., Krcmar, H.: Archetypes for Industry
4 . 0 Business Model Innovations. 24th Am. Conf. Inf. Syst. (AMCIS), New Orleans. 110
(2018).
36. Zolnowski, A., Christiansen, T., Gudat, J.: Business Model Transformation Patterns of Data-
Driven Innovations. ECIS 2016 Proc. 116 (2016).
37. Rizk, A., Bergvall-Kåreborn, B., Elragal, A.: Towards a Taxonomy for Data-Driven Digital
Services. Proc. 51st Hawaii Int. Conf. Syst. Sci. 9, 10761085 (2018).
38. Bryman, A.: Integrating quantitative and qualitative research: How is it done? Qual. Res. 6,
97113 (2006).
39. Bryman, A.: Social Research Methods. Oxford University Press, Oxford (2012).
40. Saldana, J.: The Coding Manual for Qualitative Researchers. Sage, London (2009).
41. Cohen, J.: A coefficient of agreement for nominal scales. Educ. Psychol. Meas. 20, 3746
(1960).
42. Landis, J.R., Koch, G.G.: The Measurement of Observer Agreement for Categorical Data.
Biometrics. 33, 159174 (1977).
43. Han, J., Kamber, M.: Data Mining: Concepts and Techniques. Morgan Kaufmann, San
Francisco (2006).
44. Ward, J.H.: Hierarchical grouping to optimize an objective function. J. Am. Stat. Assoc. 58,
236244 (1963).
45. Sokal, R. R., & Michener, C.D.: A statistical method for evaluating systematic relationships.
Univ. Kansas Bull. 38, 14091438 (1958).
46. Rousseeuw, P.J.: Silhouettes: a graphical aid to the interpretation and validation of cluster
analysis. J. Comput. Appl. Math. 20, 5365 (1987).
47. Kaufman, L., Rousseeuw, P.J.: Finding groups in data: an introduction to cluster analysis.
John Wiley & Sons (1990).
48. Rust, R.T., Huang, M.-H.: The Service Revolution and the Transformation of Marketing
Science. Mark. Sci. 33, 206221 (2014).
49. Schüritz, R., Wixom, B., Farrell, K., Satzger, G.: Value Co-Creation in Data-Driven
Services : Towards a Deeper Understanding of the Joint Sphere. ICIS 2019 Proc. 19 (2019).
50. Hinz, O., van der Aalst, W.M.P., Weinhardt, C.: Blind Spots in Business and Information
Systems Engineering. Bus. Inf. Syst. Eng. 61, 133135 (2019).
... In the wake of the ongoing digital transformation, new ways of collecting and transmitting large volumes of data are established and advances in information systems technology provide the means to analyze them [3,4]. As a result, service systems have turned into data-rich environments that yield opportunities to create new customer value in novel service offerings [5,6]. However, systematic approaches to leverage data and create new value propositions for customers have not been sufficiently explored and, as a consequence, organizations still experience difficulties in finding new, innovative ways to exploit these technological trends [7,8]. ...
... Similar, smart services allow organizations to leverage sensor-equipped, connected products in the field for additional service offerings [3,19]. Using the provided data, organizations are able to offer contextual and preemptive services to their customers [6,21], leading to a closer relationship throughout the entire life cycle of their products. ...
... For the second evaluation episode, we decided to apply the refined artifact ex-post to a 'typical case' [31] of a real-world ABS use case described in [6]. The objective in this episode was to demonstrate the prototype's ability to orchestrate the activities of ABS in service systems and we used logic argumentation to evaluate the outcome's validity [40]. ...
Conference Paper
Full-text available
The digital transformation has led to the rise of data-rich service systems , in which organizations engage in data collection and analytics activities. Analytics-based services (ABS) are offered, enabling customers to make better decisions and providing assistance in solving more complex problems-thereby creating new customer value. The successful application of ABS requires the configuration of data and analytical models. These are usually spread across different organizational entities within the service system. Therefore, the orchestration of collaborative activities is key in creating customer value. This brings a new level of complexity for service design teams to develop new and innovative ABS solutions. This paper presents generic design knowledge that enables researchers to develop methodological assistance for these teams in the field of ABS in the future. Furthermore, it illustrates this design knowledge by presenting a canvas-based tool that offers a systematic view on the system-wide key activities required in ABS.
