Content uploaded by Steven Alter
Author content
All content in this area was uploaded by Steven Alter
Content may be subject to copyright.
OPINION PAPER
Defining information systems as work systems:
implications for the IS field
Steven Alter
School of Business and Management, University
of San Francisco, San Francisco, CA, U.S.A.
Correspondence: Steven Alter, School of
Business and Management, University of San
Francisco, 2130 Fulton Street, San Francisco,
CA, U.S.A.
Tel: þ1 415 422 6383;
E-mail: alter@usfca.edu
Received: 4 August 2008
Revised: 20 August 2008
Accepted: 2 September 2008
Abstract
The lack of an agreed upon definition of information system (IS) is one of many
obstacles troubling the academic IS discipline. After listing a number of
definitions of IS, this paper defines IS as a special case of work system as defined
in Alter (1999a). This definition has many desirable characteristics: it is easy to
understand; differentiates IS from information technology (IT); covers totally
manual, partially automated, and totally automated ISs; links to a life cycle
model that generates many insights about development and implementation
problems; provides a simple guideline that helps in interpreting common IS/IT
jargon; and has other useful implications related to IS concepts, IS terminology,
and the analysis and design of ISs. The paper presents the proposed IS
definition and evaluates the definition in terms of simplicity, clarity, scope,
systematic power, explanatory power, validity, reliability, and fruitfulness. An
Appendix summarizes previously published concepts and two frameworks that
flow from the proposed definition and are useful for appreciating many points
in the evaluation section.
European Journal of Information Systems (2008) 17, 448–469.
doi:10.1057/ejis.2008.37
Keywords: information system; work system; definition of information system; IS
discipline; IT antifact
The challenge of defining information system
In an EJIS editorial, Ray Paul (2007) identified five challenges related to the
state of the information system (IS) discipline. One of those challenges was
to produce a definition of the term IS. Many definitions have been
proposed over the years by researchers and textbook authors, but most are
viewed as unsatisfactory for one reason or another.
One of the reasons for the dissatisfaction is that the term IS is used to
refer to different types of objects that have many aspects in common.
Carvalho (2000) identifies four types of objects that can be viewed as ISs:
IS1: Organizations (autonomous systems) whose business (purpose) is to
provide information to their clients.
IS2: A subsystem that exists in any system that is capable of governing
itself (autonomous system). The information system (IS2) assures the
communication between the managerial and operational subsystems of
an organization – that is its purpose. When this communication is
asynchronous, a memory to store the messages is necessary. IS2 includes
such memory.
IS3: Any combination of active objects (processors) that deal only with
symbolic objects (information) and whose agents are computers or
computer-based devices – a computer-based system.
IS4: Any combination of active objects (processors) that deal only with
symbolic objects (information).
European Journal of Information Systems (2008) 17, 448 –469
&2008 Operational Research Society Ltd. All rights reserved 0960-085X/08
www.palgrave-journals.com/ejis
He says that the four types of objects ‘all deal with
information; they all are somewhat related to organiza-
tions or to the work carried out in organizations; and
they all are related to information technology, either
because they can benefit from its use or because they are
made with computers or computer-based devices.’ Table 1
lists a number of IS definitions that represent or combine
the different views of the type of object that is being
discussed. These definitions are listed more or less in
order based on the extent to which the definition
emphasizes social or organizational concerns vs technical
or mathematical concerns.
Adding to the confusion is the fact that many systems
that process information are integral parts of other
systems that do things in addition to processing informa-
tion. Remove the system that is being served and the IS
becomes meaningless. Remove the information proces-
sing and the larger system grinds to a halt.
Much of the value of this paper’s proposed definition,
and of any other definition, derives from the concepts,
Table 1 Alternative definitions of IS
Source Definition
F. Land (1985, p. 215), cited by
Magalha
˜es (1999, p. 6)
‘An information system is a social system, which has embedded in it information technology. The
extent to which information technology plays a part is increasing rapidly. But this does not prevent the
overall system from being a social system, and it is not possible to design a robust, effective information
system, incorporating significant amounts of the technology without treating it as a social system.’
Symons (1991, pp. 186–187),
cited by Magalha
˜es (1999, p. 6)
An information system is ‘a complex social object which results from the embedding of computer
systems into an organization ywhere it is not possible to separate the technical from the social factors
given the variety of human judgments and actions, influenced by cultural values, political interests and
participants’ particular definitions of their situations intervening in the implementation of such a
system.’
Paul (2007, pp. 194–195) ‘The IS is what emerges from the usage that is made of the IT delivery system by users (whose strengths
are that they are human beings, not machines). This usage will be made up of two parts: (1) First the
formal processes, which are currently usually assumed to be pre-determinable with respect to decisions
about what IT to use. y(2) Second, the informal processes, which are what the human beings who
use the IT and the formal processes create or invent in order to ensure that useful work is done.’
Davis (2000, p. 67) ‘A simple definition might be that an information system is a system in the organization that delivers
information and communication services needed by the organization.’
‘This can be expanded to describe the system more fully. The information system or management
information system of an organization consists of the information technology infrastructure, application
systems, and personnel that employ information technology to deliver information and communica-
tion services for transaction processing/operations and administration/management of an organization.
The system utilizes computer and communications hardware and software, manual procedures, and
internal and external repositories of data. The systems apply a combination of automation coming
human actions and user machine interaction.’
Lyytinen & Newman (2006, p. 4) ‘An organizational system that consists of technical, organizational and semiotic elements which are all
re-organized and expanded during ISD [information system development] to serve an organizational
purpose.’
Buckingham et al. (1987, p. 18),
cited by Avison & Myers (1995)
‘A system which assembles, stores, processes and delivers information relevant to an organization (or to
society) in such a way that the information is accessible and useful to those who wish to use it,
including managers, staff, clients and citizens. An information system is a human activity (social) system
which may or may not involve computer systems.’
UKAIS, United Kingdom Academy
for Information Systems (1997)
‘Information systems are the means by which organizations and people, utilizing information
technologies, gather, process, store, and use and disseminate information.’
Gray (2006, p. 305) ‘An automated or manual collection of people, machines, and/or methods to gather, process, transmit,
and disseminate data. Information systems are used to acquire, store, manipulate, manage, display,
transmit, or receive data. It includes both hardware and software.’
Huber et al. (2007, p. 392) ‘An organized collection of people, information, business processes, and information technology
designed to transform inputs into outputs, in order to achieve a goal.’
O’Brien (2003, p. G-10) ‘(1) A set of people, procedures, and resources that collects, transforms, and disseminates information
in an organization.
(2) A system that accepts data resources as input and processes them into information products as
output.’
Checkland & Holwell (1998, pp.
110–111), cited by Magalha
˜es
(1999, p. 6)
‘Any and every information system can always be thought of as entailing a pair of systems, one a system
which is served (the people taking the action), the other a system that does the serving [i.e., the
processing of selected data (capta) relevant to people undertaking purposeful action].’
Defining information systems as work systems Steven Alter 449
European Journal of Information Systems
frameworks, and types of analysis that are motivated by
the definition. For example, in an article about the deep
structure of ISs, Wand & Weber (1990) implicitly define
ISs as representations of real world systems: ‘Information
systems are primarily intended to model the states and
behavior of some existing or conceived real world
system.’ y‘In other words, when modeling an informa-
tion system we are not concerned with the way it is
managed in organizations, the characteristics of its users,
the way it is implemented, the way it is used, y’ Wand
and Weber’s clarity about their purpose led to a stream of
important research related to conceptual modeling (e.g.,
see Wand & Weber, 2002). On the other hand, their
approach does not take into account sociotechnical issues
that many other researchers believe important.
This paper’s definition of IS as a special case of work
system (Alter, 1999a, 2003a) leads directly to a set of
concepts, frameworks, and analysis methods that are
relevant to the four types of objects that Carvalho (2000)
identified. It also has implications in many areas of IS
practice and research. The next sections explain the
definition and evaluate it based on a slight modification
of criteria proposed by Ja
¨rvelin & Wilson (2003) for
evaluating conceptual models:
Simplicity: Simpler definitions are better, other things
being equal.
Clarity: Good definitions contain concepts that are
clear and explicit.
Scope: Good definitions cover the scope of the area of
interest and do not overlook important phenomena
and issues.
Systematic power: Good definitions help in organizing
concepts, relationships, and information related to
whatever is being defined.
Explanatory power: Good definitions help in describing
and explaining phenomena and predicting outcomes.
Validity: Good definitions lead to valid representations
and findings across the full range of relevant situations.
Reliability: Good definitions lead to relatively similar
observations and understandings when applied to the
same situation by different observers.
Fruitfulness: Good definitions lead to important ques-
tions for research and practice, and help in answering
those questions.
A concluding section reviews advantages of the pro-
posed definition and poses a challenge for other defini-
tions. The Appendix summarizes previously published
Table 1 Continued
Source Definition
Falkenberg et al. (1998, p. 73) ‘An information system is a subsystem of an organisational system, comprising the conception of how
the communication- and information-oriented aspects of an organisation are composed (e.g. of specific
communicating, information-providing and/or information-seeking actors, and of specific information-
oriented actands) and how these operate, thus describing the (explicit and/or implicit) communica-
tion-oriented and information-providing actions and arrangements existing within that organisation.’
Kroenke (2008, p. 6) A group of components that interact to produce information. [The five components of an IS are
hardware, software, data, procedures, and people.]
Laudon & Laudon (2007, p. G-7) Interrelated components working together to collect, process, store, and disseminate information to
support decision making, coordination, control, analysis, and visualization in an organization.
Rainer et al. (2007, p. 393) A process that collects, processes, stores, analyzes, and disseminates information for a specific purpose;
most ISs are computerized.
Watson (2008, p. 9) ‘An information system is an integrated and cooperating set of software directed information
technologies supporting individual, group, organizational, or societal goals.’
Jessup & Valacich (2008, p. 567) ‘Assumed to mean computer-based systems, which are combinations of hardware, software, and
telecommunications networks that people build and use to collect, create, and distribute useful
information.’
TechWeb (2008) ‘A business application of the computer. It is made up of the database, application programs and
manual and machine procedures. It also encompasses the computer systems that do the processing.’
McLeod & Schell (2007, p. 19) ‘Information systems are virtual systems; their data represent the physical system of the firm.’
Wand & Weber (1990, pp. 62–63) ‘Information systems are primarily intended to model the states and behavior of some existing or
conceived real world system.’ y‘We conceive of an information system as an object that can be
studied in its own right, independently of the way it is deployed in its organizational and social context,
and the technology used to implement it. In other words, when modeling an information system we
are not concerned with the way it is managed in organizations, the characteristics of its users, the way it
is implemented, the way it is used, the impact it has on such factors as quality of working life or the
distribution of power in organizations or the type of hardware or software used to make it operational.’
Pawlak (2002, p. 182) ‘An information system is a data table, whose columns are labeled by attributes, rows are labeled by
objects of interest and entries of the table are attribute values.’
Tadeusz & Rybnik (1992, p. 182) ‘An information system is a pair A ¼(U, A), where U is a nonempty, finite set called the universe and A is
a nonempty, finite set of attributes.’
Defining information systems as work systems Steven Alter450
European Journal of Information Systems
concepts, two frameworks, and a systems analysis
method, all of which flow from the proposed definition
and are useful for appreciating many points in this
paper’s evaluation section.
Defining an IS as a work system
The definition of an IS is based on the more general
concept of work system. Businesses operate through work
systems. Typical business organizations contain work
systems that procure materials from suppliers, manufac-
ture physical and/or informational products, deliver
products to customers, find customers, create financial
reports, hire employees, coordinate work across depart-
ments, submit tax payments, and perform many other
functions.
A work system is a system in which human participants
and/or machines perform work (processes and activ-
ities) using information, technology, and other re-
sources to produce specific products and/or services for
specific internal or external customers.
