Content uploaded by Harri Jalonen
Author content
All content in this area was uploaded by Harri Jalonen on Nov 25, 2014
Content may be subject to copyright.
Predictive business –
fresh initiative or old wine in
a new bottle
Harri Jalonen
Turku University of Applied Sciences, Loimaa, Finland, and
Antti Lo
¨
nnqvist
Department of Business Information Management and Logistics,
Tampere University of Technology, Tampere, Finland
Abstract
Purpose – The purpose of this paper is to present a conceptual analysis of the theoretical and
managerial bases and objectives of predictive business. Predictive business refers to operational
decision-making and the development of business processes on the basis of business event analysis. It
supports the early recognition of business opportunities and threats, better customer intimacy and
agile reaction to changes in business environment. An underlying rationale for predictive business is
the attainment of competitive advantage through better management of information and knowledge.
Design/methodology/approach – The approach to this article is conceptual and theoretical. The
literature-based discussion and analysis combines the perspectives of business performance
management, business intelligence, and knowledge management to provide a new model of thinking
and operation.
Findings – For a company predictive business is simultaneously a practical challenge and an
epistemic one. It is a practical challenge because predictive business presupposes a change in the
company’s modes of operation. It is also an epistemic challenge, since it concerns the company’s ability
to find appropriate balance between knowledge exploitation and knowledge exploration.
Research limitations/implications – Further research should be carried out on the functionality
of practical applications as well as the attitudinal and technical preparedness of companies to adopt a
new mode of operation. As a subject of investigation, the world of business events offer interesting
methodological possibilities, since the basis of the work is the gathering and analysis of large
quantities of information on operational activities.
Originality/value – There has been little research concerning business events in knowledge
management context. This article presents a theoretically founded basis for predictive business,
combining the concept of analysing business events with previous research in the field of knowledge
management.
Keywords Performance management, Business performance, Knowledge management,
Decision making
Paper type Conceptual paper
1. Introduction
A salesperson enters an order into the company’s ERP system. However, due to a
router malfunction, the order never reaches delivery stage. When the agreed delivery
deadline has expired, the customer contacts the company’s customer service in order to
find out what is going on with the order. The customer service employee, however,
faces a difficult task, as from his or her point of view the whole order never even
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/0025-1747.htm
Predictive
business
1595
Received March 2009
Revised August 2009
Accepted August 2009
Management Decision
Vol. 47 No. 10, 2009
pp. 1595-1609
q Emerald Group Publishing Limited
0025-1747
DOI 10.1108/00251740911004709
existed, because of the router malfunction. After several failed attempts to trace the
order, the frustrated customer gives up and orders the product from a rival company
(Lundberg, 2006). Meanwhile, elsewhere: in a company producing dairy products, the
lift of a fully automated logistics centre stops for four hours. For products requiring
cold storage, the company has no choice but to forward the products to a waste
disposal site for destruction. The direct costs of the operation reach tens of thousands
of euros. In addition to the spoiled dairy products, the company incurs costs from a
breach of the service level agreement (SLA). The company’s reputation as a reliable
supplier also suffers a blow, and its restoration generates costs with long-term effects.
Introducing the phenomenon
The loss of a customer to a competitor, the destruction of spoiled products, a breach of
agreement or damage to the corporate image may generate many kinds of direct and
indirect costs for companies. However, the lesson in these examples lies not in this
relatively simple observation, but in the fact that in both examples, an identifiable
event triggers off a specific chain of acts: the faulty router leads to the termination of a
customer relationship and the broken lift results in damage to the company’s
reputation. This article concentrates on observing correlations between this kind of
events. Our hypothesis is that the emergent whole which is formed by individual
events represents either a threat or an opportunity for the company’s business. An
emergent whole, in this context, means the higher-level behaviour generated by the
combined effect of the events. Instead of individual events, we might speak of a kind of
cloud of events, in which the whole may be qualitatively more or less than the sum of
its parts. Due to the emergence, even apparently quite unrelated events, small in
themselves, may in the right conditions combine to form a chain of events with
sometimes quite unexpected effects. Events often acquire real business significance
only in combination with other events. The business significance of the malfunctioning
router is concretised when the window salesperson loses his or her bonuses and the
company loses a customer – possibly for good. In the same way, the costs related to the
destruction of spoiled dairy products may be just a fraction of the costs necessary to
restore the company’s damaged image. To put it symbolically: the individual raindrops
form a puddle, from which water then starts to flow in one direction or another. The
essential concern here is not the puddle but the source of the rain. In the examples, this
means the factors behind the series of events that caused the termination of the
customer relationship and the damage to the corporate image. The fundamental
question here is: how can the business costs generated by the breakup of a machine
(router, lift, etc) be limited to a tolerable level, or – perhaps even more importantly –
how can quick responses to customer problems be turned into a competitive advantage
and a means of strengthening customer loyalty and the corporate brand value?