... Yet, the general understanding of how services can be designed from data still develops. Although companies are eager to "cash in" on their data [3], they still lack the comprehensive knowledge of how to fully take advantage of data and analytics to expand their portfolio with data-driven service offerings [4,5]. ...
... On the one side, personal data are used for exploration purposes to build a deeper understanding of the customer and his needs, e.g., from social media platforms [78]. On the other side, the data is seen as a new commodity, which itself is the design material for new value propositions [4,34]. Similarly, our research confirms that service designers and experts in data management still operate in separate worlds regarding the use of personal data. ...
... As the digital transformation continues, new ways of capturing and transmitting large amounts of data are being established and advances in information systems technology provide the means to analyze them [20]. Service systems are thus evolving into digital and cybernized environments providing the potential to create new service offerings [21,40]. This liquefaction of data throughout the service system is accompanied by a steadily increasing access to analytical capabilities [22,27]. ...
... Thus, the successful conceptualization of a service can be decoupled from individual experts through modular reuse and leads to service design teams jointly exploring and testing new solutions. Nonaka and Krogh [29] describe this process as knowledge transformation from implicit to explicit knowledge in organizations -an effect, that seems promising for the design of digital and cybernized services, which often build on specific data sources or analytical applications [21,40]. ...
Conference Paper
Full-text available
The ability to effectively conceptualize services plays a key role in service design. Yet, action-relevant insights for service designers on how this crucial phase of service design could be systematically pursued remain surprisingly limited. This paper explores "modular reuse", a frequently used principle to efficiently implement and modify service processes in downstream, i.e. implementation, activities as a possible underlying mechanism to inspire and facilitate the conceptualization phase. In a Design Science Research approach, we develop and formatively evaluate three design principles, which enable the purposeful reuse of existing service elements for the conceptualization of new services. This research contributes by 1) tapping into novel ways to improve the conceptualization phase of service design, proposing an underlying mechanism to increase its effectiveness; 2) providing initial design principles that serve as prescriptive guidelines in the creation of new methods and tools to leverage the idea of modular reuse for the conceptualization of new services.
... Folglich weist der entsprechende Diskurs in Wissenschaft und Praxis einstimmig darauf hin, dass analysebasierte Dienstleistungen ein vielversprechender Ansatz für Unternehmen sind, um neue Wettbewerbsvorteile zu erlangen, indem sie eine tiefergehende Kundenbindung beziehungsweise die Erschließung völlig neuer Märkte ermöglichen. Die systematische Erschließung der Potenziale für analysebasierte Dienstleistungen hat daher einen Schwerpunkt in der Forschung erhalten und wird intensiv von Unternehmen vorangetrieben (Ostrom et al. 2015;Hunke et al. 2020). ...
... Diese analysebasierten Dienstleistungen ermöglichen es den Kunden, bessere Entscheidungen zu treffen und komplexere Probleme zu lösen, um letztendlich ihre Ziele effizienter oder effektiver zu erreichen (Hunke et al. 2019a). Die steigende Verfügbarkeit von Daten und den entsprechenden analytischen Fähigkeiten, diese auszuwerten, bietet Unternehmen zahlreiche Möglichkeiten, ihr Dienstleistungsangebot mit Daten und Analyseverfahren anzureichern (Hunke et al. 2020). Eine frühe Unterscheidung in dieser Hinsicht wird zwischen Data-as-a-Service und Analytics-as-a-Service getroffen (Chen et al. 2011). ...
Article
Full-text available
Zusammenfassung Analysebasierte Dienstleistungen stellen einen neuartigen Servicetyp dar. Sie stützen ihre Wertschöpfung im Schwerpunkt auf die Anwendung statistischer Analyseverfahren auf Daten, um es ihren Kunden zu ermöglichen bessere Entscheidungen zu treffen oder komplexere Probleme lösen zu können. Während jedoch ihr Potenzial für neue Wertschöpfungsmöglichkeiten viel diskutiert ist, bleibt das Service Design als Instrument zur Erarbeitung neuer Dienstleistungen in Bezug auf diesen Servicetyp bisher unspezifisch. Als Folge haben Service Design Teams nach wie vor Schwierigkeiten, analysebasierte Dienstleistungen systematisch zu entwickeln. In diesem Beitrag entwickeln wir ein neues methodisches Werkzeug, das Service Designer dabei unterstützt Servicekonzepte analysebasierter Dienstleistungen zu entwickeln. Hierfür stellen wir allgemeingültige Gestaltungsrichtlinien auf und nutzen diese anschließend, um ein neues methodisches Werkzeug – das Analysebasierte-Dienstleistungs-Canvas – vorzuschlagen.
