This paper is a DRAFT - it should not be cited without the permission of the author
CLASSIFICATION IN THE SPOTLIGHT: THE
IMPORTANCE OF CLASSIFICATION TO THE
PROGRESSION OF BUSINESS MODEL RESEARCH
Susan C Lambert
Email: [email protected]
Telephone: 618 82012766
Centre for Accounting, Governance and Sustainability
University of South Australia
PO Box 2471, Adelaide,
South Australia,
Australia 5000
This paper is a DRAFT - it should not be cited without the permission of the author
ABSTRACT
Classification of research objects is critical to many management research studies with
taxonomies and typologies providing the means by which research objects are organized and
differentiated from one another. However, the nature and structure of the classifications are
frequently presupposed with little attention given to the objectives, functions and
characteristics of the classifications, not only to the immediate study but to future related
research. I bring to the fore the scientific significance of classification and its role in theory
development in organizational science and business model research. The alternative
philosophies of classification are described and their relevance to management research
proposed. Premised on the need for both deductive and inductive empirical research a case is
made for a general classification of business models that can progress the research towards
mid-range theory building
Keywords: classification, taxonomy, typology, mid-range theory, business model, research
schema
This paper is a DRAFT - it should not be cited without the permission of the author
INTRODUCTION
In this paper the nascent field of business model research is used to illustrate the significance
of classifications in progressing research from conceptualizations to mid-range theorizing. A
business model depicts the value creation and transfer strategies of the firm including
relationships with other entities in the value domain.
Business model research is an emerging area of management research that suffers from a lack
of empirical research that can pave the way to mid-range theorising and is dominated by
studies that investigate and propose definitions, components and frameworks. However, ‘the
research community is yet to invent a common language, in terms not only of terminology
but basically in terms of conceptualization, for discussing and analyzing business models’
(Pateli & Giaglis, 2004, p.312). The conceptual nature of the research has resulted in there
being a myriad of concepts, ontologies and frameworks of business models all of which have
merit, but none of which have been universally accepted. The problem is not unique to
business model research. Hanks, Watson et al. (1993a, p.11) recognised a similar problem in
relation to research into organisational life cycle models concluding that due to ‘the absence
of careful empirical analysis, a plethora of conceptually based models have emerged’, and
that evaluation of the models could only be achieved through systematic empirical research.
Small enterprise and accounting researchers found themselves in a similar position in the late
twentieth century recognising that progress would only be made through exploratory,
inductive, empirical research (McMahon, 1998).
Existing business model research is predominately conceptualisation based research
producing metaphors, definitions, descriptions, simple classifications and explanations of
relationships between business model components aimed at producing a coherent unit of
analysis. The focus of the research is on the concept itself drawing from established research
areas including organisational theory, strategy theory and information systems theory.
Frameworks and ontologies of business models provide representations of relationships
between business model components some of which are causal models with the potential for
explanatory research (Gregor, 2006).
In parallel to the concept-focussed research, studies emerged that focus on the relationships
between the business model and other management phenomena such as change management
(Chaharbaghi, Fendt, & Willis, 2003; Samavi, Yu, & Topaloglou, 2009; Voelpel, Leibold,
Tekie, & Krogh, 2005), adoption factors (Afuah, 2004; Voelpel et al., 2005) and evaluation
(Mäkinen & Seppänen, 2007; Seppänen, 2008).
The existing research has progressed the unification of the business model conceptualisation
with several recurring themes emerging (Zott, Amit, & Massa, 2011) however because of the
highly specific context in which the research is couched it is severely limited in terms of
theory building. Findings that can be generalised to the whole population under any
circumstances represent the basis for grand theories which are rare (Gregor, 2006; Neuman,
2003). More commonly, generalisations relate to different classes of objects and hold true
under specific contexts. These context-bound generalisations form the basis of mid-range
theories which explain concepts in the context of their setting (Llewellyn, 2003). Mid-range
business model theories are proposed to hold true under particular circumstances and for
This paper is a DRAFT - it should not be cited without the permission of the author
particular categories of business models rather than for all contexts or for all instances of
business models.
The need to understand what business models are being used in practice and the
characteristics that differentiate one from another is evident from the number of classification
schemes derived from or applied in the empirical studies. Numerous classification schemes
exist, each devised to meet the specific needs of the researcher however there is no exhaustive
classification of business models (Keen & Qureshi, 2006; Morris, Schindehutte, Richardson,
& Allen, 2006; Pateli & Giaglis, 2004). It is not possible for a single classification scheme
to serve all purposes although each well-conceived classification brings knowledge to the
area of research (Gilmour, 1951; Gilmour, 1940). Distinguishing one classification from
another, evaluating their utility for future research and understanding the underlying
decisions on which the classifications are based, is not always possible because to date, very
little consideration has been paid to taxonomical issues (Mäkinen & Seppänen, 2007; Morris
et al., 2006).
The paper proceeds with an explanation of the relationships between the purpose of a
classification and the necessary functions and characteristics of the classification. The
connection between the functions and characteristics, and the various philosophies of
classification is drawn followed by an overview of classification research in the
organisational sciences. The role of classification in business model research is described and
a case is made for a general classification of business models that can progress the research to
generalisations and mid-range theorising.
Based on the classification theory presented in the earlier sections of the paper a conceptual
framework (Classification Design Framework) is proposed that can guide researchers in
classification scheme design and articulation. The paper concludes with an appeal for explicit
and detailed consideration and articulation of classification design decisions that can
encourage cumulative research.
THE UNIVERSAL SIGNIFICANCE OF CLASSIFICATION
Classification is critical to the understanding of objective reality. It involves the ordering of
objects into groups or classes on the basis of their similarity and ordering of objects into
classes provides meaning to reality (Bailey, 1994; Bailey, 2005; Simpson, 1961).
Classification also aids our understanding of a knowledge domain as ‘we do not perceive,
remember and talk about each object and event as unique, but rather an instance of a class or
concept that we already know something about’ (Smith & Medin, 1981).
The action of putting things which are not identical into a group or class is so
familiar that we forget how sweeping it is. The action depends on recognizing a
set of things to be alike when they are not identical. We order them by what it is
that we think they have in common, which means by something that we feel to
be a likeness between them (Bronowski, 1951, p.21).
For centuries biological scientists have undertaken the theoretical study of classification,
taxonomy and, more recently, organizational and social scientists have recognized the
importance of classification to their research fields and adopted a more scientific approach to
This paper is a DRAFT - it should not be cited without the permission of the author
classification. Classification is a form of analytic theory that contributes to the explanation of
simple relationships between objects and attributes of objects.