An IS is a work system whose processes and activities
are devoted to processing information, that is, captur-
ing, transmitting, storing, retrieving, manipulating,
and displaying information.
Thus, an IS is a system in which human participants
and/or machines perform work (processes and activ-
ities) using information, technology, and other re-
sources to produce informational products and/or
services for internal or external customers.
Examples of ISs include work systems devoted to
generating corporate plans, creating computer programs,
generating financial statements, creating digital products
such as software and electronic games, performing
economic analysis, writing music, rebalancing stock
portfolios, and determining prices of airline seats based
on complex yield management calculations.
This definition covers all four types of objects identi-
fied by Carvalho (2000):
IS1: Entire organizations that provide information for
clients can be viewed as work systems. The frameworks
and analysis methods described in the Appendix were
developed based on the premise that most organiza-
tions contain multiple work systems. Consequently,
the frameworks and analysis methods are somewhat
useful for viewing smaller organizations as single work
systems and less useful for large organizations that
contain multiple work systems that should be analyzed
separately.
IS2: ISs that assure communication between manage-
rial and operational subsystems of an organization can
be viewed as work systems. Throughout the develop-
ment of the work system approach, management and
communication systems were viewed as typical work
systems.
IS3: Pure computer systems can be viewed as work
systems in which machines do all of the work. The
intention to include totally automated systems within
the work system approach is one of the reasons for
using the term work system instead of Checkland’s
(1999) term human activity system.
IS4: Any combination of people and machines that deal
only with symbolic objects (information) can be
viewed as IS if that combination exists to produce
specific products and services for specific customers.
IT-reliant work systems. The fact that a work system uses
information technology (IT) extensively does not imply
that it is an IS. The following are examples of work
systems that use IT extensively but are not ISs: fulfillment
systems for physical goods, package delivery systems,
highly automated manufacturing systems, medical sys-
tems that include physical examination or treatment of
patients, and transportation systems that use IT exten-
sively. In such cases, an IS may produce intermediate
products and services that are meaningful and useful
primarily in the context of a larger work system that
involves activities beyond processing of information.
Alternatively, the processing of information may be so
intertwined with the work system that it is barely
meaningful to speak of the IS as a separate system.
Increasing reliance on computerized ISs has led to
increasing degrees of overlap between work systems and
the ISs that support them. This trend implies that a clear
IS definition should distinguish between ISs and IT-
reliant work systems (of which ISs are a special case whose
processes and activities are devoted to information and
that produce informational products and services). This
distinction is often ignored because IT-reliant work
systems are sometimes treated as subject matter within
in the IS field, an inclusion that may have significant
benefits for the IS field (Alter, 2003a, b).
Other categories of work systems. IS is only one of a
number of special cases of work system that are
important in the IS field.
Aproject is a work system designed to go out of
existence after producing specific products and services
for customers. Each major project phase or subproject
might also be viewed as a separate work system with its
own processes and activities, participants, and pro-
ducts.
Avalue chain can be viewed as a work system that
crosses several functional areas of business and whose
participants typically reside in different departments.
Each major step in a value chain can be viewed as a
subsystem that is also a work system.
Asupply chain is an interorganizational work system
devoted to procuring materials and other inputs
required to produce a firm’s economic products and
services. The firm and specific suppliers are participants
in processes and activities that use specific information
and technology to create, monitor, and fulfill orders.
An e-commerce web site can be viewed as a self-service
work system in which a customer uses the seller’s web
site in a process of matching requirements to product
Defining information systems as work systems Steven Alter 451
European Journal of Information Systems
offerings and then making the purchase. By focusing
attention on the customer’s actions and desired out-
come, a work system view helps in recognizing why an
attractive interface does not assure a web site’s success.
The fact that systems of any of these types can be viewed
as work systems implies that the same basic concepts and
analysis ideas apply for all of these cases. For example, it
is possible to use work system elements (identified by the
work system framework, in the Appendix) to summarize
any work system, including any IS, project, supply chain,
or e-commerce web site. Being able to start from the same
big picture ideas and models makes it easier to think
about a broad range of situations even though each
special case may have its own specialized terminology for
specific topics.
Previous use of ‘work system.’ Before proceeding, it is
worthwhile to note that the term work system was used
occasionally in a number of articles over the last 30 years,
including two articles by Bostrom & Heinen (1977a, b) in
the first volume of MIS Quarterly (see Table 2). It may not
have been defined clearly and treated as an analytical
concept until the third edition of an IS textbook (Alter,
1999b) and an article called ‘A General, but Useful Theory
of Information Systems’ (Alter, 1999a). The term high-
performance work system has appeared occasionally (with-
out a definition of work system) in the popular business
press and in some consulting circles to describe organiza-
tions with high degrees of participation and self-manage-
ment.
In contrast to the relatively informal use of the term
work system in the examples in Table 2, a careful
definition that overlaps with our proposed definition
but defines a work system as a context (rather than a
system) appears in Jasperson et al. (2005):
The work system represents the context within which
organizational members perform their assigned work. Thus,
the work system includes organizational members, the work
tasks undertaken by members, work processes, technology
features that enable or support work tasks and processes,
and social structures that direct organizational members
both in their work-related behaviors and in their interac-
tions with each other. yAn organization’s members are
obviously core elements of the work system, both in
performing work-related roles and as users of work-enabling
technologies. Most important, given that an organization’s
members continuously interpret their work context their
work system sensemaking becomes an especially critical
subcomponent of the work system. (p. 535)
The value of defining IS as a special case of work system is
based on the applicability of ideas and analysis methods
related to work systems. The Appendix summarizes a set
of concepts, two frameworks, and an analysis method
related to work systems. Various parts of that material
were published in various evolving versions over the last
decade. Even readers familiar with the work system
framework and work system life cycle (WSLC) model
might want to glance at the Appendix because the rest of
the paper assumes familiarity with its contents.
Evaluating the proposed definition
This section evaluates the proposed definition of IS based
on each of the criteria mentioned at the outset.
Table 2 Examples of previous uses of the term work system
Authors Representative statement
Bostrom & Heinen (1977a,
pp. 17–18
The sociotechnical systems approach ‘is used for redesigning existing work systems as well as for new
site designs.’ y‘When one intervenes in a work system, two potential improvements are possible. The
first is an improvement in task accomplishment yThe second is an improvement in the quality of
working life.’
Mumford & Weir (1979, p. 3) ‘The ETHICS method consists of a set of steps which must be taken in the design and implementation of
a new work system.’
Davis & Taylor (1979, p. xv) ‘Job design yhad its origins in attempts at comprehensive work systems design, including the social
systems within which the work systems are embedded.’
Trist (1981, p. 11) ‘Primary work systems yare the systems which carry out the set of activities involved in an identifiable
and bounded subsystem of a whole organization – such as a line department or service unit.’
Sumner & Ryan, (1994) ‘While social aspects are important in the design of work systems, these aspects are not supported by
CASE tools. yCurrent CASE tools do not fully support the process of generating multiple design
alternatives for the new work system.’
Mitchell & Zmud (1999,
pp. 425, 434)
‘When the viability of [a] new work system depends on IT, the process innovation is IT-enabled. y
Usually substantial gaps between the organization’s IT infrastructure and the IT requirements of the
adopted work system would be recognized and dealt with before planners develop and implement the
new work system.’
Land (2000, p. 116) ‘Sociotechnical methods focus on design of work systems to improve the welfare of employees. The
prime aim of redesigning work systems is the improvement of the quality of working life.’
Defining information systems as work systems Steven Alter452
European Journal of Information Systems
Simplicity
The proposed two-part definition is relatively simple and
understandable.
An IS is a work system whose processes and activities
are devoted to processing information, that is, captur-
ing, transmitting, storing, retrieving, manipulating,
and displaying information.
A work system is a system in which human participants
and/or machines perform work (processes and activ-
ities) using information, technology, and other re-
sources to produce specific products and/or services for
specific internal or external customers.
Alter (2007b) discusses how work system ideas have been
applied in over 200 (more recently over 300) student
papers. Students analyzing specific situations are reason-
ably able to identify human participants, information,
technology, products and services, and customers. They
also understand the meaning of the six types of informa-
tion processing activities and recognize examples of ITs to
perform each type of activity. For example, in a classroom
environment Petkov & Petkova (2006) found the work
system framework useful to undergraduate students
trying to understand an enterprise resource planning
(ERP) implementation situation.
Straightforward but not simplistic. Note the term ‘reason-
ably able’ in the paragraph above. Although the work
system framework is reasonably easy to understand as an
abstraction, it is sometimes challenging to apply in
situations in which the scope of the system is not known
in advance. That condition occurs in most real world
analyses of sociotechnical systems. Alter (2007b) notes
that the most experienced student teams tend to report
longer and more contentious debates about the identity
and boundaries of the system they should analyze to
produce recommendations related to a real world pro-
blem or opportunity. The less experienced student teams
sometimes seem not to realize why this is an important
issue. In both instances, as the student teams progress
with the analysis they sometimes find that their original
definition of the system is inadequate. In some cases they
even redefine the problem they are trying to solve.
Challenges in applying the proposed definition are
probably no greater than challenges in applying any
other sociotechnical definition of IS in real world
situations because the boundary and scope of the system
are usually in question. In contrast, computer-oriented
definitions can be simpler in cases where the computer
system and/or software of interest are highly localized
and defined in advance.
Clarity
Good definitions contain concepts that are clear and
explicit. The proposed definition of IS and the frame-
works that follow from it are clear enough to express and
underscore important distinctions in the IS field.
Differentiation between IS and IT. Many authors have
expressed concerns about the common blurring of IS and
IT, which is clarified by defining IS as a special case of
work system. Technology and infrastructure are elements
of the work system framework. The definition implies
that a computer is not an IS because a computer does not
produce specific products and services for specific
customers. Similarly, word processing programs and
suites of software such as ERP products are not ISs.
Rather, depending on the purpose of the analysis, they
should be treated either as part of the technology within
a specific IS or as part of the technical infrastructure
shared among multiple ISs.
Treatment of IS as a system, rather than a tool. The IS
discipline is ostensibly about systems, but many of its
fundamental ideas and viewpoints are about tools, not
systems (Alter, 2004). For example, our basic vocabulary
implies that IT vendors and IT groups provide tools and
that an IT group’s ‘users’ use them. Similarly, typical
concepts about IS success imply that a tool’s success is
measured by whether it fits specifications, how well it is
used, and what is its impact. Likewise, system develop-
ment often refers to developing software tools that meet
requirements and satisfy perceived needs of users, rather
than developing or modifying a work system in an
organization.
Table 3 compares a ‘tool view’ of IS with a ‘system view’
of IS. With a tool view, the headline is the tool that is
being used. In contrast, a system view focuses on a system
of doing something. With a tool view, the people are
users of the tool, whereas a system view of a system in an
organization treats people as participants in the system.
Recognition of the sociotechnical nature of IS. The
proposed definition of IS is a sociotechnical definition
because it includes people both as system participants
and as internal or external customers of the system.
With both participants and customers clearly in view,
the description or analysis of a system tends to include
topics such as the skills, interests, incentives, and social
relations of the people in the system. In contrast,
IS definitions that focus on hardware and software do
little to focus attention on sociotechnical issues and
concerns.
Although clearly in the sociotechnical camp, the
proposed definition does not try to establish a separation
between a technical system and social system as is
discussed by many proponents of sociotechnical ap-
proaches (e.g., Mumford & Weir, 1979; Hirschheim &
Klein, 1994). Simplicity is the main reason for taking the
more integrated view expressed in the work system
framework, whose nine elements provide useful headings
for summarizing the different parts of a work system.
Trying to separate the social system from the technical
system is more difficult. For example, in many cases
specific activities and business process steps can be
viewed simultaneously as part of the social system and
the technical system. In such cases, it is unlikely that
most business people, IT professionals, and non-Ph.D.
analysts would be willing or able to make that type of
distinction.