For companies to be able to change their modes of operation towards predictive
business, new methods and analytical tools are needed. But what does predictive
business mean and how does it relate to the existing management tools and
viewpoints? The field is teeming with varied terminology, such as business activity
monitoring (Peipert, 2005), business event processing (Zeng et al., 2007), business
process intelligence (Grigori et al., 2003), complex event processing (Luckham, 2007),
predictive analytics (Hair, 2007), competing analytics (Davenport and Harris, 2007),
MD
47,10
1596
real-time business intelligence (Sahay and Ranjan, 2008) and real-time knowledge
management (El Sawy and Majchrzak, 2004).
What the abovementioned terms have in common is that they all refer to a specific
mode of operation and its supporting knowledge system, used to produce analytical
information based on event data from business processes, at very short delays, mainly
to support operative decision making. Some of the terms refer to management models
and some, in a more concrete way, to the technologies with which the information on
business events is analysed “in real time”. However, it seems that the development
being done in this field is company – and application-oriented, and that researchers
have failed to analyse the conceptual framework.
Aims and objectives
The aim of this article is to present a conceptual analysis of the theoretical and
managerial bases and objectives of predictive business. The article contains no
empirical material. The literature-based discussion and analysis combines the
perspectives of business performance management, business intelligence, and
knowledge management to provide a new model of thinking and operation. In the
article, predictive business means operational decision-making and development of
business processes on the basis of event analysis. The prediction of product demand
and market shares, the evaluation of macro-economic development, scenario work to
probe the changes in the company’s business environment and other similar activities
aiming to analyse the future are left outside the scope of the article.
2. The concept of business event
In spoken language, the concept of event generally refers to either ongoing or finished
activity. An event consists of a series of actions, some of which are intentional and
some unintentional. Ordering a product from an online store may be considered an
example of an intentional and planned event, whereas the collision of a bus with a train
is usually an unintentional and unplanned series of actions. In this article, a business
event refers to an empirical whole, which consists of a series of intentional or
unintentional actions, verifiable through experiential knowledge or registered data,
and of circumstances, and has material or immaterial consequences (Figure 1).
Whether the material and immaterial consequences represent an opportunity or a
threat to the company depends on the business context[1]. In other words, the event is a
Figure 1.
The material and
immaterial consequences
of an event
Predictive
business
1597
sort of driving power presented by the situation, whose exploitation depends on the
company’s internal and external factors. Instead of the individual event, the business
significance of events often springs from their combination with other events. In
accordance with Baraba
´
si (2002, p. 14), the article estimates that “events are connected
to an enormous number of other pieces of a complex universal puzzle, are caused by
them and are in interaction with them”. In assessing the business significance of
events, it is thus important to learn to understand the correlations between events.
Some of these are clear, and companies usually have no difficulty understanding them.
It is easy to see a delay in delivery, for instance, as an event, which forms a threat to the
company unless it is responded to. But some correlations between events are less
evident. These typically occur in situations which in superficial analysis seem to hold
no surprises, but which at closer inspection prove to be something else altogether. Even
if a company’s delivery reliability on the whole is at a good level, a closer analysis may
point out a significant divergence, one of the company’s functional units constantly
having lower delivery reliability than the others. An analysis may reveal such causes
as an inadequate job briefing, for instance, due to which the unit’s personnel has been
unable to meet their tasks. From the viewpoint of this article, it is in fact important to
note that even something that fails to happen (e.g. a job briefing according to quality
instructions) is an event with business significance.
