UCC Library and UCC researchers have made this item openly available.
Please let us know how this has helped you. Thanks!
Title
Towards comparable business model concepts: resource description
framework (RDF) schemas for semantic business model representations
Author(s)
Schwarz, Johannes; Terrenghi, Nicola; Legner, Christine
Editor(s)
Maedche, Alexander
vom Brocke, Jan
Hevner, Alan
Publication date
2017
Original citation
Schwarz, J., Terrenghi, N. and Legner, C. 2017. 'Towards Comparable
Business Model Concepts: Resource Description Framework (RDF)
Schemas for Semantic Business Model Representations'. In: Maedche,
A., vom Brocke, J., Hevner, A. (eds.) Designing the Digital
Transformation: DESRIST 2017 Research in Progress Proceedings of
the 12th International Conference on Design Science Research in
Information Systems and Technology. Karlsruhe, Germany. 30 May - 1
Jun. Karslruhe: Karlsruher Institut für Technologie (KIT), pp. 101-109
Type of publication
Conference item
Link to publisher's
version
https://publikationen.bibliothek.kit.edu/1000069452
Rights
©2017, The Author(s). This document is licensed under the Creative
Commons Attribution – Share Alike 4.0 International License (CC
BY-SA 4.0): https://creativecommons.org/licenses/by-sa/4.0/deed.en
http://desrist2017.kit.edu/
Access to the full text of the published version may require a
subscription.
https://creativecommons.org/licenses/by-sa/4.0/deed.en
Item downloaded
from
http://hdl.handle.net/10468/4447
Downloaded on 2021-12-05T16:40:35Z
Towards Comparable Business Model Concepts:
Resource Description Framework (RDF) Schemas for
Semantic Business Model Representations
Research in Progress
Johannes Schwarz1,2, Nicola Terrenghi1,2 and Christine Legner1
1
University of Lausanne, Faculty of Business and Economics (HEC), Lausanne, Switzerland
2 SAP (Schweiz) AG, Innovation Center Network, St.Gallen, Switzerland
{johannes.schwarz,nicola.terrenghi,christine.legner}@unil.ch
Abstract. Scholars have demonstrated that business model (BM) choices have a
significant impact on the success of products, innovations and organizations.
However, knowledge about key elements of BMs is disseminated across a large
body of literature and builds on different conceptualizations. We take a step back
and provide a new approach to formalize BM concepts and related BM
knowledge, based on concepts from the semantic web. We introduce and evaluate
the Resource Description Framework (RDF) as a data model for comparable and
extensible BM descriptions. Moreover, we use this new perspective to analyze
commonalities and differences between BM concepts, to reflect critically on the
process of translating concepts to RDF and evaluate its relevance for BM design
practice.
Keywords: Business Model · Business Model Representation · RDF · Semantics
1
Introduction
The business model (BM) is a highly interesting object for product owners, innovation
managers and strategists alike [1]–[3]. It can represent the logic and capabilities of a
business in a “remarkably concise way” [4] and serve as a holistic approach to renew
and innovate organizations in times of digitization and change [5], [6]. Research seeks
to develop “conceptual toolkit[s] that enables entrepreneurial managers to design their
future business model, as well as to help managers analyze and improve their current
designs to make them fit for the future.” [7].
A main challenge in the BM domain, however, is that multiple definitions, representations and formats of BMs exist [8], [9]. These conceptualizations are either very formal in terms of ontologies or taxonomies or less formal and result in many different
perspectives of what the “key” constructs of the BM concept are [10]. At the same time,
arguments have been made to make research “more cumulative in nature, and to effect
a more efficient transfer of research results into practice” [11].
101
We assume that each (re-)conceptualization of a BM adds novel, partially overlapping yet equally relevant facets that, together, are a valuable source of knowledge for
BM innovation and decision making. However, to translate these insights from rigor to
relevance we address the challenges of how to find, integrate and use different BM
conceptualizations. Consequently, the research question is the following: How can varying BM conceptualizations be integrated to make key aspects of the concepts as well
as attached BM knowledge comparable?
We take a step back and provide a new approach to formalizing BM concepts and
related BM knowledge, based on ideas from the semantic web. We introduce and evaluate the Resource Description Framework (RDF) as a data model for comparable and
extensible description of BMs. This approach allows not only the representation of very
formal BM ontologies but also of less-structured concepts that are primarily text-based
– in a common format (schema). We do not focus on a specific kind of BM concept but
on the underlying mechanisms of describing, comparing and transferring BM
knowledge. Moreover, we use this new perspective to analyze commonalities and differences between BM concepts, to reflect critically on the process of translating concepts to RDF and evaluate its relevance for BM design practice.