Defining information systems as work systems Steven Alter 453
European Journal of Information Systems
Inclusion of highly automated ISs. While recognizing the
sociotechnical nature of IS, the proposed definition also
covers highly automated ISs because it says that human
participants and/or machines perform work (processes
and activities) using information, technology, and other
resources.
Examples such as online search and e-commerce
illustrate how participants fit into a work system view
of a highly automated IS. In the case of online search,
someone wanting to retrieve information needs to
formulate the query and needs to evaluate the response
from the search engine. Even a relatively minimal view of
online search involves at least five steps: the framing of
the issue by the participant, the formulation and entry of
the query by the participant, the automated search for
the relevant links, the display of search results to the
participant, and the participant’s evaluation of whether
to accept the result or perform another query. The person
in this situation might be viewed as a user of a tool or as a
participant in an IS. Viewing the situation as an IS with a
human participant rather than a tool with a human user
leads to a richer analysis because the scope starts with
formulation of needs and ends with evaluation of the
search engine’s response. The tool/user view focuses on
the form, affordances, and limitations of the interface.
The system view includes those factors, but goes much
further in showing why the limiting factor is sometimes
the person rather than the technology.
A more extreme example is a computerized facility
monitoring system that is set up and then operates
autonomously for months or years. That type of situation
is basically a boundary case because the participant slot
becomes less and less interesting unless the analysis
focuses on how the system is initialized and/or on how
people respond to the detection of exception conditions.
Also, people tend not to be totally absent from ISs whose
main activities and processes are totally automated. In
such cases they are part of the human infrastructure that
keeps the IS operating.
Avoidance of confusions with IT artifacts. Thinking of IS as
a type of work system bypasses the unfortunate confusion
related to the term ‘IT artifact’ (Alter, 2003b). This term
was popularized by Orlikowski & Iacono’s (2001) widely
cited article that discussed five different ways of treating
the IT artifact in IS research. Unfortunately, the term
artifact has a number of meanings such as an object
produced or shaped by human craft, something viewed as
a product of human conception or agency, or an
inaccurate observation, effect, or result or an error (e.g.,
see www.dictionary.com). To many people in the IS field,
the term IT artifact strongly suggests computerized
objects that are produced and have readily discernible
Table 3 Comparing a ‘tool view’ of IS with a ‘system view’ of IS (Alter, 2004)
Area of comparison Tool view System view
Headline The tool that is used. The system of doing something.
Role of people Users of the tool. Participants in the system.
Information Whatever information is stored or processed by the tool. Whatever codified or non-codified information is produced
or used by the system.
Technology The tool is the technology or is a part of the technology. The system may use a variety of technologies that may or
may not involve IT.
Customers Users of the tool or whatever the tool produces. People who receive and use whatever the system produces.
Performance
indicators
related to
operation
Measure how well the tool operates and how well it is
used. Typical metrics include user satisfaction, uptime,
energy usage, ease of use, and degree of use.
Measure how well the system operates internally and how
good are the products and services it produces. Typical
metrics include speed, consistency, rate of output, rate of
rework, reliability, quality, and total cost to the customer.
Life cycle model A project-oriented model related to defining, creating or
acquiring, and installing the tool.
Model of long-term change in a sociotechnical system that
evolves through a series of iterations of system in operation,
initiation of changes, development efforts, and implemen-
tation of changes in the organization. The iterations often
encompass planned and unplanned change.
Ownership A tool may be owned by the organization that uses it or
by an organization that controls tools or provides shared
infrastructure.
A system is owned by the organization a part of whose work
it performs.
Performance
indicators
related to change
In a new setting, measure the tool’s diffusion and
acceptance. In a setting where the tool is already used,
measure the tool’s usefulness, success, and cost-effec-
tiveness.
For a new system that is being created, measure the
implementation effort and extent to which the system is
institutionalized in its originally desired form. For an existing
system, measure the effort involved in defining, imple-
menting, and stabilizing a change.
Main issues in ana-
lysis and
design
Produce a tool that meets requirements in a cost-
effective manner, is installed successfully, and is used as
intended.
Create or improve a sociotechnical system, assuming that
technical and social issues may be intertwined and that the
system will evolve over time.
Defining information systems as work systems Steven Alter454
European Journal of Information Systems
form and boundaries, such as a computer program,
cellphone, or iPod. In contrast, Orlikowski & Iacono’s
definition of IT artifact, ‘bundles of material and cultural
properties packaged in some socially recognizable form
such as hardware and/or software,’ leads some people to
say that anything that IT affects in a significant way is an
IT artifact. The result is a significant lack of clarity, as
illustrated by gaps between three definitions of IT artifact
in Table 4. For example, what guidance would these three
definitions provide in discerning the boundary between
situations that are affected enough by IT that they should
be called IT artifacts and other situations that should not
be called IT artifacts even though IT happens to be
present? If something that uses IT can be called an IT
artifact, why should not something that uses electricity
be called an electricity artifact? The clarity of the
proposed IS definition sidesteps these terminological
confusions.
Scope
Good definitions cover the scope of the area of interest
and do not overlook important phenomena and issues.
Aside from clarifying what is and is not an IS, the
proposed definition is broadly inclusive and can be
associated directly with most of the subject matter of
the IS field. In contrast, some of the other definitions of
IS are basically about computer systems, traditional
organizational control systems, or organizations in gen-
eral. Such definitions do little to include various types of
ISs that may not have been important in the past but are
important today.
Scope of the IS field. The distinction between ISs and IT-
reliant work systems leads to a question about the scope
of the IS field. Alter (2003b) argues that the core of
today’s ISs discipline is IT-reliant work systems that
include both pure ISs and IT-reliant work systems that
produce both physical and informational products. The
broader view including IT-reliant work systems is neces-
sary because the narrower view including only pure ISs
would place many important phenomena concerning
business processes, organizational change, and competi-
tive uses of IT outside the boundaries of the IS field.
Consistent with observations by Ackoff (1993) and
Checkland (1999) concerning the necessity of under-
standing the system that is being served whenever
analyzing a system, important aspects of IS research look
beyond the operation of ISs per se. Viewing the IS field as
the study of IT-reliant work systems, rather than just ISs
per se leads to better approaches for developing models of
IS success, understanding IS costs, understanding the
productivity paradox, understanding IS-related risks,
communicating with business professionals, and pene-
trating techno-hype and techno-centrism. For other
reasons see Alter (2003a).
Inclusion of a range of situations. Although a very
detailed look at the nine elements of the work system
framework is beyond this paper’s scope, even a brief look
at the elements illustrates the wide range of situations
covered by the definition of IS.
Customers may include both internal and external
customers who may be viewed as recipients of whatever
the IS produces or as co-producers of value in self-
service situations.
Products and services may be produced by an IS, and
different groups of customers may benefit from offer-
ings of different products and services.
Processes and activities cover much more than totally
structured processes that appear in some IS definitions
as ‘procedures.’ As happens in collaboration systems
and many other situations, various processes and
activities performed by participants may be structured,
semi-structured, or unstructured.
Table 4 What is an IT artifact?
Source Definition
Orlikowski & Iacono
(2001)
‘By and large, IT artifacts (those bundles of material and cultural properties packaged in some socially recognizable form
such as hardware and/or software) continue to be under theorized.’ (p. 121)
Five premises for theorizing about IT artifacts (p. 131) include:
1. ‘IT artifacts, by definition, are not natural, neutral, universal, or given.’
2. ‘IT artifacts are always embedded in some time, place, discourse, and community.’
3. ‘IT artifacts are usually made up of a multiplicity of often fragile and complementary components, whose
interconnections are often partial and provisional and which require bridging, integration, and articulation in order for
them to work together.’
4. ‘IT artifacts are neither fixed nor independent, but they emerge from ongoing social and economic practices.’
5. ‘IT artifacts are not static or unchanging, but dynamic.’
Benbasat & Zmud
(2003)
‘We conceptualize the IT artifact as the application of IT to enable or support some task(s) embedded within a
structure(s) that itself is embedded within a context(s).’ (p. 186). The four elements of an IT artifact include information
technology, task, task structure, and task context (Figure 1, p. 188).
Agarwal & Lucas
(2005)
‘We also recommend expanding the definition of the IT artifact from ‘‘enabling or supporting some tasks’’ to specify IT
as the integration of the processing logic found in computers with the massive stores of databases and the connectivity
of communications networks. The IT artifact includes IT infrastructure, innovations with technology, and especially the
Internet.’ (p. 394)
Defining information systems as work systems Steven Alter 455
European Journal of Information Systems
Participants include both IT users and non-users, there-
by emphasizing that the relevant participants are
people who do some of the work, not just people
who use IT.
Information includes codified and non-codified infor-
mation used and created as participants perform their
work.
Technologies include IT and other tools that should not
be categorized as IT even if they contain embedded IT
applications.
Infrastructure includes relevant human, informational,
and technical resources that are managed outside of the
work system (or IS) and are shared with other work
systems.
Environment includes the relevant organizational, cul-
tural, competitive, technical, and regulatory environ-
ment within which the work system operates, thereby
recognizing that an IS’s success depends partly on
surrounding factors that are not part of the IS.
Strategies of the work system and/or organization help
in understanding any work system but may or may not
be articulated.
Systematic power
Good definitions help in organizing concepts, relation-
ships, and information related to whatever is being
defined. Defining IS as a work system devoted to
processing information implies that most or all of the
concepts and knowledge that apply to work systems in
general also apply to ISs. For example, many concepts
and much knowledge exist related to processes, informa-
tion, and people at work. That same knowledge is equally
applicable to work systems in general and to ISs. In turn,
any additional concepts and knowledge that are relevant
to ISs should be relevant to special cases of ISs, such as
transaction processing systems, decision support systems,
or computer-aided design systems. Inheritance of con-
cepts from work systems in general to ISs in general and
then to special cases of ISs has a number of implications
for the IS field.
Body of knowledge for the IS field. The challenge of
defining the domain and core concepts of the IS field is a
perennial topic at major IS conferences (e.g., see Hirsch-
heim & Klein, 2003). Many of the frustrations with the
current IS field reflect its existence as a loose, unsettled
conglomeration of partly overlapping slices of terminol-
ogy and knowledge related to a wide range of fields such
as conceptual modeling, organization behavior, total
quality management, human communication, coordina-
tion theory, information theory, computer science, and
microeconomics.
An offshoot of the attempt to develop the work system
approach was an attempt to develop Sysperanto (Alter,
2005), a model-based ontology of the IS field. The
architecture of Sysperanto is organized around the work
system framework and the inheritance of concepts
(properties) from more general types of work systems to
less general types. Since ISs, projects, and supply chains
are all special cases of work systems, each of those special
cases should inherit concepts and knowledge that exist
for work systems in general. Work system types are
summarized in terms of the nine elements included in
the work system framework. Each of those elements is
understood through a series of ‘slices.’ For example,
decision making, communicating, and coordinating are
three of the slices that are often useful when analyzing
processes and activities within a specific work system.
Each of the slices provides a vocabulary of concepts
related to work systems or their elements. The concepts
themselves may refer to components, actions or func-
tions, characteristics, performance indicators, relation-
ships, phenomena, and generalizations.
An attempt to flesh out the concepts at the work
system level is not yet complete. The next step in
developing Sysperanto will attempt to determine the
extent to which those concepts are truly inherited by ISs
and projects, and also the extent to which the special
cases have their own unique concepts. The underlying
hope, definitely not proved at this point, is that most of
the important concepts at each level below ‘work system
in general’ will be inherited from a higher level. Figure 1
illustrates these inheritance relationships using the
example of success factors. A related paper (Sherer &
Alter, 2004) noted that over half of the risk factors in a
convenience sample of 46 papers related to IS risk were
actually risk factors related to work systems in general,
rather than IS in general, particular types of IS, or IS
projects.