... Second, new data-driven ecosystems emerge based on "big data" (Fromm & Bloehdorn, 2014) availability, among others driven by sensorization of industrial assets, i.e. the emergence of the Internet of Things. The systematic exploitation of data in analyticsbased services (Hunke et al., 2020), or specifically in "smart services" (Beverungen et al., 2019;Maleshkova et al., 2020), will lead to "datatization as the next frontier of servitization" (Schüritz et al., 2017). Research needs to support the innovation of services and business models, posing particular challenges for distributed data sourcing in ecosystems, e.g., tackled by federated or transfer learning approaches (Yang et al., 2019;Hirt et al., 2020). ...
Preprint
Short description Across many industries we observe trends to change business models in favor of service offerings ("servitization") as well as trends to apply digital technologies for value creation ("digitalization"). First, we perform a conceptual analysis to explain their interdependence. Second, we combine a topic modeling literature study and a qualitative expert interview approach to identify key research fields for the service and information system research communities. The work should deepen the understanding of service concepts to inform the purposeful design and application of digital technologies for service innovation along the identified research roadmap-as "high-tech meets high-touch".
... Such ABS enable customers to make better decisions and to solve more complex problems to ultimately achieve their goals [5,6]. The proliferation of data provides companies with numerous opportunities to infuse their service offerings with data [22]. Huang and Rust stress that customer data hold the potential to "figure out […] why customers make decisions they make and why they behave in a certain way" [ [23], p. 255]; thereby facilitating a much deeper access to the customer. ...
Conference Paper
Full-text available
Analytics-based services (ABS) apply analytical methods to data in order to enable customers to make better decisions and solve more complex problems. While it is widely acknowledged that ABS pave the way for new value creation opportunities, surprisingly little is known about their systematic design. Service design teams still struggle to create ABS solutions systematically, i.e. to define what is to be done, how this is going to be achieved and how decisions are taken during ABS design projects. In this research, we report on the first iteration of our design science research project which aims to build design knowledge on methodological tools that can support service design teams in this particular context. We derive and evaluate four meta-requirements and four design principles-thus contributing to a more profound design knowledge base that can support researchers in developing new methodological tools in the field of ABS in the future.
Chapter
The digital transformation has led to the rise of data-rich service systems, in which organizations engage in data collection and analytics activities. Analytics-based services (ABS) are offered, enabling customers to make better decisions and providing assistance in solving more complex problems – thereby creating new customer value. The successful application of ABS requires the configuration of data and analytical models. These are usually spread across different organizational entities within the service system. Therefore, the orchestration of collaborative activities is key in creating customer value. This brings a new level of complexity for service design teams to develop new and innovative ABS solutions. This paper presents generic design knowledge that enables researchers to develop methodological assistance for these teams in the field of ABS in the future. Furthermore, it illustrates this design knowledge by presenting a canvas-based tool that offers a systematic view on the system-wide key activities required in ABS.
Conference Paper
Full-text available
While the co-creation of value between provider and customer(s) is a common theme in service research, continued theoretical conceptualization is required to guide more effective design of services. In order to advance understanding of value co-creation, we have conducted a pre-study consisting of 16 interviews in the application domain of data-driven services-services that support the decision-making process of customers through the provision of data and analytics. Adopting the theoretical lens of joint spheres, we present empirical evidence that the proportions of the joint sphere can vary with differences in interaction, access to customer processes or behaviors and decision power. As a next step, we propose a final case study phase to further explore factors that influence the proportion of joint sphere and, thus, the provider's real value creation. Understanding how real value creation is impacted by the resource integration of provider and customer(s) will ultimately enable the purposeful design of data-driven services for effective value co-creation.
Conference Paper
Full-text available
The abundance of data accompanied by advances in analytics technologies increasingly drive companies to introduce analytics-based services, i.e. customer-facing services in which data and analytics help customers make decisions. Despite its growing application in practice, theoretical and conceptual work on analytics-based services is still scarce. In this paper, we develop a taxonomy of analytics-based services unveiling their conceptually grounded and empirically validated characteristics. Applying an established taxonomy building method, we draw upon an analysis of 85 use cases of analytics-based services. The results of an expert evaluation indicate both the usefulness and robustness of our taxonomy. The developed taxonomy of analytics-based services contributes in two ways: First, we add to the descriptive knowledge on this new service type, establish a common language among researchers and equip them with the means to analyze analytics-based services in a structured manner-thus laying the foundation for a deeper theorizing process in the future. Second, we provide a concrete conceptualization of analytics-based services for practitioners for initial guidance in new service development.