2
Conceptual foundations
This section introduces the two key concepts – BMs and the resource description framework – before proposing why and how both can benefit from each other.
2.1
Business models
BMs have become a critical element for business success and the concept is identified
“as the missing link between business strategy, processes, and Information Technology” [11]. Scholars from various disciplines use the concept to understand how organizations create, capture and deliver value in different markets [12]. [13] classify BM
research in three streams: overarching concept (“meta-models that conceptualize
them”), taxonomies (generic BM types with common characteristics), and the instances
(that “consists of either concreate real world business models or […] descriptions of
real world business models”). Many scholars have tried to define BMs formally by
developing ontologies, taxonomies or frameworks [13]–[15]. No definition seems to
satisfy all purposes [9], [16]. Thus, our research provides a data model to describe different BM concepts (meta-models) and demonstrate the application of this data model
based on BM types and instances.
2.2
Resource description framework
The resource description framework (RDF) is a standard model and abstract syntax to
represent information [17] and is primarily used in the context of the semantic web
[18]. In RDF, information is represented in a set of triples. Each triple consists of a
102
subject, predicate and object. By forming triples, we build statements about the relationship (predicate) between two resources (subject and object) [17]. For example, one
could state that the city “Berlin” (subject) “has-major” (predicate) “Michael Müller”
(object). Or that the customer segment “professionals” (subject) “have-a-need-for”
(predicate) “seamless online shopping” (object).
Fig. 1. An RDF graph with two nodes (subject and object), a triple connecting them (predicate)
[17], additional properties and examples (“bms:” prefix serves as identifier for a namespace)
2.3
RDF schemas as flexible, comparable and reusable ontologies
Without constraints and additional semantics, the triple-logic could be used to describe
any kind of (un-)meaningful data. For example, the following statements are meaningless in a BM context:
:large-enterprises :type :software
:customer-segment :customer-segment :customer-segment.
Therefore, predefined RDF vocabularies are available in the RDF namespaces that can
be used to create simplified, domain-specific ontologies, called schemas, which provide
a set of definitions and constraints for the underlying RDF data. For example, a schema
could define “Customer Segment” as a meaningful “type” of RDF Resource, which can
have a Property “has-a-need-for”1. Schemas provide the meta-data for the actual information.
2.4
Towards comparable and extensible business model concepts based on RDF
schemas
This research is motivated by the fact that the body of BM research has similar characteristics to the World-Wide-Web where data “covers diverse structures, formats, as well
as content […] and lacks a uniform organization scheme that would allow easy access
to data and information” [18]. We assume that the business model is a complex, multifaceted concept with different, equally relevant aspects that differ across context or
purpose and that the concept will evolve even further in the future. Here, RDF and RDF
1
Please refer to the official specifications [17] for additional information on the RDF concept
and the set of predefined vocabularies.
103
schemas can help to create comparable, extensible and processible descriptions of BM
(meta-)information. In specific, we see the following advantages of RDF:
Properties: Unlike traditional object- and class-oriented data models, RDF provides
a rich data model where relationships are also first class objects, “which means that
relationships between objects may be arbitrarily created and be stored separately
from the two objects. This nature of RDF is very suitable for dynamically changing,
distributed, shared nature of the Web” [18]. In other words, relationships between
objects can be added without changing the definition of the class. An existing BM
construct such as “Customer Segment” can be enriched with idiosyncratic properties
(e.g. linking customer segments with atomic values or other resources).
Namespaces: A unique feature of RDF is that is uses the XML namespace mechanism: “A namespace can be thought of as a context or a setting that gives a specific
meaning to what might otherwise be a general term. […] using namespaces, RDF
provides ability to define and exchange semantics among communities.” [18]. The
advantage of this is that each BM concept can be associated with its own namespace
and BM information building on different concepts be exchanged.
Mixing definitions: One of the most interesting features of RDF is its extensibility
and shareability. It “allows metadata authors to use multiple inheritance to mix definitions and provide multiple views to their data. In addition, RDF allows creation of
instance data based on multiple schemas from multiple sources” [18]. Scholars, who
document information about BM instances, often combine different BM conceptualizations.