Explanatory power
Ideally, the definition and its implications should
enhance our ability to explain and predict phenomena.
This section demonstrates that the proposed IS definition
and the related ideas lead to insights in interpreting past
research, understanding the relationship between IT
spending and business results, identifying and over-
coming common confusions, explaining why IT matters,
and explaining the incoherence of the IS field.
Success factors for WS
in general
Success factors
for any IS
Issues related to IS, but
not other types WS
Issues related to projects,
but not other types of WS
Success factors for
any project
Success factors for
specific types of IS
Issues related to
specific types of IS
Issues related to specific
types of projects
Success factors for specific
types of projects
Figure 1 Inheritance of success factors from work systems in
general to special cases.
Defining information systems as work systems Steven Alter456
European Journal of Information Systems
Very general questions for understanding empirical research.
Based on the definition of IS, even a rudimentary
understanding of the IS (or ISs) being studied by an
empirical research study requires knowledge of the
customers, products and services, processes and activities,
participants, information, and technologies. When read-
ing an empirical research article, an obvious question to
ask is whether the article was clear about these six
elements and how they shaped the IS or sample of
systems. Relatively obvious questions for interpreting the
design, the sample, and the results include:
Customers: Did the IS have any customers? If so, were
the customers satisfied with whatever information the
system produced?
Products and services: What did the IS(s) produce?
Processes and activities: Were the processes and activities
structured, semi-structured, or unstructured; simple or
complex; tightly coupled or loosely coupled; frequent
or infrequent; relatively controlled or relatively uncon-
trolled? And so on for other big picture characteristics.
Participants: What are the skills, interests, incentives,
and social concerns of the participants? To what extent
was the use of IT genuinely important to the partici-
pants in doing their work and being successful?
Information: What codified and uncodified information
was important in the situation and how did its quality
and other characteristics affect the outcome?
Technologies: What technologies were important in this
situation, and which characteristics and affordances of
those technologies affected the results?
These questions are often revealing. As noted by Orli-
kowski & Iacono (2001) using the terminology of IT
artifacts, a substantial amount of research in the IS field
does not refer to specific ISs and their reliance on IT.
Rather, a specific capability may be tested, such as a
computer users’ ability to discern certain information.
Alternatively, the research might look at something like
the diffusion of a particular technology without reference
to how it is actually used. Where there is an IS, it may not
be clear that the IS is actually producing something for
customers. In other cases, the research report may be
unclear about whether the IS supports structured or
unstructured business processes and activities. It may say
little or nothing about the skills, interests, or incentives
of the IS participants. Also, it may say little or nothing
about uncodified information that was important in the
situation(s) studied.
Recognizing missing links in IT success stories. Especially at
a time when the IS field is sensitive about the competitive
importance of IT, it is important to be able to identify the
missing links in IT success stories. Obviously, every story
stresses some things and de-emphasizes others in order to
attain its own goals and maintain coherence. However, a
knowledgeable interpretation of an IT success story
should not ignore what is left out. The questions
mentioned above for interpreting empirical research are
equally useful for understanding IT success stories and
identifying the missing links.
Understanding IS jargon. Discussions of systems and the
IS field as a whole are rife with confusion because words
such as system, IS, user, and implementation are used by
different people to denote different things. Work system
ideas are often useful in interpreting jargon related to ISs.
The key is to remember that an IS is a special type of work
system and to ask a simple question whenever jargon
terms such as DSS, CRM, ERP, and BI are used:
Is an information system being discussed, and if so, what are
its elements (i.e. customers, products and services, processes
and activities, participants, information, technologies)?
Sometimes it will turn out that an IS is being discussed,
such as a particular way in which management informa-
tion is provided and used for decision making in a
particular setting. In other cases, the topic will be a type
of software that is being touted, sold, or installed. For
example, a software vendor selling what it calls a
customer relationship management (CRM) system actu-
ally is selling software and possibly value-added services.
Viewing CRM software as ‘the system’ is problematic if it
leads one to forget that software is installed as part of an
IS in which business professionals perform work that
happens to use CRM software.
The unusually high failure rate of CRM has been
attributed to many factors, including the immaturity of
CRM software, difficulty in integrating CRM software
with other software of the organization, and confusion
about what CRM is supposed to do. A work system
perspective explains part of the problem. The business
goal in using CRM software is creating better ISs for
selling to customers, entering orders, providing customer
service, and performing other customer-facing work.
Starting from the premise that those are the goals, rather
than implementing CRM software, might avoid some of
the confusion that sometimes occurs when companies
launched CRM initiatives (e.g., see Schrage, 2005).
ERP is another example of jargon with multiple mean-
ings. Based on the definition of IS, commercially
purchased ERP software is not an IS. Rather, it is technical
infrastructure that is shared among multiple work
systems, some or all of which may be ISs. ERP software
and its impacts certainly fall within the scope of the IS
field, however. In this case, having a clear definition of IS
makes it easier to understand the true role of ERP software
and the necessity of configuring various modules in a
way that reflects reasonable trade-offs between possibly
inconsistent requirements of many different ISs for
various business functions.
Interpreting IS-related theories and ideas. The proposed
definition of IS is useful in interpreting a number of IS-
related theories and ideas. As an example, consider the
Delone and McLean IS Success Model.
SYSTEM QUALITY and INFORMATION QUALITY singularly
and jointly affect both USE and USER SATISFACTION.
Additionally, the amount of USE can affect the degree of
Defining information systems as work systems Steven Alter 457
European Journal of Information Systems
USER SATISFACTION – positively or negatively – as well as
the reverse being true. USE and USER SATISFACTION are
direct antecedents of INDIVIDUAL IMPACT; and lastly, this
IMPACT on individual performance should eventually have
some ORGANIZATIONAL IMPACT. (DeLone & McLean,
1992)
As noted by DeLone & McLean (2003) in an article that
updated the original model in several ways, ‘nearly 300
articles in refereed journals have referred to, and made
use of, the IS Success Model.’
Use of the definition of IS reveals that most of the
concepts in the IS success model (system, system quality,
use, user satisfaction, individual impact, and organiza-
tional impact) are problematic because the model does
not recognize the distinction between software, an IS
(defined as human participants and/or machines per-
forming work y) and the work system that the IS serves.
It is not clear whether the user of the system is using its
hardware and software to enter information or, alterna-
tively, using information it produces. If either qualifies as
IS use, then it is possible for data entry users to be
satisfied and information users to be dissatisfied, and vice
versa. In many real world situations, it is not obvious
which system (the IS or the work system that is served)
exhibits ‘system quality’ and ‘information quality,’ and
which exhibits ‘success.’ Likewise, many of the individual
and organizational impacts and satisfactions or dissatis-
factions may be more directly related to the work system
and its environment rather than to the IS.
It would be interesting to look at a variety of theories in
the IS field to see whether the definition of IS provides
insights about the theory and whether the theory
provides insights about ISs. (The web site of the Associa-
tion for Information Systems contains a page called
‘Theories in IS Research’ that summarizes over 50 theories
that are used in IS research.) In cases such as the theory of
planned behavior, the theory is stated in terms of
concepts that are relevant in most ISs (behavioral
intentions, subjective norms, behavior) and probably
describes the way most IS participants do their work. In
other cases, such as the technology acceptance model
(TAM), the theory leads to different concerns in different
situations. For example, the application and validity of
TAM depends on whether the IS involves mandatory use
of the technology (e.g., an airline’s reservation call
center), voluntary use of the technology (e.g., use of
spreadsheet software by an individual), or trial use and
experimentation related to a technology that might be
used in an operational IS but has not yet been adopted. In
yet other cases, it is not obvious how the definition of IS
helps in understanding the theory or vice versa.
Understanding why IT matters. A 2003 Harvard Business
Review article with the inflammatory title ‘IT Doesn’t
Matter’ (Carr, 2003) caused great consternation among IS
professionals and academics because it challenged the
value of most IT investments. Carr’s argument was that IT
doesn’t matter because it is a commodity that anyone can
purchase. Therefore firms should minimize their spend-
ing on IT.
The definition of IS leads to a quick and simple
response that IT matters because it allows ISs and the
work systems they support to operate in ways that
otherwise would have been difficult or impossible. By
the definition of system, skimping on the technology
within a system will have impacts on the system’s
performance unless the capabilities of the technology
truly are not relevant to system participants performing
processes and activities to produce products and services
for customers. The only time when skimping on
technology will have no impact on an IS’s performance
is when the existing system is designed with excess
technical capabilities.
Understanding the competitive impacts of IT. The defini-
tion of IS helps in understanding the competitive impacts
of IT because IT investments affect productivity and other
results only when those investments are incorporated
into ISs and the work systems they support. This leaves a
question of how to define IT investment, that is, whether
IT investment involves anything other than the purchase
and technical installation of computers, software, and
networks.
Past research related to whether IT investments lead to
greater productivity and profitability is directly relevant.
Brynjolfsson (2003) and others concluded that IT invest-
ments are positively correlated with business results, but
that ‘IT is only the tip of a much larger iceberg of
complementary investments that are the real drivers of
productivity growth. yFor every dollar of IT hardware
capital that a company owns, there are up to $9 of IT-
related intangible assets, such as human capital – the
capitalized value of training – and organizational capital
– the capitalized value of investments in new business-
process and other organizational practices.’ These busi-
ness process and organizational practices are ISs and the
IT-reliant work systems they support. Most of the
investment is in ISs and work systems rather than in IT
per se. Consequently, the competitive impacts occur only
when those additional investments occur.
Explaining the incoherence of the IS field. A conjecture in
the 2005 Sysperanto paper mentioned earlier (Alter,
2005) may explain some of the difficulties underlying
the ‘identity crisis’ (Benbasat & Zmud, 2003) of the IS
discipline. IS as a category includes transaction proces-
sing systems, MIS, DSS, CAD systems, e-commerce web
sites, expert systems, group support systems, commu-
nication systems, and many other types of IS. Worth
considering, but not fundamental to the structure of
Sysperanto, is the conjecture that the various types of ISs
differ so greatly in form and function that ISs in general
have few concepts or generalizations in common beyond
those inherited from work systems in general.
According to this ‘level-skipping conjecture,’ most of
the concepts and generalizations related to ISs in general
are inherited from work systems in general; very few
additional concepts and generalizations are related to ISs
Defining information systems as work systems Steven Alter458
European Journal of Information Systems
in general but not work systems in general; most of the
additional concepts and generalizations related to ISs are
related to unique features of specific types of IS.
This conjecture might help explain why it is so difficult
to generalize about ISs and why the IS field seems to lack
a conceptual core. It may turn out that almost all of the
useful concepts and generalizations about ISs are either
about work systems in general or about the various
special cases of ISs. The conjecture would be shown
invalid if one could identify a substantial set of valuable
concepts and generalizations that apply to all of the IS
types mentioned above but do not apply to work systems
in general.
Reliability and validity
Good definitions should lead to valid representations and
findings across the full range of relevant situations. They
should also lead to relatively similar observations and
understandings when applied to the same situation by
different observers.
Although work system ideas have been applied in over
300 student papers, the reliability and validity of the
proposed IS definition and other IS definitions have not
been tested formally, at least to the author’s knowledge.
The concepts and frameworks in the work system
approach were developed iteratively starting around
1996. At each stage, the then current version was tested
informally by evaluating the areas of success and the
difficulties experienced by MBA and Executive MBA
students trying to use it for a practical purpose. After
each semester, the papers were re-examined for implica-
tions about how the then current version of the work
system approach might be improved.
Classroom experience and personal testimonials to
date suggest that the work system framework and WSLC
model are useful to MBA and Executive MBA students,
both in class work and in their own professional work.
Field-testing of the usefulness of the IS definition and
related frameworks would require experiments or pilot
studies. After training, users would be compared to non-
users trying to perform similar tasks related to recogniz-
ing, understanding, analyzing, and/or designing ISs. That
research remains to be done.