Embedded in numerous management research papers are classifications of research objects as
diverse as activities within the strategy process (Eppler & Platts, 2009) and reasons for
financial report restatement (Gertsen, van Riel, & Berens, 2006). Firms are classified
according to size (Brews & Purohit, 2007), entrepreneurial orientation (Jambulingam,
Kathuria, & Doucette, 2005), industry (Yip, Devinney, & Johnson, 2009) or business model
(Baden-Fuller & Morgan, 2010; Sabatier, Mangematin, & Rousselle, 2010). Classifications
play an important role in research making it possible to study and make generalizations about
discrete, homogeneous groups of objects and, ultimately, propose mid-range theories (Rich,
1992). Without some level of consensus on the classification of objects within a field of
research, knowledge accumulation and meta-analysis are impeded (Hollenbeck, Beersma, &
Schouten, 2012).
PURPOSES, FUNCTIONS AND CHARACTERISTICS OF CLASSIFICATIONS
Throughout the classification literature, authors allude to the purpose of classification by
listing desirable characteristics, objectives or functions often without distinguishing them.
These descriptors are interrelated and deserve to be distinguished from one another (Walker,
2003). The objective of classification is the same as the purpose of the classification as it is
what the classification will be used for and therefore relates to the user. The functions and
characteristics of a ‘good’ classification scheme depend largely on the purpose of the
classification; it is necessary to know the purpose of the classification before desirable
functions can be identified.
Classifications are often developed to serve a specific purpose such as to provide a means of
reference to the groups of objects through nomenclature (Bailey, 1994; Mäkinen & Seppänen,
2007; McKelvey, 1982; Simpson, 1961; Sokal & Sneath, 1963) or to facilitate information
storage and retrieval so that data can be accessed and used in future research (Chrisman,
Hofer, & Boulton, 1988; McKelvey, 1982). With reference to political strategy, Meznar
(2002) points out that a lack of coherence among studies is due, at least in part, to the
inability to store and retrieve related research due to the absence of a general classification
scheme. Classification provides a foundation for comparative studies, the best classification
being one that permits the most useful comparative investigations (Bock, 1973). Once
objects are classified, generalizations can be made about homogeneous groups of objects
rather than the whole population (Chrisman et al., 1988; Gilmour, 1951; Mäkinen &
Seppänen, 2007; Mayr, 1982; Meznar, Chrisman, & Carroll, 1990; Simpson, 1961). The
purposes are not necessarily mutually exclusive and a carefully designed classification may
serve multiple purposes.
The terms ‘function’ and ‘characteristic’ are used interchangeably throughout the
classification literature; however it is helpful to distinguish between these terms. The
propensity for a classification to meet its stated purpose depends on functionality and it is the
characteristics of the classification that enable this.
This paper is a DRAFT - it should not be cited without the permission of the author
Some common functions of classification schemes include the ordering of objects (Bailey,
1994; McKelvey, 1982; Rich, 1992; Simpson, 1961; Sokal & Sneath, 1963), the provision of
an exhaustive and, perhaps, even definitive array of types or taxa (Bailey, 1994), to reduce
complexity and achieve parsimony (Bailey, 1994; Meznar et al., 1990) or, in contrast, to
provide versatility, i.e. suit many needs and display different aspects of the data (Bailey,
1994). Classifications can be designed to identify similarities between objects and facilitate
differentiation between objects so that homogeneous groups can be formed (Chrisman et al.,
1988). Classifications can also possess the function of identifying ideal types or archetypes
that can be used as criterion for measurement: one type can be used as the reference point and
others can be measured relative to that criterion (Baden-Fuller & Morgan, 2010; Bailey,
1994).
The characteristics of the classification enable its potential functions. The classification
systems might be hierarchically structured meaning that lower classes in the classification
system are nested within higher classes (Mayr & Bock, 2002). The classification may
generate mutually exclusive groups, present an exhaustive list of object attributes, be
conceptually elegant and use as few concepts as possible, and/or be intuitively sensible which
means the groups must reflect reality rather than simply be statistically valid (Scott, 1987).
Some functions and characteristics are broadly relevant to all classifications but the relevance
of others depends on the purpose of the classification. When designing a classification
scheme for a specific purpose, the functions and characteristics necessary for satisfying that
purpose must be incorporated into its design. Some functions and characteristics are
complementary while others conflict. For example, there is a trade-off between versatility
and parsimony. To be useful in multiple settings and for multiple purposes, many object
attributes need to be considered in the classification. However, this introduces complexity.
A parsimonious classification uses as few distinguishing attributes as possible but this limits
its utility. For example, a general purpose classification of objects that is intended for
multiple applications requires a large number, perhaps an exhaustive list, of attributes. On
the other hand, a classification that is intended for a specific, narrow purpose should, in the
interests of parsimony, use as few attributes as possible.
PHILOSOPHIES OF CLASSIFICATION
It is widely recognized that classification is a necessary step in understanding a research area,
however throughout history there has been continuous debate about the best way to classify
objects, what criteria to use, and what purpose the classification can serve. In this section, a
short historical overview of classification philosophies and principles are presented with a
view to explaining the issues associated with classification in general and then
contextualizing for management research.
Since ancient times, the natural historians worked to ‘bring order to the apparent chaos of the
natural world’ (Huxley, 2007, p.12) and for centuries, biologists have understood the
importance of classifying objects according to a general, widely accepted classification
scheme which facilitates the naming of objects and provides a common language within the
entire domain (Huxley, 2007). The importance of classification is not, however, peculiar to
biological science research. Researchers in the organizational sciences (Carper & Snizek,
This paper is a DRAFT - it should not be cited without the permission of the author
1980; Chrisman et al., 1988; McKelvey, 1982; Scott, 1987; Sells, 1964), behavioral sciences
(Mezzich & Solomon, 1980), social sciences (Bailey, 1994), and information and computer
sciences (Fettke & Loos, 2003; Vessey, Ramesh, & Glass, 2005), also recognize the value of
an empirically derived, general classification scheme for their fields of research.
Three distinct theories of classification have emerged over two thousand years of
classification research in the natural sciences: essentialism, empiricism and ancestry. The
suitability of each theory depends on the purpose of the classification and subject matter. An
overview of each theory follows.
Essentialism
Essentialism stems from the Aristotelian (384-322 BC) view that there exists a few essential
attributes which define the essence of an organism and that, by identifying these attributes,
classes of organisms can be created. Classes based on a small number of attributes
considered essential to defining the essence of the group are called monothetic groups. For
objects to qualify for membership of the group they must possess the attribute(s) used to
define the group and possession of the attribute(s) is both sufficient and necessary for
membership of the group (Bailey, 1994; McKelvey, 1982). Classifications that are the
product of essentialist philosophy are called typologies.
Arguably the most well known typology was that developed by Carl Linnaeus in 1763.
Using a technique called ‘downward classification’ Linnaeus classified organisms firstly as
belonging to the plant or animal Kingdom, then to the relevant Phylum, Class, Order, Family,
Genus, and Species. Although this classification system has recognized shortcomingsi, it
provides a simple, practical classification system that serves the study of natural history. The
enduring contribution of Linnaeus’ classification scheme is the binomial naming system
(genus-species) for all living organisms that remains in use today (Huxley, 2007; Mayr,
1982).