Query language: With SPARQL (an acronym for SPARQL Protocol and RDF
Query Language) a powerful tool to query RDF data is available (see the following
example that gets all BMs with a customer in the software industry).
@prefix bm: http://bm.example.com/exampleBmOntology#
SELECT ?businessModel ?customer
WHERE {
?businessModel bm:hasCustomer ?customer .
?customer bm:isInIndustry bm:Software;
}
3
Methodology
This research follows a design science paradigm [19] and Peffers et al’s. [20] specific
guidelines. Peffers et al. suggest the following phases: problem identification & motivation, objectives of a solution, design & development, demonstration, evaluation,
communication. Problem, motivation and objective were already outlined in the previous sections. Communication takes place in academic conference proceedings. The following sub-sections explain the remaining phases.
104
3.1
Design and development: eliciting conceptual meta-constructs and creating
an exemplary BM schema vocabulary
We leverage an exemplary set of six BM conceptualizations (see table 2) and translate the underlying implicit or explicit constructs into RDF schemas. Conceptual literature is defined as peer-reviewed, scientific articles that explicitly discuss the nature of
the BM concept (in contrast to case studies that use the BM lens, to understand how a
company works, for example [21]) and have a significant number of citations (>150).
In contrast to previous BM ontology mapping approaches [22], [23], we do explicitly
include also more qualitative concept definitions. For example, [7] consider BMs as
activity systems which are “a set of interdependent organizational activities centered
on a focal firm” with two relevant design parameters “design elements and design
themes”. The concept is not as formal as a taxonomy and rather implicit but holds valuable information about key BM constructs. Specifically, we translate text and concepts
into a set of triple statements which will then be consolidate in an RDF schema in the
namespace of the authors. To remain with the example of [7], the sentence “an activity
in a focal firm’s business model can be viewed as the engagement of human, physical
and/or capital resources of any party to the business model” is translated into the following schema (extract):
@prefix za: http://schema.bm.org/2010/Zott_and_Amit
za:BusinessModel
za:consistsOf
za:Activity
za:Activity
za:linkedTo
za:ActivityLink
za:Activity
za:uses
za:Resource
za:ActivityLink
za:hasNovelty
rdfs:Bag [“Novel”, “NotNovel”]
za:Physical
a
za:Resource
In a parallel step, we review these concepts for commonalities to create a common BM
schema within its own namespace (BMS) that represents a custom schema mapping –
yet again extensible and comparable in RDF.
3.2
Demonstration: Representing business models and BM knowledge
To demonstrate the value of compatible BM RDF schemas for BM design purposes we
select exemplary BMs from the BM literature (e.g. [21]) and represent them by manually selecting constructs from the concept schemas. Moreover, we want to demonstrate
that BM knowledge that builds currently on different BM concepts (e.g. [24]) can be
described and identified. We select in total 10 exemplary articles from the body of BM
research that build on one (or a combination of) the above BM concepts. These articles
represent BM design knowledge either by BM type or instance. We use an instance of
Apache Jena, an open source framework for semantic web and linked data applications2, to store and query the RDF data.
2
http://jena.apache.org/index.html
105
3.3
Evaluation
We adopt the following evaluation criteria:
Table 1. Evaluation criteria
Criterion
RDF translatability
Construct-schema
coverage
Knowledge
extraction and comparability
Definition and measurement
In general, any text or concept can be translated into more formal ontologies, regardless of the data format. Thus, we reflect on our experiences
with RDF and provide qualitative insights, whether we perceived this
process as easy or difficult and whether enhanced text-to-ontology
methods could support this process. Moreover, we provide suggestions
for researchers who want to code other BM information bases.
An interesting aspect is, to what extend different BM conceptualizations
build on similar meta-constructs and whether a common schema is
meaningful. We build the BMS iteratively, based on the body of
knowledge identified as describe in section 3.1 and try to identify the
lowest common denominator of constructs. We provide a simple measurement of construct coverage in our schema. This will support our understanding of differences and commonalities between BM concepts.
The main research objective is to assess whether RDF supports the representation and extraction of meaningful BM knowledge for BM design
purposes. We evaluate this aspect based on ten BM instances and interviews with at least three different BM experts (who have more than two
years of experience working on BM innovation or BM development). In
particular, we will assess whether the underlying semantics help to identify a) problems such as inconsistency between BM elements and b) additional, previously unknown BM knowledge to improve the BM design.