Fruitfulness
Good definitions lead to important questions for research
and practice, and help in answering those questions. In
addition to motivating the author’s research in developing
the work system approach, the proposed definition has
been applied by a number of researchers who used work
system ideas. Ramiller (2002) reports using a version of the
work system framework within a method for ‘animating’
the idea of business process within an undergraduate class.
Siau et al. (2004) used the work system framework to
analyze the value of mobile commerce to customers. Petrie
(2004) used the work system framework in a Ph.D. thesis
examining 13 e-commerce web sites. Petkov & Petkova
(2006) demonstrated the usefulness of the work system
framework by comparing results from students who did
and did not learn about the framework before trying to
interpret the same ERP case study. Ralph & Wand (2007)
used the work system framework as an underpinning of a
proposed ontology of design concepts.
The proposed IS definition leads toward possible
insights and developments in many areas including IS-
related innovation, analysis and design from a socio-
technical perspective, communication between business
and IT professionals, systems analysis for business
professionals, better customer focus, and elevated cen-
trality of service and service metaphors.
Understanding of IS-related innovation. Understanding IS-
related innovation from a work system viewpoint is
consistent with the previously mentioned observation by
Brynjolfsson (2003) that $9 is spent on intangible assets
such as processes, training, and organizational capital for
each $1 of IT hardware capital. Seeing IS-related innova-
tion using the work system framework leads to a number
of different starting points for thinking about IS-related
innovations:
Customers: Meet different needs of existing customers;
satisfy needs of new customers.
Products and services: Improve the products and services
produced by the IS.
Processes and activities: Change big picture character-
istics of processes and activities such as degree of
structure, level of integration, complexity, rhythm,
treatment of exceptions, and so on. Alternatively,
change the details by adding, removing, or combining
steps or by changing the way specific steps are
performed.
Participants: Provide new skills; change incentives;
change social relations.
Information: Use different information or provide
information in a different form or level of detail.
Technologies: Re-configure, upgrade, or replace techno-
logies.
The assumption that IS innovation need not start with IT
and might not involve IT changes might seem unusual,
or possibly self-contradictory, but this broader view
could motivate deeper understandings of important
phenomena. Consider the disappointing adoption com-
puter-aided software engineering (CASE) technology. It
once seemed that CASE would revolutionize system
development in IT groups, but the extent of adoption
has been disappointing, as demonstrated by a lengthy
‘assimilation gap’ between acquisition of CASE and use
for 25% of new applications (Fichman & Kemerer, 1999).
Based on data through 1993, after 54 months only 24% of
CASE acquisitions in their survey had resulted in that
level of use. Around the same time, Orlikowski (1993)
compared two CASE implementations and concluded
that ‘the adoption and use of CASE tools should be
conceptualized as a form of organizational change and
that such a perspective allows us to anticipate, explain,
and evaluate different experiences and consequences
Defining information systems as work systems Steven Alter 459
European Journal of Information Systems
following the introduction of CASE tools in organiza-
tions.’ Noting that ‘even with its many benefits, most
organizations have found it difficult to implement CASE,’
Sumner & Ryan (1994) concluded that CASE tools
support technical analysis of ‘the processes and data
which are needed for correct task accomplishment by the
work system,’ but that ‘social analysis is not well-
supported by existing CASE tools.’ Looking at the work
within the IS organization, they concluded, ‘for CASE to
be effective, an organization may need to view informa-
tion systems development as a work system.’
Recognition that IS should be analyzed and designed as
sociotechnical systems. The work system method (WSM)
treats work systems, including ISs, as sociotechnical
systems. Important topics in the analysis include the
skills, knowledge, interests, and social relations of the
participants and various aspects of the environment,
including the surrounding organization’s culture, struc-
ture, and policies and procedures.
In contrast, typical systems analysis textbooks used by
IS students usually mention sociotechnical concerns in
passing but focus primarily on the creation of technical
artifacts rather than the improvement of sociotechnical
systems. Table 5 demonstrates that pattern using brief
excerpts from three systems analysis textbooks. Consis-
tent with an emphasis on building technical artifacts,
most systems analysis and design textbooks treat systems
analysis and design as the special domain of IT profes-
sionals. With some exceptions such as Mathiassen et al.
(2000), most systems analysis and design textbooks
ignore or barely mention soft systems methodology
(Checkland, 1999) and other techniques that emphasize
the social or sociotechnical nature of systems in organi-
zations.
Better communication between business and IT
professionals. There is widespread agreement about the
importance of user involvement in system development
and maintenance, yet the level and quality of user
involvement are often inadequate. Users often have
difficulty saying what they want. Even if the software
totally reflects what they requested, it often omits impor-
tant capabilities that they failed to request. At a different
organizational level, but in a similar vein, misalignment
between business and IT is an ongoing source of
frustration and inefficiency. As reported in 2004, 2005,
and 2006 issues of MIS Quarterly Executive, annual surveys
of IT executives in the Society for Information Manage-
ment (in 2003, 2004, and 2005 all identified ‘IT
and business alignment’ as the #1 management concern
(Luftman & Mclean, 2004; Luftman, 2005; Luftman
et al., 2006). The 2003 and 2005 surveys asked about
key enablers and inhibitors of alignment. For both years,
the #1 enabler was ‘IT understands the firm’s business
environment.’ In 2005, the #1 inhibitor was ‘business
communication with IT.’
These issues have been discussed for several decades
and the same issues will surely appear for years to come.
Using the proposed definition of ISs and everything that
follows from it (e.g., the work system framework and
WSLC model) could help in attaining a mutual under-
standing of how systems should be improved. This might
be an approach for improving the efficiency and quality
of collaboration between business and IT professionals.
Obviously it is necessary to compile details needed for
programming and software configuration. However,
many business professionals find it difficult to discuss
software-related details that are not directly linked to
their everyday work. In addition, IT-centered discussions
may miss many important big picture issues that can
engage business professionals and that should be dis-
cussed before launching into technical details. For
example, jumping quickly to ‘tell me what you want this
software to do’ might miss big picture issues such as
whether current processes and activities are too struc-
Table 5 Textbook views of analysis and design for ISs
Textbook Brief excerpt from description of systems analysis and design as phases of the system development life cycle (SDLC)
Dennis et al. (2002) Analysis phase: ‘The analysis phase answers the questions of who will use the system, what the system will do, and
where and when it will be used.’ (p. 5)
Design phase: ‘The design phase decides how the system will operate, in terms of the hardware, software, and network
infrastructure; the user interface, forms, and reports, and the specific programs, databases, and files that will be
needed.’ (p. 6)
Hoffer et al. (2002) Analysis phase: ‘In [the requirements definition subphase] yanalysts work with users to determine what users want
from the proposed system yThe output of the analysis phase is a description of (but not a detailed design for) the
alternative solution recommended by the analysis team. Once the recommendation is accepted yyou can begin to
make plans to acquire any hardware and system software needed to build or operate the system as proposed.’ (p. 21)
Design phase: ‘You must design all aspects of the system from input and output screens to reports, databases, and
computer processes. You must then provide the physical specifics of the system you have designedy’ (p. 21)
Kendall & Kendall
(2002)
Analysis phase: [subdivided into three phases] (pp. 10–12) y‘identifying problems, opportunities and objectives’
y.‘determining information requirements for the particular users involved’ y‘analyzing system needs’ [using tools
such as data flow diagrams and data dictionaries].
Design phase: The design phase includes: ylogical design of the IS ydata entry procedures yuser interface design
yfile or database design ycontrols and backup procedures. (pp. 12–13)
Defining information systems as work systems Steven Alter460
European Journal of Information Systems
tured or not structured enough, whether processes and
activities are too complex or not complex enough, and
whether the rhythm of the work might change for the
better. Improvements at the detailed level may yield only
marginal results if big picture issues that would have been
revealed through the proposed IS definition are never
discussed.
Systems analysis and design methods for business profes-
sionals. One of the main causes of the abysmal rate of
disappointment and failure in IT projects is inadequate
user involvement in the early stages of these projects. For
example, in the Standish Group’s biennial survey of
thousands of projects, ‘lack of user involvement tradi-
tionally has been the number one reason for project
failure. Conversely, the number one contributor to pro-
ject success has been user involvement. Even when
delivered on time and on budget, a project can fail if it
does not meet user needs or expectations’ ( Johnson et al.,
2001). Markus & Mao’s (2004) much more nuanced and
detailed survey of the current understanding of user
participation proposes that the quality of engagement by
business professionals is important because ‘[user] parti-
cipation richness is related to [both] solution develop-
ment and solution implementation success’ (p. 535).
There is little or no evidence that existing systems
analysis and design methods for IT professionals are
effective tools for business professionals working without
the direct support of IT professionals. Discussions at
conferences usually find substantial agreement about the
difficulty of teaching techniques such as data flow
diagrams and entity relationship diagrams to non-
technical business students in introductory IS courses.
Discussions questioning whether Unified Modeling
Language (UML) is used effectively by IT professionals
(e.g., Erickson & Siau, 2004; Dobing & Parsons, 2006) give
inconsistent results about how UML is used and the
extent to which it is used by IT professionals. It seems
quite unlikely that it is an appropriate tool for direct use
by typical business professionals, even if they can under-
stand verbal explanations of use case diagrams that have
been produced for them. The WSM and other methods
with greater sociotechnical content and closer associa-
tion with the proposed IS definition might lead to better
results.
Keeping the customer in sight. Given that just about
everyone agrees (or at least claims publicly) that organi-
zations exist to serve their customers and that the
customer comes first (or at worst second), it would be
good if a definition of IS helped in keeping customers in
sight. The proposed definition does exactly that.
Defining an IS as a work system devoted to processing
information keeps the customer in sight because work
systems exist to produce products and services for
internal and/or external customers. The form and con-
tent of the work system framework (Figure 2 in the
Appendix) emphasize the importance of customers by
placing the customers at the top and showing that the
purpose of the system’s processes and activities is to
produce products and services for customers.
Going a step further, the elements of a work system can
be used as a basis for evaluating the customer-centricity
of any work system (or IS) and for adjusting the system to
attain the right degree of customer-centricity. The idea of
customer-centricity has become commonplace, but is
often vague. The classification of an IS as customer-
centric or not is far less important than the use of
dimensions of customer-centricity to respond to custo-
mer needs. Table 6 shows how work system elements
point to 12 dimensions of customer-centricity. Any
existing or proposed IS can be evaluated along each of
these dimensions (e.g., on a scale from 0 to 3) as part of
the effort to find the appropriate trade-offs between
T
N
E
M
N
O
R
I
V
N
E
CUSTOMERS S
T
R
A
T
E
G
I
E
S
I N F R A S T R U C T U R E
PRODUCTS & SERVICES
PROCESSES & ACTIVITIES
INFORMATIONPARTICIPANTS TECHNOLOGIES
Figure 2 The work system framework (slightly updated) (Alter, 2007a, 2008).
Defining information systems as work systems Steven Alter 461
European Journal of Information Systems
various aspects of customer-centricity and other impor-
tant design factors (Alter, 2007c).
Elevating the centrality of service and service metaphors in
the IS field. Services comprise nearly 75% of the U.S.
economy (Horn, 2005). Recognizing the large percentage
of its revenues that services produce, IBM has promoted a
major initiative to encourage the development of
‘services science’ along with the development of instruc-
tional programs in SSME (services science, management,
and engineering). The July 2006 edition of the Commu-
nications of the ACM contained a special section on
services science that included 13 papers such as
Chesbrough & Spohrer (2006), Bitner & Brown (2006),
and Maglio et al. (2006). Editorial notes in Information
Systems Research (Rai & Sambamurthy, 2006) covered
similar territory with special emphasis on opportunities
for IS scholars.