Empiricism
In contrast to essentialism, empiricism entails upwards classification and is based on
observations of many attributes. The classification scheme is designed to be exhaustive and
free from researcher bias (failings of the Linnaeus hierarchy). Polythetic groups of objects
are formed. Polythetic groups ‘have the greatest number of shared character [attribute] states,
and no single state is either essential to group membership or sufficient to make an [object] a
member of the group’ (Sneath & Sokal, 1973, p.21). Classifications that are the product of
empiricist philosophy are called taxonomies.
Empirically derived classification is based on Adansonian principlesii which were largely
overlooked until the advent of computer processing in the middle twentieth century that
provided the means of handling the large amounts of data necessary to produce such
classifications. These neo-Adansonian classifications have come to be known as numerical
taxonomy (Sneath & Sokal, 1973). Numerical taxonomy is a form of classification that
groups taxonomic units into taxa based on their numerically derived similarity of attributes.
According to neo-Adansonian principles of classification that form the basis of numerical
taxonomy, taxonomy is an empirical science and the quality of the classification is directly
This paper is a DRAFT - it should not be cited without the permission of the author
related to the number of attributes used to define the taxa and amount of information
contained in the taxa. A priori, all attributes have equal weighting and similarity between
entities is a function of the similarity between each of their many individual attributes. The
correlation of characters differs significantly from one taxa to another and classifications are
based on phenetic similarityiii
(Sneath & Sokal, 1973).
Numerical taxonomy requires objects to be grouped according to their observed similarity of
many attributes. The taxonomic groupings are polythetic, meaning that the members of the
group possess ‘a large number of common characters, that each character is possessed by
many [objects] in a group, but that no character is possessed by all the [objects] in the group’
(McKelvey, 1982, p.45). Numerical taxonomies evaluate affinity between objects
numerically (using multivariate techniques) and order the objects according to their degree of
affinity (McKelvey, 1982; Sokal & Sneath, 1963).
Ancestry
Since the middle nineteenth century, classifications in biological science have been phyletic
which means they are based on ancestry. Two types of phyletic classification exist, one
attributed to Charles Darwin, evolutionismiv and the other, a variation on evolutionism
attributed to Hennig (1950) called cladism (Mayr, 1982). According to evolutionism entities
are grouped based on their overall ancestral (genetic) affinities and under cladism entities are
grouped strictly according to lines of descent emphasizing the recency of branching points.
Group members have the same immediate ancestor species (McKelvey, 1982).
Ancestry relates only to living organisms and does not translate well to inanimate objects or
to objects that have no discernable ancestry and therefore has little relevance to management
research. McKelvey (1982) argues that organizations are social processes that have life-cycle
and evolutionary patterns and therefore are suitable for classification based on ancestry
although there is a dearth of support for this view. Essentialism and empiricism on the other
hand are appropriate for classification of inanimate objects and will therefore be explored and
contrasted further.
TAXONOMY V TYPOLOGY
Before proceeding further it is important to emphasize the dual meaning of taxonomy.
Taxonomical activities form part of the broader study of systematics which is ‘the scientific
study of the kinds of diversity of organisms and of any and all the relationships among them’
(Simpson, 1961, p.7). A taxonomy is an empirically derived classification of objects based
on the totality of their observable attributes. The term taxonomy is also used to refer to the
‘theoretical study of classification, including its bases, principles, procedures, and rules’
(Simpson, 1961, p.11).
This dual meaning of taxonomy can lead to confusion and misuse of the term. Researchers
who develop a classification schemes carry out taxonomic activity, yet their output, the actual
classification schemes, can be typologies or taxonomies.
Essentialist and empiricist theories of classification imply important differences in the
taxonomical approaches used to create a catalogue of objects and in the resulting catalogue
itself. The utility of those catalogues also differs. A typology is developed with a specific
This paper is a DRAFT - it should not be cited without the permission of the author
purpose in mind, is based on only a few attributes and, therefore, has limited utility
(McKelvey, 1982). By contrast, taxonomies are the result of grouping objects based on the
totality of their observable attributes. Although many researchers use the terms
interchangeably it is important to distinguish between typologies and taxonomies since they
serve different purposes and have their own limitations and strengths. Table 1 summarizes
the differences between typologies and taxonomies.
-----------------------------------
Insert Table 1 about here
-------------------------------------
Conceptually Derived Typologies
Typologies can take the form of traditional (commonsense) or theoretical classifications
(Rich, 1992; Warriner, 1984). Traditional classifications, ‘depend on implicit recognition of
the categories referred to, for there are no explicit classificatory criteria’ (Warriner, 1984,
p.134). The only function of traditional classifications is to identify and describe generic
kinds of objects that exist in the so-called ‘real world’.
Theoretical typologies are derived on the basis of a pre-existing theory such as economics,
management, strategy or entrepreneurship theory. The researcher conceptualizes and names
the ‘types’ that are relevant to the research and decides, a priori, the few attributes that
represent the essence of the object which in turn, depends on the intended purposes of the
classification. The result is a deductively-derived classification designed for a specific
purpose ‘but no matter how useful they may be in predicting certain features of special
interest to particular theories they have limited general utility’ (Warriner, 1984, p.135).
The purely theoretical-conceptual classification may have no empirical equivalents, and may
be ideal types or completely hypothetical (Bailey, 1994). Alternatively, theoretical
classifications can be populated with empirical cases what Bailey (1994) refers to as
conceptual-operational classifications: the categories have been conceptually conceived and
then empirical instances of them have been collected. All members of a category must
possess the characteristic(s) which define that category. Typologies are mostly generated
through qualitative classification rather than quantitative or statistical analysis although they
can be formed through conceptualizing types and then using cluster analysis to quantify the
empirical findings (Bailey, 1994).
As typologies categorize objects according to a limited number of attributes (often as few as
two) they are able to simplify complex concepts. However, their simplicity limits their power
to explain or predict phenomena (Hambrick, 1984). The criteria are determined by the
researcher based on their personal perspective and bias (Hambrick, 1984) making them
purpose-specific classifications. Classification can be specific in three dimensions. The
classification can be specific in terms of the taxonomic attributes being considered, the
environment or the situation to which it relates in terms of activity and timing. These
dimension are not mutually exclusive and the more specific the classification is along each of
these dimensions, the less generalisable it is (Chrisman et al., 1988).
This paper is a DRAFT - it should not be cited without the permission of the author
A typology is well suited to a specific need, conceptualizing a small number of categories
based on a few attributes and collecting data based on those categories is a practical research
method. The problem is that any increase in the number of attributes considered will lead to
a disproportionate increase in the level of complexity of the task and in the ultimate result of
the research itself.