Conference Paper
Full-text available
Sensor technology has become increasingly important (e.g., Industry 4.0, IoT). Large numbers of machines and products are equipped with sensors to constantly monitor their condition. Usually, the condition of an entire system is inferred through sensors in parts of the system by means of a multiplicity of methods and techniques. This so-called condition monitoring can thus reduce the downtime costs of a machine through improved maintenance scheduling. However, for small components as well as relatively inexpensive or immutable parts of a machine, sometimes it is not possible or uneconomical to embed sensors. We propose a system-oriented concept of how to monitor individual components of a complex technical system without including additional sensor technology. By using already existing sensors from the environment combined with machine learning techniques, we are able to infer the condition of a system component, without actually observing it. In consequence condition monitoring or additional services based on the component's behavior can be developed without overcoming the challenges of sensor implementation.
Article
Full-text available
This paper reports on an exploratory, multisite case study of four organizations from the insurance, banking, telecommunications, and e-commerce industries that implemented big data analytics (BDA) technologies to provide individualized service to their customers. Grounded in our analysis of these four cases, a theoretical model is developed that explains how the flexible and reprogrammable nature of BDA technologies provides features of sourcing, storage, event recognition and prediction, behavior recognition and prediction, rule-based actions, and visualization that afford (1) service automation and (2) BDA-enabled human-material service practices. The model highlights how material agency (in the case of service automation) and the interplay of human and material agencies (in the case of human-material service practices) enable service individualization, as organizations draw on a service-dominant logic. The paper contributes to the literature on digitally enabled service innovation by highlighting how BDA technologies are generative digital technologies that provide a key organizational resource for service innovation. We discuss implications for research and practice.
Article
Full-text available
The emergence of big data has stimulated enormous investments into business analytics solutions, but large-scale and reliable empirical evidence about the business value of big data and analytics (BDA) remains scarce. This paper presents the results of an econometric study that analyzes the direction, sign, and magnitude of the relationship between BDA and firm performance based on objective measurements of BDA assets. Using a unique panel data set that contains detailed information about BDA solutions owned by 814 companies during the timeframe from 2008 to 2014, on the one hand, and their financial performance, on the other hand, we estimate the relationship between BDA assets and firm productivity and find that live BDA assets are associated with an average of 3 to 7 percent improvement in firm productivity. Yet, we also find substantial differences in returns from BDA when we consider the industry in which a firm operates: While firms in IT-intensive or highly competitive industries are clearly able to extract value from BDA assets, we did not detect measurable productivity improvement for firms outside these industry groups. Taken together, our findings provide robust empirical evidence for the business value of BDA, but also highlight important boundary conditions.
Article
Applications of cluster analysis to marketing problems are reviewed. Alternative methods of cluster analysis are presented and evaluated in terms of recent empirical work on their performance characteristics. A two-stage cluster analysis methodology is recommended: preliminary identification of clusters via Ward's minimum variance method or simple average linkage, followed by cluster refinement by an iterative partitioning procedure. Issues and problems related to the use and validation of cluster analytic methods are discussed.
Article
Service is a key context for the application of IT, as IT digitizes information interactions in service and facilitates value creation, thereby contributing to service innovation. The recent proliferation of big data provides numerous opportunities for information-intensive services (IISs), in which information interactions exert the greatest effect on value creation. In the modern data-rich economy, understanding mechanisms and related factors of data-based value creation in IISs is essential for using IT to improve such services. This study identified nine key factors that characterize this data-based value creation: (1) data source, (2) data collection, (3) data, (4) data analysis, (5) information on the data source, (6) information delivery, (7) customer (information user), (8) value in information use, and (9) provider network. These factors were identified and defined through six action research projects with industry and government that used specific datasets to design new IISs and by analyzing data usage in 149 IIS cases. This paper demonstrates the usefulness of these factors for describing, analyzing, and designing the entire value creation chain, from data collection to value creation, in IISs. The main contribution of this study is to provide a simple yet comprehensive and empirically tested basis for the use and management of data to facilitate service value creation.