The article estimates that even complicated chains of events can be understood once
they are broken down into their parts. A careful analysis of each part helps to gain a
motivated picture of the factors and mechanisms influencing the occurrence of an
event. The identification of these factors and mechanisms, in turn, provides a basis for
the prediction of future events.
3. The elements of predictive business
Predictive business refers to an activity that aims, among other things, to eliminate
guesswork and surprises from business processes, to identify opportunities and
threats at an early stage and to respond quickly to changes in such factors as demand
and offer or changes in the financial and commodities markets (e.g. Davenport and
Harris, 2007; Ranadive
´
, 2006). The ultimate goal of predictive business may be
considered to be the attainment of operative agility through better management of
information and knowledge. In this article, information refers to data, which has a form
and a context of use, which carries some kind of meaningful content, and is intended to
be communicated. Information becomes knowledge when the receiver interprets it in a
specific context. From the viewpoint of information and knowledge management,
predictive business may be divided into three phases as shown in the Figure 2:
observation, analysis and optimisation.
Observation here means the pre-planned, active and systematic gathering of
information from the company’s internal sources (e.g. ERP system) and external sources
(e.g. news services). By visualising the information gathered from different sources,
combined with, e.g. key performance indicators (KPI), the company can get a better idea
of the state of its operational processes. In practice this could mean, for instance,
providing the process owner with a real-time visual view of the business processes in
order to chart the process bottlenecks as comprehensively as possible. If, for instance, a
product’s order-to-delivery cycle time differs from the predefined limit values, the
process owner receives an alarm which in turn allows him or her to respond quickly and
MD
47,10
1598
start corrective actions as well as avoiding customer disappointment. Visualisation
gives the process owner an idea of the chain of events that has started or is starting and
of the information needed to solve the problem and the place to seek it (Arita et al., 2007,
p. 43). Observation and visualisation are based on ex nunc information, which refers to
information generated along the process through its internal logic (see Lillrank et al.,
2004). In essence, it is a question of combining the information and the situation into a
whole and visualising it in a way that activates the cognitive processes of the mind
(perception, memory, problem solving, comprehension). In semantic terms, the real-time
graphics demonstrating the state of the order-to-delivery process functions as a sign
depicting reality, which “interpellates the mind of the user” thus initiating a specific
interpretation process[2]. Visualisation has been found to improve such processes as
intuitive thinking and the observation of unexpected elements which would otherwise
remain undetected (Tyman and Huang, 2003).
Observation provides the basis for the next level: analysing the events. Analysis aims
at understanding the correlations between events. In terms of knowledge management,
analysis means combining the ex nunc information generated by the current chain of
events with ex post information generated by earlier chains of events, stored in different
registers. The hypothesis is that the different events share at least some common
characteristics, thus allowing the exploitation of information registered from one event
for analysing another event. In practice, this implies patterns and rules which help
identify those events that are relevant to the company’s business within the general flow
of events (Luckham, 2007, pp. 55-59). The aim of event analysis is to identify the reasons
behind the higher-level behaviour. The essential concern here is the capacity to examine
the information generated by the processes in all its multiple aspects. For instance, the
factor behind the divergence in the order-to-delivery cycle time can be found by breaking
the process time into sections (e.g. order time, adjustment time, production time and
delivery time) and examining each of them separately. Likewise, a company with similar
production operations in several different locations can compare the functionality of
production processes in terms of time, quality or cost and thus identify the locations
where processes should be improved.
Event analysis provides a basis for preparing for future events and for optimising
operations. Optimisation refers to the attainment of the best possible end result within
Figure 2.
The relationship of
knowledge exploitation
and operative agility to
elements of predictive
business
Predictive
business
1599
specific limitations. Optimisation being a future-oriented activity by nature, there are
grounds for suggesting that it presupposes a comprehensive knowledge basis. This
article estimates that in addition to information and knowledge, optimisation requires a
pinch of intuition. It should undoubtedly be emphasised that intuition here does not
mean the opposite of analytical knowledge, but a complementary form of knowledge.
Intuition is understood as a sort of leap by which the decision-maker distances him- or
herself from the matter at hand and solves it in a new, insightful and sustainable way
(Kierkegaard, 1998). Etymologically, intuition means “looking” and “seeing”. In the
context of predictive business, the analytical activity of looking into the future and
seeing it can be supported by providing the decision-maker with tools to support,
e.g. modelling, simulation and “what-if” analyses (Lundberg, 2006). It is important to
note that intuition is not a question of traditional ad-hoc style problem-solving, but a
skill that can be learned and developed. From the viewpoint of knowledge
management, intuition represents ex post knowledge, in which ex post information is
combined with the existing knowledge basis of the operators.