For example, even if all dimensions [attributes] are dichotomous, the formula
for determining the number of cells [types] is 2M
, where M is the number of
dimensions. Thus for five dichotomous dimensions the typology will contain
only 25 or 32 cells, but for 12 dichotomous dimensions the number of cells is
212
or 4,096. (Bailey, 1994, p.4)
Typologies based on a few, carefully chosen attributes can serve a specific purpose well;
however as the number of attributes increases, the efficacy of the classification decreases.
Empirically Constructed Taxonomies
A taxonomy is derived empirically from inductive research (Sokal & Sneath, 1963) using
cluster or multivariate analysis (Hanks, Watson, Jansen, & Chandler, 1993b) to create taxa
(categories) based on a large number of variables. The purpose of taxonomy is to produce a
general classification of objects from which generalizations can be made, hypotheses
proposed and eventually mid-range theory generated. Eisenhart (1989, p.532) purports that
‘it is the intimate connection with empirical reality that permits the development of a testable,
relevant, and valid theory’. It is also argued that a general classification can take the place of
contingency theory because it can specify the class of objects to which the theory relates
(Carper & Snizek, 1980; McKelvey, 1975). By using a large number of variables the
researcher bias that is present in typologies is potentially reduced however there are still
many subjective decisions that need to be made. In research fields where little is known
about the object of classification and research is exploratory, the researcher must trawl the
data using as many variables as practical. The dangers with this approach are that key
variables may be overlooked and that irrelevant variables may dominate. The resultant
classification may be statistically valid but may not be intuitively sensible.
Where there is little domain knowledge, an alternative to the pure inductive method is to seek
expert opinion on variable selection, what Ketchen (2005) refers to as the cognitive approach.
In research areas that are more mature, with existing theories in relation to the object of
classification, the researcher can utilize that prior knowledge to minimize the chance of
irrelevant data obstructing the classification and to ensure all key variables are included. In
addition, where causal relationships are known, they need to be taken into account in order to
avoid overrepresentation of constructs (Ketchen, 2005). There must still be a large number of
variables; however utilizing existing theory to refine the variable set is recognized as
beneficial for classifications that are aimed at confirming existing theory.
Variables must be identified and measurement rules determined to allow data to be collected
and coded for cluster analysis. The data can be further analyzed using a range of multivariate
techniques. The aim is to minimize within-group variance and maximize inter-group
variance, thereby creating homogeneous groups. Once created, these homogeneous groups
This paper is a DRAFT - it should not be cited without the permission of the author
can be used for a multitude of research applications, enabling the study of both within-group
behavior as well as inter-group behavior.
CLASSIFICATION IN THE ORGANIZATIONAL SCIENCES
During the mid twentieth century organizational science research became dominated by an
empiricist approach which Gergen and Thatchenkery (2004) refer to as the modernist era.
The modernist era was dominated by information technology advancement in the form of
computerized data processing that facilitated an empirical approach to research. The
emphasis was to determine what was actually taking place in the ‘real world’ or as Gergen
and Thatchenkery (2004, p.242) put it, researchers were essentially ‘polishers of mirrors’
striving to isolate variables, standardize measures and assess causal relationships. The
empirical, modernist emphasis of organizational science has more recently given way to post
modern organizational science where the objective is to ‘tell it as it might become’ (Gergen
& Thatchenkery, 2004, p.242) rather than how it is; to rely less on the observations of reality
and more on potential for change. Even though the emphasis of current organizational
research has moved away from empiricism, empiricism continues to play a role albeit a less
prevalent role in organizational research.
It is postulated here that research outputs from the empirical era facilitated the move towards
postmodern research and that this empirically derived knowledge was required prior to
developing the constructionist views. Before attempting to propose theories in relation to
nascent fields of research, it is necessary to ‘polish the mirror’ and see what is actually taking
place. A critical requirement of the empirical approach to research is the classification of the
objects of interest in order that generalizations can be made.
Hanks et al. (1993) claim that empirical research was required to subjugate an explosion of
conceptually based organizational life-cycle models and that systematic empirical research
was required to evaluate the models.
McKelvey (1982) summarizes the views of other organizational science scholars (Burns,
1967; Carper & Snizek, 1980; Haas, Hall, & Johnson, 1966; Hall, 1977; Pugh, Hickson, &
Hinnings, 1969; Pugh, Hickson, Hinnings, & Turner, 1968) who reviewed and analysed
typologies of organizations arguing that the typologies were of no use for empirical research
because they failed to capture important differences among organizations due to the fact that
they rely on only a few organizational attributes. McKelvey (1982) provided the theoretical
and methodological foundation for the development of taxonomies of organizations and these
methods were applied by several scholars (Rich, 1992). Rich (1992) observes that despite the
foregoing, research obstacles that inhibit the construction of a general classification of
organizations that can aid generalizations and theory building remain. These obstacles
include the development of a generic pool of variables to use in the classification process and
the measurement, weighting and methods of collecting the data.
The twentieth century organizational scientists analysed the biological science taxonomy
research in ways that can assist with the application of taxonomy to other management
research. The theory that was developed by the natural scientists to classify living organisms
was adapted to suit an abstract object, the organization.
This paper is a DRAFT - it should not be cited without the permission of the author
BUSINESS MODELS AND CLASSIFICATIONS
Business model research finds itself in what Kuhn refers to as its ‘pre-paradigm’ period
indicating an immature science that lacks consensus and where ‘competing schools of
thought possess differing procedures, theories, even metaphysical presuppositions’ (Bird,
2008). Conceptual differences inhibit research that uses the business model as a vehicle to
collect information about other phenomena. Although some evidence of research aimed at
unifying the business model concept exists, it is underdeveloped and there is an absence of
order in the research field that obscures the path to business model theories.
The precursor to theorising is generalisation whereby the results of inductive empirical
research are used to infer general patterns of configurations and simple relationships between
variables can be hypothesised and tested. Generalisations are the product of empirical
research and ‘…it is the intimate connection with empirical reality that permits the
development of a testable, relevant, and valid theory’ (Eisenhardt, 1989 p.532).
In all forms of scientific research, including organisational science (McKelvey, 1982) and
behavioural science (Mezzich & Solomon, 1980), classification of objects within the research
domain is an important step towards other research. A good classification scheme forms the
foundation of theory development. To advance research towards theorising, it is necessary to
order the objects within the research domain since ‘theory cannot explain much if it is based
on an inadequate system of classification’ (Bailey, 1994, p.15). Classifications ‘…are
partway between a simple concept and a theory. They help to organise abstract, complex
concepts’ (Neuman, 2003 p.46). Business models are abstract, complex concepts, the
understanding of which can be enhanced through the development of a general classification
scheme.
Numerous business model classifications can be identified in the extant literature however
most treat as a ‘given’ the basis of the classification and the principles of the classification
system. That is to say, there is a lack of taxonomic research in relation to business models. It
has long been recognized that the business model literature lacks a systematic approach to the
development of taxonomies and that many of the so-called taxonomies of business models
are simply lists of existing business activities, or at best, typologies of generic kinds of
business models (Baden-Fuller & Morgan, 2010). The indiscriminate use of the terms
typology and taxonomy in the business model literature creates misunderstanding and
confusion for those attempting to understand and compare the various classification schemes.