4
Preliminary results
4.1
Representing and comparing BM concepts in RDF (ongoing)
Table 2. Business model concept meta-constructs (legend: ● key construct, explicitly defined
and ex-plained defined and explained ◑ mentioned but not explicitly defined mentioned
briefly ○ not mentioned)
Completed?
…
Level
Transaction
Value
Actor
Attribute
Group
Other
Ele-
Offering
Activity
Link
Author(s): concept name
(if available) and source
Element
Concept constructs (BMS vocabulary)
Resource
Named
ments
106
Akkermans & Gordijin: e³-value ontol- ● ● ○
ogy [14]
Osterwalder & Pigneur: business model ● ◑ ●
ontology [13]
Zott & Amit: Activity system [7]
● ◑ ◑
Johnson et al. [25]
● ◑ ●
Casadesus-Masanell & Ricart : Choices ◑ ● ◑
and consequences [26]
Demil & Lecocq : RCOV [27]
● ● ●
● ●
○ ● ● ● ● ●
○
● ● ● ● ● ◑ ◑
● ◑ ●
● ● ●
●
●
● ● ◑ ○
◑ ● ◑ ●
◑ ◑ ●
◑ ◑ ●
◑
○
◑
● ◑ ◑
◑
Our current results are mainly based on a detailed analysis of a subset of BM concepts and a simple analysis of all concepts for similar constructs (Table 2). For a subset
of concepts [7], [27], [28], we have extracted all sentences and images that include
relevant facts about the BM concept, for example “Choices, [..] are not the sole constituent of business models. As all authors highlight, choices must be connected to value
creation and value capture, or to alternative goals the company may want to pursue”
[28] and translated them into RDF statements. These statements were then consolidated
to create an initial BM concept schema within the namespace of the corresponding authors (e.g. Casadesus-Masanell_and_Ricart). In general, our impression is that the process of translating text to RDF works very well and the resulting schema is consistent
even when created independently by two authors of this paper. Moreover, we discover
some constructs that appear frequently, such as the idea to decompose a BM into elements and links between these elements and novel constructs that are usually not explicitly modeled (for example the properties “novelty” of an activity link or “switching
costs” of a customer [7]). This is also a main difference to previous ontologies and
ontology mappings because we discover additional properties that may have important
implications for the understanding of a BM. Moreover, our current impression is that,
yes, concepts differ significantly but have certain constructs in common. These common constructs can then be used the make links between the concepts explicit. For
example, [13] consider revenue models, costs and activities (besides others such as
channels, customers etc.). In contrast, [7] focus mainly on activities and consider the
revenue model as “conceptually distinct” and [28] introduce price as a “choice” and
cost as a “consequence”. We look forward to evaluate whether a standard BM schema
improves concept integration. Our preliminary results strengthen the assumption that
each BM concept is unique and that attempts to conciliate them into one ‘ideal’ ontology are likely to remain impracticable because ontology mappings and simple taxonomies neglect relevant properties. Given that these concepts are then used to capture BM
knowledge, for example about the development of cloud BMs in the software industry,
a flexible and comparable schema language, such as RDF can then help to them in a
simple, yet effective way and to model the underlying BM information.
References
1.
D. J. Teece, “Business models, business strategy and innovation,” Long Range Plann., vol.
43, no. 2–3, pp. 172–194, 2010.
107
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
M. Sosna, R. N. Trevinyo-Rodríguez, and S. R. Velamuri, “Business model innovation
through trial-and-error learning: The naturhouse case,” Long Range Plann., vol. 43, no. 2–
3, pp. 383–407, 2010.
Chesbrough and R. S. Rosenbloom, “The role of the business model in capturing value
from innovation: evidence from Xerox Corporation’ s technology spin-off companies,”
Ind. Corp. Chang., vol. 11, no. 3, pp. 529–555, 2002.
[C. Baden-Fuller and M. S. Morgan, “Business models as models,” Long Range Plann.,
vol. 43, no. 2–3, pp. 156–171, 2010.
R. Amit and C. Zott, “Business Model Design: A Dynamic Capability Perspective,”
(Working), 2014.
C. Baden-Fuller and S. Haefliger, “Business Models and Technological Innovation,” Long
Range Plann., vol. 46, no. 6, pp. 419–426, 2013.
C. Zott and R. Amit, “Business Model Design : An Activity System Perspective,” Long
Range Plann., no. APRIL, 2010.