The existing IS field focuses on services in a variety of
ways. All systematic services are produced through
service systems that rely on ISs, and in many cases are
tightly integrated with ISs that exist within or across
organizations. IT groups often think and speak of IT users
as customers. SERVQUAL, a multi-item scale developed to
assess customer perceptions of service quality in service
and retail businesses (Parasuraman et al., 1985) has been
applied to services provided by IT groups (see Association
for Information Systems, 2008). On the technical side,
the traditional concept of client-server has developed
into service-oriented architecture and extensive discus-
sions of service-oriented enterprises and service-oriented
infrastructures.
Given the groundswell of attention to services, it might
be beneficial for the IS field to elevate the centrality of
service concerns and service metaphors. The content of
the proposed IS definition and the form of the work
system framework (Figure 2) directly support that type of
thrust. Greater attention to service systems in the IS field
would place greater emphasis on customer value, the
customer experience, and the customer’s shared respon-
sibility for producing whatever a work system produces
and attaining value from it. It might also develop bridges
between social and technical views of service. An
integrated view of service in IS field might lead to new
types of systems analysis and new insights for research
about how IT-reliant work systems and organizations
operate (see Alter (2007a, 2008) for steps in that
direction).
Conclusion
This paper addressed the fifth of Paul’s (2007) five
challenges by answering the question ‘What is an
Information System?’ He said that ‘since many people
are studying IS from a variety of perspectives, maybe it
should be no surprise that there are a variety of
definitions. But then, how would Society know what IS
is and what it can do if there is no clear understanding?’
Addressing all five challenges. The IS definition offered in
response to Paul’s fifth challenge might lead to responses
to his other four challenges as well:
1. Nobody knows who we are outside of the IS community.
Response: Do things that are of greater interest outside of
the IS community. The proposed IS definition links directly
to many different fields. It does the opposite of encouraging
us to focus tightly on arcane, and often transient, issues
within the IS discipline.
2. Demand for students to study IS is generally dropping,
and it is particularly rapid in the U.S.A.
Response: The definition leads toward broadening the IS
field’s appeal by making it more understandable to most
business professionals and students. Instead of hunkering
down and focusing closely on a more limited set of topics,
this may involve broadening and establishing alliances with
other fields.
3. Research publications in IS do not appear to be publish-
ing the right sort or content of research.
Response: Addressing this assertion is beyond this paper’s
scope. However, the proposed definition, and the linkage to
other fields might result in research that is more related to
real world situations.
4. Journal League Tables (pressure to publish in a very small
number of journals)
Table 6 Dimensions of customer-centricity in work systems
Work system element Dimension
Customer Recognizing and responding fully to customer needs
Providing a satisfying customer experience
Products and services Producing customized products and services
Processes and activities Personalizing or customizing processes and activities
Using customer information to maximize benefits for customers
Relying on co-production or self-service by customers
Participants Non-customer participants recognize and emphasize customer needs and priorities
Information Availability of customer-related information to maximize benefits for customers
Technology For any technology used by customers, personalization or conformity to customer work practices, standards,
terminology, convenience, or tastes
Infrastructure Avoidance of interfering with or operating incompatibly with relevant aspects of the customer’s infrastructure
Environment Operating consistent with the customer’s environment wherever the customer is involved with co-production
Strategy Producing products and services that are consistent with the customer’s strategies
Defining information systems as work systems Steven Alter462
European Journal of Information Systems
Response: Proceeding with the broad view of IS implied by
the proposed IS definition might encourage the develop-
ment of new content, new journals and new types of
publications that address information system issues more
broadly.
Comparison with alternative definitions. In addition to
presenting the proposed definition, this paper includes
in Table 1 a lengthy list of alternative IS definitions.
Comparing the scope and relative benefits of the
proposed definition and the other definitions is beyond
this paper’s scope and surely beyond the patience of
readers. Nonetheless, this paper’s discussion of concepts
and understandings that follow from the proposed
definition presents a challenge for any other definition.
A better definition should have substantial advantages
over the proposed definition in terms of simplicity,
clarity, scope, systematic power, explanatory power,
validity, reliability, and fruitfulness.
Advantages of the proposed definition. Suffice it to say that
the proposed IS definition is very general and has a large
number of advantages:
It is understandable.
It leads to and organizes a layered set of concepts that
can be used to summarize existing and proposed
systems and to analyze those systems in some depth
(see the Appendix).
It provides a basis for communication between busi-
ness and IT professionals and between IS researchers
who focus on different aspects of the IS field using
different methods.
Potentially, it could help in cataloguing past research
and developing a body of knowledge for the IS field.
Potentially, it could help in explaining the IS field to
Society.
Potentially, it could lead to new developments and
extensions of the IS field.
This paper showed that merely providing a definition or
meaningful list of topics is not sufficient. In addition to
covering the current IS field, an IS definition that matters
needs to motivate insights and research about important
questions and problems. It should help practitioners and
educators. It should provide direction for researchers. The
paper’s coverage of a wide range of topics demonstrates
that the proposed IS definition has many beneficial
characteristics and leads in many potentially fruitful
directions. A better alternative should rate comparably in
terms of the evaluation criteria and should have other
advantages that the proposed definition lacks.
About the author
Steven Alter is Professor of Information Systems at the
University of San Francisco. He earned a Ph.D. from MIT
and extended his thesis into one of the first books on
decision support systems. He served for 8 years as Vice
President of Consilium, a manufacturing software firm
that went public in 1989 and was acquired by Applied
Materials in 1998. His research for the last decade has
concerned developing systems analysis concepts and
methods that can be used by typical business profes-
sionals and can support communication with IT profes-
sionals. His 2006 book, The Work System Method:
Connecting People, Processes, and IT for Business Results,is
a distillation and extension of ideas in 1992, 1996, 1999,
and 2002 editions of his information system textbook.
His articles have been published in Harvard Business
Review,Sloan Management Review,MIS Quarterly,IBM
Systems Journal,European Journal of Information Systems,
Decision Support Systems,Interfaces,Communications of the
ACM,Communications of the AIS,CIO Insight, and many
conference proceedings.
References
ACKOFF RL (1993) Presentation at the Systems Thinking in Action
Conference, Cambridge, MA, cited by Silver M, MARKUS ML and BEATH
CM (1995). The information technology interaction model: a
foundation for the MBA core course. MIS Quarterly 3(19), 361–390.
AGARWAL RandLUCAS Jr HC (2005) The information systems identity
crisis: focusing on high-visibility and high-impact research. MIS
Quarterly 29(3), 381–398.
ALTER S (1999a) A general, but useful theory of information systems.
Communications of the Association for Information Systems 1(13),
[WWW document] http://cais.aisnet.org/artic les/default.asp?vol ¼1
&art ¼13.
ALTER S (1999b) Information Systems: A Management Perspective 3rd edn,
Prentice-Hall, Upper Saddle River, NJ.
ALTER S (2003a) 18 reasons why IT-reliant work systems should replace
‘The IT Artifact’ as the core subject matter of the IS field. Communica-
tions of the Association for Information Systems 12(23), 365–394
[WWW document] http://cais.aisnet.org/articles/default.asp?vol ¼12
&art ¼23.
ALTER S (2003b) Sidestepping the IT Artifact, scrapping the IS Silo,
and laying claim to ‘Systems in Organizations’. Communications
of the Association for Information Systems 12(30), 494–526 [WWW
document] http://cais.ais net.org/articles/default.asp?vol ¼12&art ¼
30.
ALTER S (2004) Desperately seeking systems thinking in the IS discipline.
Proceedings of ICIS-25, the International Conference on Information
Systems, pp 757–769, Washington, DC.
ALTER S (2005) Architecture of Sysperanto – a model-based ontology of
the IS field. Communications of the Association for Information Systems
15(1), 1–40 [WWW document] http://cais.aisnet.org/articles/
default.asp?vol ¼15&art ¼1.
Defining information systems as work systems Steven Alter 463
European Journal of Information Systems
ALTER S (2006) The Work System Method: Connecting People, Processes,
and IT for Business Results. Work System Press, Larkspur, CA.
ALTER S (2007a) Service responsibility tables: a new tool for analyzing and
designing systems. AMCIS 2007 – Americas Conference on Information
Systems, Keystone, CO.
ALTER S (2007b) Pitfalls in analyzing systems in organizations. Journal of
Information System Education 17(3), 295–302.
ALTER S (2007c) Customer-centric systems: a multi-dimensional view.
Proceedings of WeB 2007, Sixth Workshop on eBusiness, pp 130–141,
Montreal, Canada.
ALTER S (2008) Service system fundamentals. IBM Systems Journal 47(1),
71–85 [WWW document] http://www.research.ibm.com/journal/sj/
471/alter.html.
ASSOCIATION FOR INFORMATION SYSTEMS (2008) Theories in IS research. Web
site viewed on 2 February 2008 at [WWW document] http://
www.istheory.yorku.ca/.
AVISON DE and MYERS MD (1995) Information systems and anthropology:
an anthropological perspective on IT and organizational culture.
Information Technology & People 8(3), 43–46.
BENBASAT IandZMUD RW (2003) The identity crisis within the IS
discipline: defining and communicating the discipline’s core proper-
ties. MIS Quarterly 27(2), 183–194.
BITNER MJ and BROWN SW (2006) The evolution and discovery of
services science in business schools. Communications of the ACM 49(7),
73–79.
BOSTROM RP and HEINEN JS (1977a) MIS problems and failures: a
socio-technical perspective. Part I: The causes. MIS Quarterly 1(3),
17–32.
BOSTROM RP and HEINEN JS (1977b) MIS problems and failures: a
socio-technical perspective. Part II: The application of socio-technical
theory. MIS Quarterly 1(4), 11–28.
BRYNJOLFSSON E (2003) The IT productivity gap. Optimize 21
[WWW document] http://ebusiness.mit.edu/erik/Optimize/pr_roi.
html.
BUCKINGHAM RA, HIRSCHHEIM RA, LAND FF and TULLY CJ Eds (1987)
Information Systems Education: Recommendations and Implementation.
Cambridge University Press, Cambridge, UK pp 204–214.
CARR N (2003) IT doesn’t matter. Harvard Business Review 81(5),
41–49.
CARVALHO JA (2000) Information system? Which one do you mean? In
Information Systems Concepts: An Integrated Discipline Emerging
Proceedings of the ISCO 4 Conference (FALKENBERG E, LYYTINEN Kand
Verrijn-STUART, Eds), pp 259–280, Leiden, Holland, 20–22 September
1999, Kluwer Academic Publishers. Viewed on 1 February 2008 at
[WWW document] http://piano.dsi.uminho.pt/Bjac/SI/zdocumentos/
IS1234.pdf.
CHECKLAND P (1999) Systems Thinking, Systems Practice (Includes a
30-year retrospective), John Wiley & Sons, Chichester, UK.
CHECKLAND PandHOLWELL S (1998) Information, Systems, and Information
Systems: Making Sense of the Field. John Wiley & Sons, Chichester, UK.
CHESBROUGH HandSPOHRER J (2006) A research manifesto for services
science. Communications of the ACM 49(7), 35–40.
DAVIS GB (2000) Information systems conceptual foundations: looking
backward and forward. In Organizational and Social Perspectives
on Information Technology (BASKERVILLE R, STAGE JandDEGROSS JI, Eds),
pp 61–82, Kluwer Academic Publishers, Boston.
DAVIS LE and TAYLOR JC Eds (1979) Design of Jobs 2nd edn, Goodyear
Publishing Company, Santa Monica, CA.
DELONE WH and MCLEAN ER (1992) Information systems success: the
quest for the dependent variable. Information Systems Research 3(1),
60–95.
DELONE WH and MCLEAN ER (2003) The DeLone and McLean model of
information systems success. Journal of Management Information
Systems 19(4), 9–30.
DENNIS AB, WIXOM HandTEGARDEN D (2002) Systems Analysis & Design:
An Object Oriented Approach with UML. John Wiley & Sons, New York,
NY.