The need for a general classification of business models remains unsatisfied.
Early business model classifications are simple traditional (otherwise known as
commonsense) classifications that ‘depend on implicit recognition of the categories referred
to, for there are no explicit classificatory criteria’ (Warriner, 1984, p.134). The only function
of the traditional classifications is to identify and describe business models that exist in the
so-called ‘real world’ and since the purpose of the classifications was to identify objects that
fit preconceived categories (for examples see Applegate, 2001; Bambury, 1998; Eisenmann,
2002; Laudon & Traver, 2003) the traditional typologies were useful. The traditional
This paper is a DRAFT - it should not be cited without the permission of the author
typologies of business models use no explicit criteria for classification and produce generic
types or shorthand descriptions of existing business models (Baden-Fuller & Morgan, 2010)
which serves the limited purpose of identification and nothing more . As few as four types
and as many as fourteen types of business model are identified in the various traditional
classifications.
Advancing from the traditional typologies, a range of theoretical classifications were created
to enable comparisons between existing business models. Most of the theoretical
classifications use market related criteria such as customer profile (Bienstock, Gillenson, & T
Sanders, 2002; Leem, Suh, & Kim, 2004), pricing and profit factors (Afuah & Tucci, 2003;
Bienstock et al., 2002; Linder & Cantrell, 2000; Weill & Vitale, 2001), transaction factors
(Wang & Chan, 2003), market configuration factors (Tapscott, Ticoll, & Lowy, 2000;
Timmers, 1998; Wang & Chan, 2003) and marketing strategy (Weill & Vitale, 2001) to
categorise business models. Internal criteria including product related factors (Bienstock et
al., 2002; Rajala & Westerlund, 2007; Timmers, 1998), resource related factors (Betz, 2002;
Weill & Vitale, 2001) and operational issues (Betz, 2002; Rajala & Westerlund, 2007) also
feature in the theoretical typologies.
Some of the theoretically derived typologies are subsequently populated with empirical
instances of the types but there are few empirically derived taxonomies of business models
present in the literature. The Italian biotechnology industry is the subject of one series of
studies (Bigliardi, Nosella, & Verbano, 2005; Nosella, Petroni, & Verbano, 2005) and two
non-industry-specific studies involve United States based firms (Malone et al., 2006; Morris
et al., 2006).
Although many specific, arbitrary classifications exist in the extant literature (Mäkinen &
Seppänen, 2007; Malone et al., 2006; Morris et al., 2006) ‘the diversity in classification
schemes, suggest much more progress is needed. In particular, the ability to generalize a
given scheme to different types of industries and ventures is limited’ (Morris et al., 2006,
p.32). As yet there is no general classification of business models that can form the basis of
mid-range theory building.
A CLASSIFICATION DESIGN FRAMEWORK
In the interests of encouraging the application of theoretical rigor to the design of
classification schemes in management research and communicating their underlying structure
to potential users, a conceptual framework for the design of classification schemes is now
proposed. Conceptual frameworks guide decision making by providing a structure that
incorporates the broad principles of the relevant practice or process. The proposed
conceptual framework for classification development can guide the development of future
classifications and assist in articulating important information regarding the classification
scheme. Based on the classification theory presented earlier in this paper a series of decision
steps are identified that lead to a classification outcome.
For taxonomy (the process) to serve a practical purpose, it is necessary to match the theory to
the application. This requires the researcher to firstly specify the purpose of the
classification. The purpose might be specific or broad; the classification might be required to
This paper is a DRAFT - it should not be cited without the permission of the author
serve several purposes or only one. Each purpose necessitates particular functions and
characteristics of the classification so the purpose must firstly be recognised.
Depending on the purpose and related characteristics and functions of the classification, the
relevant classification philosophy and related principles can be identified. For example, a
specific classification that aims only to identify types of organisational structure is consistent
with an essentialist philosophy of classification the principles of which include deriving the
types conceptually, using as few attributes as possible and forming monothetic groups.
Procedures and rules that are appropriate to the philosophy and principles must then be
selected. Continuing with the example of a specific classification, one procedure would be to
conceptualise the categories. An associated rule might be to specify the minimum and
maximum number of categories.
The steps required to design a classification scheme form a Classification Design Framework
(CDF) that is depicted in Figure 1. The first two steps in the CDF relate to the application of
the intended classification. Steps three to six equate to the definition of taxonomy proposed
by Simpson (1961); theory, principles, procedures and rules together form taxonomy, the
theoretical study of classification.
-----------------------------------
Insert Figure 1 about here
-------------------------------------
Step 1 State the purpose(s) of the classification: The three broad research purposes of a
classification are nomenclature, information storage and retrieval, and generalisations. These
purposes are not mutually exclusive and they can potentially be satisfied by a single
classification scheme although each may be better served by a specific classification designed
solely for that purpose. For example, a general classification of business models is required
to provide a foundation for comparative studies and to provide a means of reference to groups
of business models through nomenclature. Generalisations can then be made with respect to
homogeneous groups of business models as a precursor to mid-range theorising
Step 2 Identify the necessary functions and characteristics of the classification required
to satisfy the purpose(s): The relevant functions and characteristics become a checklist that
can assist in both building and evaluating the classification scheme.
The functions desirable in a general classification suitable for making generalisations and
comparative studies and providing a naming structure, include the ability to reliably and
consistently allocate business models to the appropriate category, to order the business
models, facilitate differentiation between business models and produce categories that are
collectively exhaustive and mutually exclusive and provide versatility in terms of purpose
and over time. The related functions are that it identifies similarities among business models,
produces mutually exclusive groups, presents an exhaustive list of key characteristics and it
reflect reality.
Step 3 Select the philosophical basis for the classification: The philosophical basis for the
classification must be consistent with the functions and characteristics of the classification.
For classifications of inanimate objects the choice is between essentialist and empiricist
philosophies. Essentialist philosophy of classification requires the researcher to identify the
This paper is a DRAFT - it should not be cited without the permission of the author
characteristics of the object or concept under study that capture its ‘essence’. For the
essentialist philosophy of classification to be suitable for a general business model
classification there needs to be widespread agreement on what constitutes the essence of a
business model. Business model research has not reached a level of maturity for this to be
settled. The widely varying conceptualisations of business models indicate that the ‘essence’
of the business model is not a shared perception thereby providing no support for an
essentialist approach to developing a general classification of business models.
Empiricism, on the other hand, does not rely on agreement of the underlying essence of the
concept but instead is based on all identifiable key characteristics. An empirically derived
taxonomy groups objects based on overall similarity and, because it relies on observations to
form the categories, has potential to reflect reality. The categories are collectively exhaustive
and mutually exclusive. Based on the above, an empiricist philosophy would be required for
a general business model classification.