T. Burkhart, D. Werth, J. Krumeich, and P. Loos, “Analysing the business model concept
- A comprehensive classification of literature.,” Thirty Second Int. Conf. Inf. Syst., pp. 1–
19, 2011.
B. W. Wirtz, A. Pistoia, S. Ullrich, and V. Göttel, “Business Models: Origin, Development
and Future Research Perspectives,” Long Range Plann., 2015.
B. W. Wirtz, A. Pistoia, S. Ullrich, and V. Göttel, “Business Models: Origin, Development
and Future Research Perspectives,” Long Range Plann., pp. 1–19, 2015.
D. Veit, E. Clemons, A. Benlian, P. Buxmann, T. Hess, D. Kundisch, J. M. Leimeister, P.
Loos, and M. Spann, “Business Models An Information Systems Research Agenda,” Bus.
Inf. Syst. Eng. - Res. Notes2, no. 2014, pp. 45–53, 2014.
C. Zott, R. Amit, and L. Massa, “The Business Model: Recent Developments and Future
Research,” J. Manage., vol. 37, no. 4, pp. 1019–1042, 2011.
A. Osterwalder, Y. Pigneur, and C. L. Tucci, “Clarifying business models: origins, present,
and future of the concept,” Commun. Assoc. Inf. Syst., vol. 15, no. 1, pp. 1–43, 2005.
J. Gordijn and H. Akkermans, “Designing and Evaluating E-business,” Ieee, no.
July/August, pp. 11–17, 2001.
M. M. Al-Debei and D. Avison, “Developing a unified framework of the business model
concept,” Eur. J. Inf. Syst., vol. 19, no. 3, pp. 359–376, 2010.
T. Burkhart, J. Krumeich, D. Werth, and P. Loos, “Analyzing the Business Model C Oncept
— a Comprehensive Classification,” ICIS 2011 Proc., no. September, pp. 1–19, 2011.
W3C, “RDF 1.1 Concepts and Abstract Syntax,” W3C Recommendation 25 February 2014,
no. February. pp. 263–270, 2014.
K. S. Candan, H. Liu, and R. Suvarna, “Resource Description Framework : Metadata and
Its Applications,” SIGKDD Explor., vol. 3, no. 1, pp. 6–19, 2001.
Hevner, S. T. March, and J. Park, “Design Science in Information Systems Research,” MIS
Q., vol. 28, no. 1, pp. 75–105, 2004.
K. Peffers, T. Tuunanen, M. A. Rothenberger, and S. Chatterjee, “A Design Science
Research Methodology for Information Systems Research.,” J. Manag. Inf. Syst., vol. 24,
no. 3, pp. 45–77, 2008.
B. Moingeon and L. Lehmann-Ortega, “Creation and Implementation of a New Business
Model: a Disarming Case Study,” M@ n@ gement, vol. 13, no. 4, pp. 266–297, 2010.
T. Mettler, “Towards a unified business model vocabulary: A proposition of key
constructs,” J. Theor. Appl. Electron. Commer. Res., vol. 9, no. 1, pp. 19–27, 2014.
B. Andersson, M. Bergholtz, A. Edirisuriya, T. Ilayperuma, E. Dubois, S. Abels, A. Hahn,
B. Wangler, H. Weigand, P. Johannesson, B. Grégoire, M. Schmitt, and J. Gordijn,
108
24.
25.
26.
27.
28.
“Towards a Reference Ontology for Business Models,” Proc. 25th Int. Conf. Concept.
Model., vol. 4215, pp. 482–496, 2006.
T. Kessler and J. Brendel, “Planned Obsolescence and Product-Service Systems: Linking
Two Contradictory Business Models,” Jcsm, vol. 8, no. February, pp. 29–53, 2016.
M. W. Johnson, C. M. Christensen, and H. Kagermann, “Reinventing Your Business
Model,” no. December 2009, pp. 1–10, 2008.
R. Casadesus-Masanell and J. E. Ricart, “From strategy to business models and onto
tactics,” Long Range Plann., vol. 43, no. 2–3, pp. 195–215, 2010.
B. Demil and X. Lecocq, “Business model evolution: In search of dynamic consistency,”
Long Range Plann., vol. 43, no. 2–3, pp. 227–246, 2010.
R. Casadesus-Masanell and J. E. Ricart, “Competing through business models,” Business,
vol. 3, no. 713, pp. 1–43, 2007.
109