4. Predictive business from the viewpoi nt of company architecture
It can be argued that the ultimate objective of predictive business is to respond both
efficiently and creatively to a company’s internal and external events. This article
estimates that in order to combine the efficiency gained by exploiting resources with the
creative search for new opportunities, a company must not only be sufficiently
functional internally but also open to stimuli from the environment. A company that
applies predictive business should have both internal closure and an interactive
openness to its environment (see Maula, 2006, pp. 93-98). Closure here means the
company’s “memory”, which may be understood as the company’s organised knowledge
and the process of recording its past (Maier, 2004, pp. 79-81). Correspondingly,
interaction necessitates “senses” through which the company participates in the creation
of new knowledge and coordinates its activities to fit the changing business
environment (Maula, 2006, p. 93). In practice, the company’s memory and its senses are
concretised in the company architecture, which refers to its business and information
technology as a whole, as well as the main aspects in the development of this whole
(Dragstra, 2005). A functioning company architecture will allow an efficient and creative
exploitation of both the knowledge and experience accumulated within the company and
its processes, and of knowledge external to the company.
Even if the elements of company architecture are basically specific to each
organisation, some general criteria for a good company architecture can also be
identified in the literature. In particular, if company architecture is understood as a
collection of capabilities, one approach worth considering is the division suggested by
El Sawy and Pavlou (2008), which also suits the needs of this article. El Sawy and
Pavlou divide a company’s capabilities into three mutually complementary categories.
These are operational capabilities, dynamic capabilities and improvisational
capabilities. Operational capabilities are manifested in the efficient performance of
planned daily operations (e.g. production, logistics, sales). An efficient performance of
daily routines may be considered a necessary condition of successful business,
although not sufficient in itself. In a swiftly changing business environment, a
company also needs dynamic capabilities, by which El Sawy and Pavlou mean the
company’s ability to adapt its operations to the changing demands of the environment.
In the same vein as Teece et al. (1997), El Sawy and Pavlou divide dynamic capabilities
into four components: observation of the environment, learning, knowledge
MD
47,10
1600
integration, and coordination. The observation of the environment consists in
identifying the emerging internal and external opportunities and assessing their
business effects. Learning involves the improvement of operational capabilities based
on the creation, acquisition and absorption of new knowledge. Knowledge integration
refers to commonly shared understanding and collective sense-making, which is
indispensable in order to incorporate new knowledge into the company’s daily
operations. Coordination means organising the insights and new modes of operation
generated in the previous phases and concretising them into tasks and resources.
According to El Sawy and Pavlou (2008), in addition to operational and dynamic
capabilities, companies also need improvisation – the capability to act spontaneously
when a sudden need (a threat or an opportunity) arises. This is because – particularly
in a dynamic environment, with opportunity windows opening and closing ever faster
– there is increasingly little time left for formal planning. While dynamic capabilities
are based on planning, improvisation is by nature a capability based on learning.
According to Weick (1998), improvisation is a behavioural pattern that can be learned
and then utilised in different operational situations. Dynamic capabilities refer to a
disciplined flexibility, based on a logic of opportunities, while improvisation, drawing
from a logic of responsiveness, presupposes creativity and intuition (El Sawy and
Pavlou, 2008, p. 141).
The contents of a company architecture and its combination of capabilities depend
on each particular context. This article considers context to be determined by the
nature of the events. In company architecture, the objective should be an appropriate
balance between operational capabilities, dynamic capabilities, and improvisation.
This appropriate balance may be defined as a point on the line segment connecting
efficiency and creativity. An appropriate balance also includes change and movement,
since the location of the point depends on the nature of the event. In a more complicated
and complex event environment, the balance may be considered to be closer to the side
of improvisation and dynamic capabilities, which nourish intuition and creativity.
Correspondingly, when the event environment is simple, the balance should be closer
to the operational capabilities, which support efficiency.