DOBING BandPARSONS J (2006) How the UML is used. Communications of
the ACM 49(5), 109–113.
ERICKSON JandSIAU K (2004) Theoretical and practical complexity of
UML. Proceedings of the Tenth Americas Conference on Information
Systems, pp 1669–1674, New York, NY.
FALKENBERG ED, HESSE W, LINDGREEN P, NILSSON BE, OEI JLH, ROLLAND C,
STAMPER RK, VAN ASSCHE FJM, VERRIGN-STUART AA and VOSS K (1998)
A framework of information system concepts: the FRISCO report. IFIP,
ISBN 3-901882-01-4, viewed on 2 September 2008 at [WWWdocu-
ment] http://www.mathematik.uni-marburg.de/Bhesse/papers/fri-full.
pdf.
FICHMAN RG and KEMERER CF (1999) The illusory diffusion of innovation:
an examination of assimilation gaps. Information Systems Research
10(3), 255–275.
GRAY P (2006) Manager’s Guide to Making Decisions about Information
Systems. John Wiley & Sons, Boston, MA.
HILL C, YATES R, JONES CandKOGAN SL (2006) Beyond predictable
workflows: enhancing productivity in artful business processes. IBM
Systems Journal 45(4), 663–682.
HIRSCHHEIM RandKLEIN HK (1994) Realizing emancipatory principles for
information systems development: the case for ETHICS. MIS Quarterly
18(1), 83–109.
HIRSCHHEIM RandKLEIN HK (2003) Crisis in the IS field? A critical reflection
on the state of the discipline. Journal of the Association for Information
Systems 4(5), 237–293.
HOFFER JA, GEORGE JF and VALACICH JS (2002) Modern Systems Analysis &
Design. Prentice-Hall, Upper Saddle River, NJ.
HORN P (2005) The new discipline of services science. BusinessWeek,
viewed on 1 September 2008 at [WWW document] http://www.
businessweek.com/technology/content/jan2005/
tc20050121_8020.htm.
HUBER MW, PIERCY CA and MCKEOWN PG (2007) Information Systems:
Creating Business Value. John Wiley & Sons, Hoboken, NJ.
Ja¨RVELIN KandWILSON TD (2003) On conceptual models for information
seeking and retrieval research. Information Research 9(1), Paper 163
viewed on 1 September 2008 at [ http://InformationR.net/ir/9-1/
paper163.html ].
JASPERSON J, CARTER PE and ZMUD RW (2005) A comprehensive
conceptualization of post-adoptive behaviors associated with
information technology enabled work systems. MIS Quarterly 29(3),
525–557.
JESSUP LandVALACICH J (2008) Information Systems Today: Managing in
the Digital World 3rd edn, Pearson Prentice-Hall, Upper Saddle River,
NJ.
JOHNSON J, BOUCHER KD, CONNORS KandROBINSON J (2001) Collaborating
on project success. Sponsored Supplement Software Magazine, viewed
on 2 February 2008 at [WWW document] http://www.softwaremag.
com/archive/2001feb/CollaborativeMgt.html.
KENDALL KE and KENDALL JE (2002) Systems Analysis and Design 5th edn,
Prentice Hal, Upper Saddle River, NJ.
KROENKE DM (2008) Experiencing MIS. Pearson Prentice-Hall, Upper
Saddle River, NJ.
LAND F (1985) Is an information theory enough? The Computer Journal
28(3), 211–215.
LAND F (2000) Evaluation in a socio-technical context. Proceedings of IFIP
W.G.8.2 Working Conference 2000, IS2000: The Social and Organiza-
tional Perspective on Research and Practice in Information Systems,
Aalberg, Denmark.
LAUDON KC and LAUDON JP (2007) Management Information Systems:
Managing the Digital Firm 10th edn. Pearson Prentice-Hall, Upper
Saddle River, NJ.
LUFTMAN J (2005) Key issues for IT executives for 2004. MIS Quarterly
Executive 4(2), 269–285.
LUFTMAN J, KEMPAIAH RandNASH E (2006) Key issues for IT executives
2005. MIS Quarterly Executive 5(2), 27–45.
LUFTMAN JandMCLEAN ER (2004) Key issues for IT executives. MIS
Quarterly Executive 3(2), 89–104.
LYYTINEN KandNEWMAN M (2006) Punctuated equilibrium, process
models and information system development and change: towards a
socio-technical process analysis. Sprouts: Working Papers on Information
Environments, Systems, and Organizations 6(1), 1–48 viewed on 1
September 2008 at [WWW document] http://sprouts.case.edu/2006/
060101.pdf.
MAGALHA
˜ES R (1999) Organizational implementation of information
systems: towards a new theory. Ph.D. Thesis, London School of
Economics. [WWW document] http://www.lse.ac.uk/collections/
informationSystems/pdf/theses/magalhaes.pdf.
Defining information systems as work systems Steven Alter464
European Journal of Information Systems
MAGLIO PP, SRINIVASAN S, KRUELEN JT and SPOHRER J (2006) Service systems,
service scientists, SSME, and innovation. Communications of the ACM
49(7), 81–85.
MARKUS ML and MAO JY (2004) Participation in development and
implementation – updating an old, tired concept for today’s IS
contexts. Journal of the Association for Information Systems 5(11:14),
514–544.
MATHIASSEN LA, MUNK-MADSEN PA, NIELSEN PA and STAGE J (2000) Object
Oriented Analysis & Design. Marko Publishing, Aalborg, Denmark.
MCLEOD Jr R and SCHELL GP (2007) Management Information Systems 10th
edn. Pearson Prentice-Hall, Upper Saddle River, NJ.
MITCHELL VL and ZMUD RW (1999) The effects of coupling IT and
work process strategy in redesign projects. Organization Science 4(10),
424–438.
MUMFORD EandWEIR M (1979) Computer systems in Work Design – the
ETHICS method. John Wiley & Sons, New York.
O’BRIEN JA (2003) Introduction to Information Systems: Essentials for the
e-Business Enterprise 11th edn. McGraw Hill – Irwin, Boston, MA.
ORLIKOWSKI WJ (1993) CASE tools as organizational change: investigating
incremental and radical changes in systems development. MIS
Quarterly 17(3), 309–340.
ORLIKOWSKI WJ and IACONO CS (2001) Research commentary: desperately
seeking the ‘‘IT’’ in IT research – a call to theorizing the IT artifact.
Information Systems Research 12(2), June 2001, 121–134.
PARASURAMAN A, BERRY LL and ZEITHAML VA (1985) A conceptual model of
service quality and its implications for future research. Journal of
Marketing 49(4), 41–50.
PAUL RJ (2007) Challenges to information systems: time to change.
European Journal of Information Systems 16(3), 193–195.
PAWLAK Z (2002) Rough sets, decision algorithms, and Bayes’ theorem.
European Journal of Operational Research 136, 181–189.
PETKOV DandPETKOVA O (2006) The work system model as a tool for
understanding the problem in an introductory IS project. Proceedings
of the 23rd Information Systems Education Conference (ISECON 2006),
Dallas, TX.
PETRIE DE (2004) Understanding the impact of technological disconti-
nuities on information systems management: the case of business-to-
business electronic commerce. Ph.D. Thesis, Claremont Graduate
University.
RAI AandSAMBAMURTHY V (2006) Editorial notes – the growth of interest
in services management: opportunities for information system
scholars. Information Systems Research 17(4), 327–331.
RAINER RK, TURBAN EandPOTTER RE (2007) Introduction to Information
Systems. John Wiley & Sons, Hoboken, NJ.
RALPH PandWAND Y (2007) An ontology of design concepts. Working
Paper Presented at the Theory Development Workshop of the Journal
of the Association of Information Systems. Montreal, Canada, 12
December 2007.
RAMILLER N (2002) Animating the concept of business process in the core
course in information systems. Journal of Informatics Education and
Research 3(2), 53–71 viewed on 1 September 2008 at [WWW
document] http://iaim.aisnet.org/jier/V3N2/.
SCHRAGE M (2005) IT’s hardest puzzle. CIO Magazine, (blog) 9 August.
Viewed on 1 February 2008 at [WWW document] http://www.cio.
com.au/index.php/id;1031171136.
SHERER SA and ALTER S (2004) Information system risks and risk factors: are
they mostly about information systems? Communications of the
Association for Information Systems 14(2), 29–64.
SIAU K, SHENG H and NAH FF-H (2004) The value of mobile commerce to
customers. Proceedings of the Third Annual Workshop on HCI Research in
MIS, Washington, DC, pp 65–69.
SUMNER MandRYAN T (1994) The impact of CASE: can it achieve
critical success factors? Journal of Systems Management 6(45),
16–21.
SYMONS VJ (1991) Impacts of information systems: four perspectives.
Information and Software Technology 33(3), 181–190.
TADEUSZ LandRYBNIK J (1992) Rough sets and some aspects of logic
synthesis. In Intelligent Decision Support: Handbook of Applications and
Advances of the Rough Sets Theory (SLOWINSKI R, Ed), pp 181–201,
Kluwer, Dordracht, Netherlands.
TECHWEB (2008) Information System. In TechEncyclopedia, viewed on 1
February 2008. [WWW document] http://www.techweb.com/ency-
clopedia/defineterm.jhtml?term ¼information+system.
TRIST E (1981) The evolution of socio-technical systems: a conceptual
framework and an action research program. In Perspectives on
Organizational Design and Behavior (VAN dEVEN and JOYCE W, Eds),
Wiley Interscience, NY.
UNITED KINGDOM ACADEMY FOR INFORMATION SYSTEMS (UKAIS) (1997)
Newsletter of the UKAIS. 3, cited by Monarch IA ‘‘Information Science
and Information Systems: Converging or Diverging?’’ Viewed on 1
September 2008 at [WWW document] http://www.cais-acsi.ca/
proceedings/2000/monarch_2000.pdf.
WAND YandWEBER R (1990) Toward a theory of the deep structure of
information systems. In Proceedings of the Eleventh International
Conference on Information Systems (DEGROSS J, ALAVI MandOPPELLAND
H, Eds), pp 61–71, Copenhagen, Denmark.
WAND YandWEBER R (2002) Research commentary: information systems
and conceptual modeling – a research Agenda. Information Systems
Research 13(4), 363–376.
WATSON RT Ed (2008) Information Systems, Release 6, Global Text Project,
viewed on 1 September 2008 at [WWW document] http://home-
page.mac.com/rickwatson/filechute/IS%20bookE1R6.pdf.
Appendix
Background related to the work system approach
This Appendix covers topics that have been presented in
various evolving versions over the last decade:
summary of the work system framework;
clarifications about each of the elements of a work
system;
summary of the work system life cycle (WSLC) model;
summary of the work system method (WSM).
Work system framework The work system framework
(Figure 2) was developed to help business professionals
recognize and understand IT-reliant systems in organiza-
tions. This framework emphasizes business rather than
IT concerns. It identifies nine elements that are part of
even a rudimentary understanding of a work system
(Alter, 2003a, 2004, 2006, 2008). Six of those elements are
part of the definition of IS. The three additional elements
are included because they are an important part of any
analysis of a work system, and hence an IS.
The work system framework provides an outline for
describing the system being studied, identifying pro-
blems and opportunities, describing possible changes,
and tracing how those changes might affect other parts of
the work system. The arrows within the framework
indicate that the various elements of a work system
should be in balance. The first four elements are the basic
components that actually perform the work. These
include processes and activities, participants, informa-
tion, and technologies.
Defining information systems as work systems Steven Alter 465
European Journal of Information Systems
The work system framework assumes that customers
and products and services are not part of a work system; it
includes those two elements because systems exist in
organizations in order to produce products and services
for internal or external customers. Environment and
infrastructure are included in the work system framework
because a work system’s success often depends on its fit
with the surrounding environment and on its use of
available infrastructure that is shared with other work
systems. Strategies are included in the framework as a
reminder that work systems have implicit or explicit
strategies and that those strategies should be aligned with
the organization’s strategies.