Step 4 Identify classification principles associated with the chosen philosophical basis:
The principles capture the fundamentals of the classification philosophy. These include the
number of attributes considered necessary to identify the object, the basis of similarity and
differentiation between objects and the choice of monothetic or polythetic groupings.
A general business model classification based on an empiricist philosophy creates
polymorphic groupings and a specific business model classification based on an essentialist
philosophy creates monothetic groupings.
Step 5 Choose a procedure consistent with the principles: The specific procedures are
selected according to the choices made in the first four steps of the classification building
process. Statistical programs are frequently used to analyse the results of conceptually
derived classifications that are based on essentialist philosophy although manual techniques
are also used. Modern empirical classification, numerical taxonomy, is principally carried
out using computerised statistics programs that perform cluster analysis to identify the taxa
based on the observed variables. Cluster analysis is considered appropriate when the research
is essentially exploratory (Huberty, Jordan, & Brandt, 2005; Timm, 2002) and ‘where the
number of groups, as well as their forms, may be unknown’ (Henry, Tolan, & Gorman-Smith,
2005, p.121).
Table 2 summarises the relationships between theories, principles and procedures and
illustrates the decisions in relation to essentialism and empiricism.
-----------------------------------
Insert Table 2 about here
-------------------------------------
Step 6 Decide the rules to operationalise the procedure: The rules dictate how the
procedures are to be carried out. They include decisions regarding the definition of the
objects to be classified. In relation to classifications regarding tangible objects, this is a
relatively simple step, however, when the object is abstract and is the product of multiple
definitions, the step becomes more complex. Decision rules relating to the selection and
This paper is a DRAFT - it should not be cited without the permission of the author
measurement of variables, choice of particular statistical techniques and decision rules
associated with those techniques should be made explicit. Some of the decisions that need to
be made to implement cluster analysis include; choice of the sampling unit, population and
sample size. Variables need to be selected, measured and coded, the clustering method
selected and the methods of analysis chosen.
The CDF provides a structure of decisions that need to be made to ensure the resultant
classification is consistent with the purpose of the classification. The CDF ensures that all
important issues have been considered and documented.
CONCLUSION
Classification is an integral part of business model and other management research
(Christensen & Carlile, 2009). To progress nascent management research fields such as
business model research towards mid-range theory building generalisations must first be
made and those generalisations must be based on a sound classification of business models.
It is also necessary to understand how existing classification schemes have been developed
and the decisions underlying their structure so they can be replicated and further developed.
A classification scheme, like a good theory, is seldom finished. It is only given
interim acceptance with the understanding that further studies will tend to elaborate
and refine it, or disconfirm it (McKelvey, 1982 p.30).
An overview of the theory of classification has been presented to bring to light the significant
differences between classification schemes and their relevance to research. The differences
between typologies and taxonomies have been highlighted in order to show how each serves
different research needs. Typologies produce specific classifications, derived either through
intuition (common sense classifications) or theoretically derived, some populated with ‘real
life’ instances of the types and others remaining purely theoretical, ideal types. Taxonomies
produce general classifications as a result of inductive, empirical research that takes into
account many variables and is based on analyzing many ‘real life’ instances of the object.
Thoughtful consideration of the purpose of the classification scheme that extends beyond the
immediate requirement has the potential to create bridges between current and future
research. An awareness of the implications of classification design criteria can guide
researchers in their choice of classification schemes and knowledge of the principles that
underlie existing classification schemes increases their potential for reuse. The CDF is
proposed to facilitate the classification decision process and to explicate the decision steps for
the benefit of potential users. The pervasiveness of classification throughout all fields of
research renders it an important construct worthy of careful and explicit consideration.
This paper is a DRAFT - it should not be cited without the permission of the author
REFERENCES
Afuah, A. 2004. Business Models: A strategic management approach. New York: McGraw-Hill Irwin.
Afuah, A., & Tucci, C. 2003. Internet Business Models and Strategies (2 ed.). Boston: McGraw-Hill.
Applegate, L. 2001. E-Business Models: Making Sense of the Internet Business Landscape. In G. Dickson, & G.
DeSanctis (Eds.), Information Technology and the Future Enterprise, New Models for Managers:
49-101. Upper Saddle River NJ: Prentice Hall.
Bacharach, S. B. 1989. Organizational Theories: Some Criteria for Evaluation. Academy of Management
Review, 14(4): 496-515.
Baden-Fuller, C., & Morgan, M. S. 2010. Business Models as Models. Long Range Planning, 43(2-3): 156-
171.
Bailey, K. D. 1994. Typologies and Taxonomies: An Introduction to Classification Techniques. Los Angeles:
Sage Publications Inc.
Bailey, K. D. 2005. Typology Construction, Methods and issues. Encyclopedia of Social Measurement, 3: 889-
898.
Bambury, P. 1998. A Taxonomy of Internet Commerce, First Monday, Vol. 3.
Betz, F. 2002. Strategic Business Models. Engineering Management Journal, 14(1): 21-27.
Bienstock, C., Gillenson, M., & T Sanders. 2002. A Typology of E-Commerce Business Models. Quarterly
Journal of Electronic Commerce, 3(2): 173-182.
Bigliardi, B., Nosella, A., & Verbano, C. 2005. Business models in Italian biotechnology industry: a quantitative
analysis. Technovation, 25(11): 1299-1306.
Bird, A. 2008. Thomas Kuhn, The Stanford Encyclopedia of Philosophy, Vol. Fall 2008: Stanford University.
Bock, W. J. 1973. Philosphical Foundations of Classical Evolutionary Classification. Systematic Zoology, 22:
375-392.
Brews, P., & Purohit, D. 2007. Strategic Planning in Unstable Environments. Long Range Planning, 40(1): 64-
83.
Bronowski, J. 1951. The Common Sense of Science. London: Heineman Educational Books Ltd.
Burns, T. 1967. The Comparative Study of Organizations. In V. Vroom (Ed.), Methods of Organizational
Research: 113-170. Pittsburgh: university of Pennsylvania Press.
Carper, W. B., & Snizek, W. E. 1980. The Nature and Types of Organizational Taxonomies: An Overview. The
Academy of Management Review, 5(1): 65-75.
Chaharbaghi, K., Fendt, C., & Willis, R. 2003. Meaning, legitimacy and impact of business models in fast-
moving environments. Management Decision, 41(4): 372.
Chrisman, J. J., Hofer, C. W., & Boulton, W. R. 1988. Toward a System for Classifying Business Strategies.
The Academy of Management Review, 13(3): 413-428.
Christensen, C. M., & Carlile, P. R. 2009. Course Research: Using the Case Method to Build and Teach
Management Theory. Academy of Management Learning & Education, 8(2): 240-251.
Djelic, M. 1993. Exporting the American model: The postwar transformation of European business. Oxford,
England: Oxford University Press.