The relationship between event nature and action may be demonstrated in Figure 3,
in which the “simplicity” of an event depends on several factors: context, the necessary
capabilities, the basis for understanding the event, the focus of action and the
predictability of the end result.
Figure 3.
The relationship between
environment, event nature,
capabilities and action
Predictive
business
1601
The context of a simple event may be characterised as a linear and organised
environment. Such an environment typically contains causal relationships, which are
relatively easy to identify. A simple event can be defined in its entirety, which in turn
means that all the information necessary for understanding the event and reaching the
desired end result can be listed exhaustively. From the viewpoint of architecture, the
question is one of exploitation of operational capabilities and the functionality of daily
routines. Having sufficient information on the factors influencing the event also means
that the focus of action can be put on the categorisation of events. Categorisation refers
to the act of placing events into predefined patterns and into directives connected to
these patterns (Snowden and Boone, 2007, pp. 2-7). An event identified in the
order-to-delivery process, fulfilling the criteria of a predefined pattern, may be
understood as a kind of trigger that activates the actions linked to the pattern. The
more predictable the context, the easier it is to automate a part of the actions triggered
by the events (Indart, 2005).
In chaos and disorder, events are complicated by nature. Even though chaos is often
understood as a negative phenomenon and associated with difficulty in managing
events, it may also be understood as a natural element in the functioning of any system.
In this article, chaos refers to the complicated behaviour of a complicated system, where
the end result of the development cannot be completely predicted on the basis of what is
known of its initial state. The main reason for the unpredictability of the end result lies in
the correlations between events. However, this does not imply that one could not or
should not attempt to analyse chaotic events. Aula (1999), for instance, emphasises that
regularity can be found even in chaos. The essential aspect of chaos is the lack of
regularity, as it is found in classical science: as recurrent trajectories and end results that
can be predicted with adequate accuracy on the basis of the initial conditions (Aula,
1999). According to Mitleton-Kelly (1997), a chaotic system is formed by the iteration of
rules and regularities that remain constant. The crucial challenge here is to learn to
identify the underlying regularities that influence chaotic events and the correlations
between the events. This article estimates that in order to meet this challenge, a company
should at least be able to combine the information related to ongoing events with the
knowledge and know-how already absorbed in the company’s processes and operators.
This is because regularities are supposed to exist between different events, which means
that registered information on one event can be utilised in understanding other events.
Even though the example presented in the article introduction – a lift breakdown in a
logistics centre – may be a unique event, it is quite likely that a company has experience
of events similar to a lift breakdown in terms of their consequences. An in-depth
understanding of the immaterial and material consequences of earlier breaches of
contract, for instance, may act as the necessary framework supporting the observation
of, and learning from, even apparently unique events, as well as the integration and
coordination of knowledge. Analysis as a form of activity represents the uncovering of
the “known unknowns” (Snowden and Boone, 2007, p. 3).
Unlike events in a simple or complicated environment, events in a complex
environment may have several possible end results depending on the approach and the
explanatory model. Complexity breeds uncertainty, on the one hand because events do
not follow the course of previous events or previous plans, and on the other hand
because human knowledge is always incomplete. Complexity may, however, also be
considered both a threat and an opportunity at the same time. The difference between a
MD
47,10
1602
threat and an opportunity lies in the approach to uncertainty. Even though complex
events cannot be managed in the traditional sense, facing and understanding them can
be made easier by increasing openness and fostering interaction between individuals.
Especially in business based on know-how and knowledge, this means shifting the
focus in management from control to coordination and to the removal of obstacles to
cooperation. It is important for companies to develop operational cultures that leave
room even for diversity and conflicting opinions. Instead of models and rules that
simplify reality, the best response to environmental complexity is to increase the
interaction within the company as well as between the company and its environment.
The ability to transform complexity into business depends crucially on
improvisational capabilities – i.e. the use of creativity and intuition by companies in
learning to read the emergent events in their environment. In a complex environment,
even more important than event analysis is sense-making, which in this article refers to
the placing of individual stimuli into a larger framework (Weick, 1995, p. 4).
Sense-making is understood as a process seeking not to reduce the uncertainty
springing from lack of information, but rather to reduce the equivocality due to the
ambiguity of knowledge (Daft and Lengel, 1986, pp. 554-555). While uncertainty can be
reduced by adding more information, equivocality reduction presupposes a common
framework of interpretation. Only by processing information and knowledge on the
basis of interaction can a company create new order from chaos.