The framework makes no assumptions about whether
or not IT is used. It simply reserves a location for
whatever technology is used. This is appropriate because
any particular work system might not use IT, or might use
IT only in a minor way. Furthermore, the framework does
not create an artificial separation between the work
system that produces products and services for the
customer and the IS that often overlaps with the work
system. Regardless of whether the topic is an IS or an IT-
reliant work system that is supported by an IS, it is the
analyst’s responsibility to decide the content and bound-
aries of the system of interest.
The framework says nothing about how long a system
will exist. Some work systems exist and produce their
outputs over extended time spans. Others are projects,
temporary systems designed to produce a particular output
and then dissolve. In other words, the same framework
can be used to summarize work systems in general, ISs,
and projects. The concept of work system therefore pro-
vides a consistent starting point for thinking about work
systems in general, about ISs that support work systems,
and about projects that build or modify work systems.
The work system framework is easy to understand as an
abstraction and has many uses in practice. At the
beginning of an analysis, it can help people agree about
the scope of the system, and hence, what the analysis
should include or exclude. Later, it is a useful reference
point for keeping the analysis on target and recognizing
whether the initial definition of the system and problem
proves inadequate in relation to the realities that are
uncovered.
The individual ideas underlying the work system
framework are not revolutionary, but the framework
combines these ideas to provide an organizing perspec-
tive that is quite different from the way people in many
organizations talk and think about computerized sys-
tems. Unless a work system approach or some other
consciously sociotechnical approach is taken, informal
conversations, formal presentations, and written propo-
sals and documentation about systems often tend to
emphasize computerized features and functions. At the
same time they tend to de-emphasize how people
produce the work system’s products with the help of
technology, how people deal with exceptions and work-
arounds, how well the various components of the work
system are expected to operate, and how infrastructure
and surrounding environment affect the work system.
More about work system elements A number of distinc-
tions in the work system framework are useful in
understanding work systems. And since ISs are work
systems, the same distinctions stated in terms of work
systems in general also apply to ISs.
Customers include the direct beneficiaries of whatever a
work system produces, plus other customers whose
interest and involvement is less direct. This definition
includes both internal and external customers, and
therefore is less like a marketing definition and more
like an operations management or Six Sigma definition.
Customers may also be participants, both in self-service
situations and in other work systems in which they play a
significant co-production role.
As introduced in discussions beyond the scope of this
paper, an additional framework called the service value
chain framework (Alter, 2007a, 2008) is based on the
assumption that services tend to be co-produced by
providers and customers, implying that customers should
be viewed as part of a work system. Since most work
systems can be viewed as service systems (Alter, 2007a,
2008), at least some of the customers should also be
considered participants, even in work systems that are
not viewed as self-service systems. (Self-service systems
are work systems in which a service provider provides
resources that are used by customers to create value for
themselves and possibly for the provider.)
Products and services produced by a work system are the
combination of physical things, information, and ser-
vices that the work system produces for its various
customers. A work system’s products and services may
take various forms, including physical products, informa-
tion products, services, intangibles such as enjoyment
and peace of mind, and social products such as arrange-
ments, agreements, and organizations.
Processes and activities cover a full range of situations
that might involve highly structured workflows and/or
‘artful processes’ whose sequence and content ‘depend on
the skills, experience, and judgment of the primary
actors’ (Hill et al., 2006). Perspectives for thinking about
processes and activities in depth include workflow,
decision making, communication, coordination, control,
and information processing, among others. Each of these
perspectives brings a distinct set of concepts and general-
izations.
Participants are people who perform the non-auto-
mated work in the work system.
The term participants (not users) is included because
non-users of IT may play important roles in work
systems, and because the usage of technology is of
secondary importance to important participants in many
work systems.
Information includes codified and non-codified infor-
mation used and created as participants perform their
work. Typical codified information is the pre-defined
Defining information systems as work systems Steven Alter466
European Journal of Information Systems
information used in tracking packages, entering orders,
and performing repetitive financial transactions. In each
case, each data item must be defined precisely, and the
information is often processed using explicit rules.
Typical uncodified information includes computerized
or handwritten documents, verbal agreements, and
formal or informal conversations. Information not
related to the work system is not directly relevant,
making the common distinction between data and
information secondary when describing or analyzing a
work system.
Technologies (not IT) is included because multiple
technologies may be relevant to the analysis. Technolo-
gies may be general purpose or tailored to a specific
situation. Technologies tailored to specific business
situations usually involve a combination of general-
purpose tools and specialized techniques. The separation
between tools and techniques is worth considering
because it is often possible to use a different general-
purpose tool without changing the technique and vice
versa.
Infrastructure resources that the work system relies on
even though these resources are managed outside of it
and are shared with other work systems.
Environment includes the organizational, cultural, com-
petitive, technical, and regulatory environment within
which the work system operates. Factors in the environ-
ment affect system performance even though the system
may not rely on those factors directly in order to operate.
The organization’s general norms of behavior are part of
the culture in the environment that surrounds the work
system, whereas behavioral norms and expectations
about specific activities within the work system are
considered part of its processes and activities.
Strategies consist of the guiding rationale and high-level
choices within which a work system, organization, or
firm is designed and operates. A work system’s strategy
includes its production strategy and its value proposi-
tions for customers. Other strategies of relevance for
understanding a work system include the business
strategies of the organization or firm and technical
strategies such as enterprise architectures (if used).
Work System Life Cycle (WSLC) model In addition to
leading to a framework for describing and analyzing an IS
as it exists at a particular point in time, the proposed
definition also leads to the WSLC model (Alter, 2006).
That model describes how persistent work systems (work
systems that are not projects) change over time through a
combination of planned change (explicit projects with
initiation, development, and implementation phases)
and unplanned change (adaptations and experimenta-
tion). As shown in Figure 3, the WSLC is fundamentally
different from the frequently cited system development
life cycle (SDLC). First, the SDLC is basically a project
model rather than a system life cycle. Even current
versions of the SDLC that contain iterations are basically
project models. Second, the term system in the acronym
SDLC is basically a technical artifact that is being
programmed. In contrast, the system in the WSLC is a
work system that evolves over time through multiple
iterations. Unlike control-oriented versions of the SDLC,
the WSLC treats unplanned changes as part of a work
system’s natural evolution.
Unanticipated adaptations Unanticipated opportunities
OPERATION and
MAINTENANCE Redesign
Continue
Terminate
I NITIATION
Accepted for
operation
Recognition of
non-adoption
or excessive
workarounds
Ready for
development
Recognition of
infeasibility in
vision, goals, or
resources
IMPLEMENTATION Recognition of infeasibility in
vision, goals, or resources
Ready for implementation
DEVELOPMENT
Unanticipated adaptations Unanticipated opportunities
Figure 3 The work system life cycle model (Alter, 2006).
Defining information systems as work systems Steven Alter 467
European Journal of Information Systems
The WSLC presents a picture of punctuated change
(e.g., Lyytinen & Newman, 2006) whereby work systems
operate in a fairly stable configuration for extended
periods of time, during which work system participants
may make minor tweaks and adjustments without
changing the fundamental structure or operation of the
work system. Eventually management decides to initiate
a project that will create a major change in the work
system. Punctuated change occurs as the new work
system is implemented.
Work system method The definition of work system (and
hence IS) and the two frameworks mentioned above are
the basis of the WSM, an organized way to analyze any
work system (and hence any IS) from a business view-
point. The design of WSM assumes that business profes-
sionals frequently lack two important things: first,
organized frameworks, terminology, and methods for
understanding systems in organizations; second, proce-
dural knowledge about the process of analyzing systems
and making justified recommendations for improving
those systems. The structure and content of WSM
attempt to provide both conceptual and procedural
knowledge in a readily usable form, and try to express
that knowledge in everyday business language. The WSM
is organized around the work system framework. Just
agreeing on the identity and scope of its nine elements in
a particular situation can eliminate fundamental confu-
sions in many situations (Alter, 2006).
WSM is based on the assumption that a single, totally
structured analysis method is not appropriate for all
situations because the specifics of a situation determine
the nature of the understanding and analysis that is
required. In some situations, a manager simply wants to
ask questions to make sure someone else has done a
thoughtful analysis. At other times, a manager may want
to establish a personal understanding of a situation
before discussing it with someone else. When collaborat-
ing with IT professionals, managers can use WSM to
clarify and communicate their own understanding of the
work system and to make sure that the IT professionals
are fully aware of business issues and goals that software
improvements should address. Managers and business
professionals wishing to perform a thorough analysis can
follow a set of specific questions and can use a set of
templates for pursuing the analysis in depth. From the
other side, IT professionals can use WSM at various levels
of detail to confirm that they understand the business
professionals who are the customers for their work.
WSM focuses on IT-reliant work systems rather than on
IT applications that support them. It is designed to
produce shared understandings rather than detailed
specifications that are required to produce high-quality
software. It can be used for existing systems or for totally
new systems. Starting with the work system framework
and related characteristics and performance indicators,
WSM uses business concepts that are readily under-
standable to business professionals. These concepts are
well enough defined to provide a reasonable degree of
rigor in the description of the system and in the analysis.
WSM’s basic form assumes that a set of problems or
opportunities motivate the analysis of an existing work
system.
The most basic application of WSM encourages the user
to think about the situation in work system terms, but
provides minimal guidance other than saying that each
of three main steps should be included. For example,
assume that several people are speaking in general about
whether a particular CRM software package might be
beneficial. Even if they use WSM in the most basic way
they would not focus initially on the features and
purported benefits of the CRM software. Instead, they
would discuss:
System and problem (the SP step): What work system
are we talking about? From a business viewpoint, what
are the problems and opportunities related to the work
system?
Analysis and possibilities (the AP step): What are the
shortcomings of each part of the work system and what
are the possibilities for eliminating or minimizing
these shortcomings? Adopting CRM software might be
one possible change, but there are probably others.
Recommendations and justification (the RJ step): What
changes in the work system do we recommend and
how could we justify those changes?
Recognizing the varied goals of WSM applications, WSM
can be used at three levels of detail and depth. The user’s
goals and the need to communicate and negotiate with
others determine the level to use in any particular situation.
Level one: Be sure to remember the three main steps
when thinking about a system in an organization: SP
(system and problem), AP (analysis and possibilities),
and RJ (recommendation and justification).
Level two: Within each main step, answer a set of
specific questions related to each work system element
and other factors that are typically important.
Level three: Use checklists, templates, and diagrams to
identify and consider topics and issues that might
otherwise be overlooked.
In the discussion of CRM, merely using these questions to
stay focused on the work system instead of plunging into
software details and features would probably make the
initial discussion more productive. The discussion would
treat the CRM software as part of a larger system. It would
be clearer that addressing the business problems or
opportunities requires many changes beyond just adopt-
ing CRM software.
Executive MBA, MBA, and undergraduate students
have used various abbreviated versions with varying
degrees of success. The most recent version attained the
best balance thus far in terms of excessive simplicity, on
the one hand, and excessive completeness and cognitive
overload, on the other. Classroom experience plus
informal testimonials from employed students (e.g., ‘I
Defining information systems as work systems Steven Alter468
European Journal of Information Systems
am using it to help in my software sales cycles.’) support
the possibility that WSM might help business and IT
professionals improve current, often unsatisfactory levels
of user involvement and business/IT alignment. Field
experiments and action research studies are required to
learn more about WSM’s strengths and limitations and to
make it more powerful.
The development of WSM is continuing. New tools
based on a new service value chain framework (Alter,
2007a, 2008) will be integrated into WSM to attain
greater descriptive and analytical depth and to improve
business/IT communication without overloading busi-
ness professionals with an excessive number of technical
distinctions and details.
Defining information systems as work systems Steven Alter 469
European Journal of Information Systems