Eisenhardt, K. M. 1989. Building Theories form Case Study Research. Academy of Management Review,
14(4): 532-550.
Eisenmann, T. 2002. Internet Business Models, Text and Cases. New York: McGraw-Hill Irwin.
Eppler, M. J., & Platts, K. W. 2009. Visual Strategizing: The Systematic Use of Visualization in the Strategic-
Planning Process. Long Range Planning, 42(1): 42-74.
Fettke, P., & Loos, P. 2003. Classification of reference models: a methodology and its application. Information
Systems and eBusiness Management, 1(1): 35-53.
Gergen, K. J., & Thatchenkery, T. J. 2004. Organization Science as a Social Construction: postmodern potential
Journal of Applied Behavioral Science, 40(2): 228-249.
Gertsen, F. H. M., van Riel, C. B. M., & Berens, G. 2006. Avoiding Reputation Damage in Financial
Restatements. Long Range Planning, 39(4): 429-456.
Ghiselin, M. T. 2004. Mayr and Bock versus Darwin on genealogical classification. Journal of Zoological
Systematics & Evolutionary Research, 42: 165-169.
Gilmour, G. 1951. The development of taxonomic theory since 1851. Nature, 168: 400-402.
Gilmour, J. S. L. 1940. Taxonomy and Philosphy. In J. Huxley (Ed.), The New systematics: 461-474. Oxford:
Oxford University Press.
Gregor, S. 2006. The nature of theory in information systems. MIS Quarterly 30(3): 611-642.
Haas, J. E., Hall, R. H., & Johnson, N. J. 1966. Towards an Empirically Derived Taxonomy of Organizations. In
R. V. Brown (Ed.), Behavior in Organizations: 157-180. Athens,Ga: University of Georgia Press.
Hall, R. H. 1977. Organizations: Structure and Process (2nd ed.). Englewood-Cliffs, N.J.: Prentice-Hall.
Hambrick, D. C. 1984. Taxonomic Approaches to Studying Strategy: some Conceptual and Methodological
Issues. Journal of Management, 10(1): 27-41.
This paper is a DRAFT - it should not be cited without the permission of the author
Hanks, H. S., Watson, C. J., Jansen, E., & Chandler, G. N. 1993a. Tightening the Life-Cycle Construct: A
Taxonomic Study of Growth Stage Configurations in High-Technology Organizations.
Entrepreneuship, Theory & Practice: 5-30.
Hanks, H. S., Watson, C. J., Jansen, E., & Chandler, G. N. 1993b. Tightening the Life-Cycle Construct: A
Taxonomic Study of Growth Stage Configurations in High-Technology Organizations.
Entrepreneurship, Theory & Practice, Winter 1993: 5-30.
Henry, D. B., Tolan, P. H., & Gorman-Smith, D. 2005. Cluster Analysis in Psychology Research. Journal of
Family Psychology, 19(1): 121-132.
Hollenbeck, J. R., Beersma, B., & Schouten, M. E. 2012. Beyond Team Types and Taxonoies: A Dimensional
Scaling Conceptualization for Team Description. Academy of Management Review, 37(1): 82-106.
Huberty, C. J., Jordan, E. M., & Brandt, W. C. 2005. Cluster Analysis in Higher Education Research. In J. C.
Smart (Ed.), Higher Education: Handbook of Theory and Research, Vol. XX: 437-457: Springer.
Huxley, R. 2007. The Great Naturalists. London: Thames & Hudson.
Jambulingam, T., Kathuria, R., & Doucette, W. R. 2005. Entrepreneurial orientation as a basis for classification
within a service industry: the case of retail pharmacy industry. Journal of Operations Management,
23(1): 23-42.
Keen, P., & Qureshi, S. 2006. Organizational Transformation through Business Models: A Framework for
Business Model Design. Paper presented at the 39th Hawaii International Conference on System
Sciences - 2006, Hawaii
Lambert, S. 2006. Do We Need a General Classification Scheme of Business Models, Australasian Conference
on Information Systems (ACIS). Adelaide, Australia.
Laudon, K., & Traver, C. 2003. E-commerce; business, technology, society. Upper Saddle River, New Jersey:
Pearson Prentice Hall.
Leem, C. S., Suh, H. S., & Kim, D. S. 2004. A classification of mobile business models and its applications.
Industrial Management + Data Systems, 104(1): 78-87.
Linder, J. C., & Cantrell, S. 2000. Changing business models: surveying the landscape: Accenture Institute for
Strategic Change.
Llewellyn, S. 2003. What counts as "theory" in qualitative management and accounting research? Introducing
five levels of theorizing. Accounting, Auditing and Accountability Journal, 16(4).
Mäkinen, S., & Seppänen, M. 2007. Assessing business model concepts with taxonomical research criteria.
Management Research News, 30(10): 735-748.
Malone, T. W., Weill, P., Lai, R., D'Urso, V. T., Herman, G., Apel, T., & Woerner, S. 2006. Do Some Business
Models Perform Better than Others? MIT Sloan Research Paper (4615-06).
Mayr, E. 1982. The Growth of Biological Thought: Diversity, Evolution, and Inheritance. Cambridge:
Harvard University Press.
Mayr, E., & Bock, W. J. 2002. Classifications and other ordering systems. Journal of Zoological Systematics &
Evolutionary Research, 40(4): 169-194.
McKelvey, B. 1975. Guidelines for the empirical classification of organizations. Administrative Science
Quarterly, 20(4): 509-525.
McKelvey, B. 1982. Organizational Systematics: Taxonomy, Evolution, Classification. Berkeley: University
of California Press.
McMahon, R. G. P. 1998. Business growth and performance and the financial reporting practices of
Australian manufacturing SMEs. Unpublished PhD Research Methodology, University of New
England, Armidale.
Meznar, M. 2002. The theoretical foundations of public affairs and political strategy: Where do we go from
here? Journal of Public Affairs, 1(4): 330-335.
Meznar, M. B., Chrisman, J. J., & Carroll, A. B. 1990. Social Responsibility and Strategic Management:
Towards an Enterprise Strategy Classification. Business & Professional Ethics Journal, 10(1): 47-66.
Mezzich, J. E., & Solomon, H. 1980. Taxonomy and Behavioral Science: Comparative Performance of
Grouping Methods. London: Academic Press Inc (London) Ltd.
Morris, M., Schindehutte, M., Richardson, J., & Allen, J. 2006. Is the business model a useful strategic concept?
Conceptual, theoretical, and empirical insights. Journal of Small Business Strategy, 17(1): 27-50.
Neuman, W. L. 2003. Social Research Methods: Qualitative and Quantitative Approaches (5th ed.). Boston:
Pearson Education Inc.
Nosella, A., Petroni, G., & Verbano, C. 2005. Characteristics of the Italian biotechnology industry and new
business models: the initial results of an empirical study. Technovation, 25(8): 841-855.
Padian, K. 2004. For Darwin, 'genealogy alone' did give classification. Journal of Zoological Systematics &
Evolutionary Research, 42: 162-164.