5. Predictive business – bubble or opportunity?
Implications for management
Business analysis and ERP have previously been approached from such angles as
business performance measurement and management, business intelligence, and
knowledge management. Different information systems and application tools, such as
ERP systems, reporting applications and data mining tools, have also been discussed
in this context. It is unclear to what extent the different approaches – especially the
terms mentioned in the introduction – are interrelated and what their specific added
value is when compared to more traditional approaches. The conceptual apparatus
should be analysed so that researchers could discuss this new phenomenon in a
meaningful way and so that corporate management, too, could understand the role of
the different methods.
In order to have a more precise idea of predictive business, it is useful to compare its
characteristics with more traditional approaches to business management. The
comparison has its problems, though, since traditional approaches too are ambiguous.
Business intelligence, for instance, may be understood in several different ways
(Pirttima
¨
ki, 2007). It is sometimes seen as a method of analysing the business
environment (e.g. competitor analyses, market studies), which mainly produces
qualitative information. At other times, it is taken to mean an IT application with
which the information contained in the company’s knowledge systems is refined into
visual reports for the management. In spite of the related challenges, a comparison
between the typical characteristics of different methods makes it possible to
demonstrate differences between approaches and thus also to discuss whether there is
a need for a new type of predictive business. Below, the characteristics of business
intelligence and performance management are described and then mirrored against
predictive business.
Predictive
business
1603
Business intelligence generates analyses and reports on trends in the business
environment and on internal organisational matters. Analyses may be produced
systematically and regularly or they may be ad-hoc ones, related to a specific
decision-making context. This knowledge is utilised by decision-makers at different
organisational levels, but also by experts (e.g. exploitation of a news service). The
process results in the generation of both numerical and textual information.
Performance management is based on the continuous measurement of objective
factors identified as being important (see, e.g. Lo
¨
nnqvist, 2004). This is done at both
operational and strategic levels, which means that the objects of measurement vary as
well. As a rule, measurement results are numerical, although qualitative information is
also used, e.g. in customer satisfaction measurements. The results of the indicators are
often reported in some sort of dashboard, sometimes including a possibility for the
in-depth examination of more detailed data. Compared to business intelligence, the
main difference lies in the fact that performance management is limited to a selected set
of measurable items, while business intelligence fathoms a larger field through
variables that have not been delimited beforehand. On the other hand, it sometimes
seems that in practice, business intelligence products are really just sophisticated
indicator reporting systems, and there are thus many similarities as well.
When the approaches described above are compared with predictive business,
many similarities are found: in fact, all of the approaches aim to produce information to
support decision-making. However, the purpose of the information produced by
predictive business is to serve operational-level personnel rather than management. It
emphasises the speed of analysis – the striving for real-time monitoring and
decision-making. In addition, the object of monitoring is a large number of
process-level events. The number of monitored factors is larger than in performance
measurement, and this also helps to point out unexpected events. In predictive
business analysis, the IT solution has a central role, while measurements and business
intelligence can be gathered even with simple technical tools, in principle. Table I
describes some of the essential characteristics in the different approaches.
The novelty value of predictive business would seem to lie in the systematic
examination of business process data on a more detailed level than before, with analyses
that support quick decision-making and development of operational activities. Even
though this article does not evaluate the functionality or usefulness of the applications
on the market, it should be noted that different solutions are already available.
In terms of knowledge management, however, quick responses and high-quality
operational decision-making are often hampered by either lack of information or
alternatively by too ambiguous information. Because of knowledge-related problems, the
reality of decision-making can resemble the situation in Figure 4, which demonstrates the
relationship between decision significance, available information and time.
Since the most important decisions are often made in the initial phase of a process,
operators rarely have enough relevant information at their disposal (Moensted, 2006,
p. 19). In spite of the ignorance in the initial phase, companies and individuals have to
make decisions based on the knowledge available at any given moment (see Pollack,
2003, p. 214). In terms of predictive business, the crucial challenge here is reducing the
uncertainty and equivocality that are an integral part of any future-oriented
decision-making. In a nutshell, it can be argued that by investing in business event
observation and analysis and in the optimisation of its activities, a company may
MD
47,10
1604
simultaneously both increase its response speed (T
1
! T
2
) and improve the
knowledge basis and quality of its operational decisions (Q
1
! Q
2
).