Pateli, A., & Giaglis, G. 2004. A Research Framework for Analysing eBusiness Models. European Journal of
Information Systems, 13: 302-314.
This paper is a DRAFT - it should not be cited without the permission of the author
Pugh, D. S., Hickson, D. J., & Hinnings, C. R. 1969. An Empirical Taxonomy of Structures of Work
Organizations. Administrative Science Quarterly, 14(1): 115-126.
Pugh, D. S., Hickson, D. J., Hinnings, C. R., & Turner, C. 1968. Dimensions of Organizational Structure.
Administrative Science Quarterly, 13: 65-105.
Rajala, R., & Westerlund, M. 2007. Business models - a new perspective on firms' assets and capabilities:
observations from the Finnish software industry. International Journal of Entrepreneurship and
Innovation, 8(2): 115-125.
Rich, P. 1992. The Organizational Taxonomy: Definition and Design. Academy of Managment Review, 17(4):
758-781.
Sabatier, V., Mangematin, V., & Rousselle, T. 2010. From Recipe to Dinner: Business Model Portfolios in the
European Biopharmaceutical Industry. Long Range Planning, 43(2-3): 431-447.
Samavi, R., Yu, E., & Topaloglou, T. 2009. Strategic Reasoning about Business Models: a Conceptual
Modeling Approach. Information Systems e-Business Management, 7: 171-198.
Scott, W. R. 1987. Organizations Rational, Natural, and Open Systems (second ed.). New Jersey: Prentice-
Hall.
Sells, S. B. 1964. Toward a Taxonomy of Organizations. In W. W. Cooper, H. J. Leavitt, & S. I. M.W (Eds.),
New Perspectives in Organization Research: 515-531. New York: John Wiley & Sons. Inc.
Seppänen, M. 2008. Business Model Concept: Building on Resource Components. Tampere University of
Technology, Tampere.
Simpson, G. G. 1961. Principles of Animal Taxonomy. New York: Columbia University Press.
Smith, E. E., & Medin, D. L. 1981. Categories and Concepts. Massachusetts: Harvard University Press.
Sneath, P. H., & Sokal, R. R. 1973. Numerical Taxonomy: The Principles and Practice of Numerical
Taxonomy. San Francisco: W.H. Freeman and Company.
Sokal, R., & Sneath, P. 1963. Principles of Numerical Taxonomy: W H Freeman and Company.
Tapscott, D., Ticoll, D., & Lowy, A. 2000. Digital Capital: Harnessing the Power of Business Webs: Harvard
Business School Press.
Timm, N. 2002. Appplied Multivariate Analysis. New York: Springer-Verlag Inc.
Timmers, P. 1998. Business Models for Electronic Markets. Electronic Markets, 8(2): 3-8.
Vessey, I., Ramesh, V., & Glass, R. L. 2005. A unified classification system for research in the computing
disciplines. Information and Software Technology, 47(4): 245-255.
Voelpel, S., Leibold, M., Tekie, E., & Krogh, G. v. 2005. Escaping the Red Queen Effect in Competitive
Strategy: Sense-testing Business Models. European Management Journal, 23(1): 37.
Walker, R. G. 2003. Objectives of Financial Reporting. Abacus, 39(3): 340-355.
Wang, C., & Chan, K. 2003. Analysing the taxonomy of Internet business models using graphs, First Monday,
Vol. 8.
Warriner, C. K. 1984. Organizations and Their Environments: Essays in the Sociology of Organizations.
London: JAI Press.
Weill, P., & Vitale, M. 2001. Place to Space. Boston: Harvard Business School Press.
Yip, G. S., Devinney, T. M., & Johnson, G. 2009. Measuring Long Term Superior Performance: The UK's
Long-Term Superior Performers 1984-2003. Long Range Planning, 42(3): 390-413.
Zott, C., Amit, R., & Massa, L. 2011. The business model: recent developments and future research. Journal of
Management, 37(4): 1019 - 1042.
This paper is a DRAFT - it should not be cited without the permission of the author
TABLES
Typologies Taxonomies
The product of essentialist philosophy The product of empiricist philosophy
Categories (types) are conceptually derived Categories (taxa) are empirically derived
Few attributes considered Many attributes considered
Reasoning by deduction Reasoning by inference
Mostly qualitative classifications Quantitative classifications
Monothetic groupings Polythetic groupings
Specific classification General classification
Provides a basis for only limited generalizations Provides a basis for wider generalisation
Table 1
Summary of the Differences between Typologies and Taxonomies (adapted from Lambert, 2006))
Figure 1: Classification Design Framework (CDF)
This paper is a DRAFT - it should not be cited without the permission of the author
Philosophy Essentialism Empiricism
Associated
Principles
Categories derived conceptually
Few characters
Monothetic groups
Categories derived through observation
Many characters
Polythetic groups
Related
Procedures Conceptualize the categories
Define the sampling unit and determine the
population
Identify objects that fit the categories
Analyse the results quantitatively and/or
qualitatively
Define the sampling unit and determine the
population
Discover and measure the variables
Code the variables
Form clusters
Analyse results quantitatively
Table 2
Principles and Procedures for the Development of an Essentialist and an Empiricist Classification Scheme
This paper is a DRAFT - it should not be cited without the permission of the author
i Fails to classify all organisms and requires researcher discretion to allocate exceptions Mayr, E. 1982. The Growth of
Biological Thought: Diversity, Evolution, and Inheritance. Cambridge: Harvard University Press..
ii At about the same time that Linnaeus was proposing his classification scheme, Michel Adanson, a French botanist,
produced a ‘natural’ classification scheme designed to be exhaustive and to be free from researcher bias. This classification
scheme reflects the principles of John Ray’s ‘natural classification’ (proposed almost a century earlier) in that it is based on
observations of many attributes of the organism and it is a compositional (upward) classification Huxley, R. 2007. The Great
Naturalists. London: Thames & Hudson, Mayr, E. 1982. The Growth of Biological Thought: Diversity, Evolution, and
Inheritance. Cambridge: Harvard University Press..
iii Phenetics is a ‘system of classification based on similarities between organisms without regard to their evolutionary
relationships’ (Collins English Dictionary, 1991, p.1168). iv It should be noted that Mayr and Bock Mayr, E., & Bock, W. J. 2002. Classifications and other ordering systems. Journal
of Zoological Systematics & Evolutionary Research, 40(4): 169-194. argue strongly that Darwin’s classification of
organisms was based on two criteria, overall similarity (systematics) and descent however this is refuted by other respected
researchers Ghiselin, M. T. 2004. Mayr and Bock versus Darwin on genealogical classification. Journal of Zoological
Systematics & Evolutionary Research, 42: 165-169, Padian, K. 2004. For Darwin, 'genealogy alone' did give classification.
Journal of Zoological Systematics & Evolutionary Research, 42: 162-164.