Implications for research
The aim of this article was to analyse and bring to the academic community’s attention
a new model of thinking and operation for predictive business. Until now there has
Approach Focus on attention Aim of activity
Exploiting
organisational
level
Time span
of
examination
Business
intelligence
Phenomena related to the
decision-making
situation
To identify trends in the
business environment and
to produce analysed
information for the
decision-maker
Management
and experts
Monthly –
yearly
Performance
measurement
Factors identified as
important (performance
objectives), for which an
indicator has been
designed
ERP, monitoring objective
attainment
Management
and superiors
at different
organisational
levels
At process
level: daily
– monthly
At
corporate
level:
monthly –
yearly
Predictive
business
Predictable and
unexpected events in
business processes
To quickly gain knowledge
on significant events with
the objective of fixing the
problem or exploiting the
opportunity
Operative
personnel
Continuous
monitoring
Table I.
Characteristics of
different approaches to
management
Figure 4.
The relationship between
the significance of choices,
information quantity and
quality, and time
Predictive
business
1605
been little research available in this field of knowledge management and business
management, and it seems quite a promising subject of investigation. In the future,
more research will be necessary for analysing the concepts and understanding the
phenomenon. Research should also be carried out on the functionality of practical
applications as well as the attitudinal and technical preparedness of companies to
adopt a new mode of operation. As a subject of investigation, the world of business
events may offer interesting methodological possibilities, since the basis of the work is
the gathering and analysis of large quantities of information on operational activities.
This material can probably be utilised in many ways to research the connections
between different factors. It may also be necessary to develop new analytical methods
for analysing these connections.
In addition to possible beneficial consequences, there might also be some
unintentional negative outcomes like alert overload and power conflicts. An increase in
IT applications amplifies the risk of “overuse” of the real-time capability, which may,
in turn, overload decision-makers with too many alerts in much the same way that
e-mails, text messages, etc., have done (Welke et al., 2007). It is also important to
remember that information technology that is technically and economically feasible
can be socially infeasible (Bostrom and Heinen, 1977). This is because the technology
affects bargaining power both inside and outside the organization, disrupts social
processes, and violates social taboos (Markus, 1983). In predictive business, there is, for
example, the issue of too much transparency, which may cause a negative experience
of the “big brother” effect. In other words, the intentions are positive but the
consequences may be negative. This “hidden” side of the predictive business is
certainly worth further research.
Finally, the limitations of this study should be discussed in order to be taken into
account in further research. First, this conceptual paper strongly reflects the authors’
own thinking, although it builds on the work of others. Thus, there could have been
other ways to conceptualise the phenomena discussed in this paper. Second, the
research theme discussed here covers several research streams such as those of
knowledge management and business performance management. Due to the large
volume of potentially valuable literature it is difficult to capture all the relevant
material in one paper. Nevertheless, these shortcomings can be improved in
forthcoming research.
6. Conclusions
For a company, the shift to predictive business is simultaneously a practical challenge
and an epistemic (knowledge-related) one. It is a practical challenge because predictive
business presupposes a change in the company’s modes of operation. In particular, it
concerns the company’s ability to shift from planning-oriented operation (plan-execute)
to response-oriented operation (sense-respond) (Welke et al., 2007). While the
planning-oriented mode of operation relies on the operators’ capacity to remove
uncertainty related to the future and the business environment through rational
planning, the response-oriented approach emphasises the company’s sensitivity to
internal and external events. This sensitivity appears as a resonance and a
responsiveness through which the company adapts its operations to suit internal and
external events. Responsiveness is important for the company in order to learn to
identify and interpret unpredictable events in both internal and external environments.
MD
47,10
1606
Predictive business is also an epistemic challenge, since it concerns the company’s
ability to exploit information and knowledge in different problem-solving and
decision-making situations. This epistemic challenge may be analysed by dividing
knowledge management into knowledge exploitation and knowledge exploration.
Knowledge exploitation is based on the knowledge and know-how incorporated in the
established routines and modes of operation, while knowledge exploration emphasises
the identification of new opportunities and different alternatives (March, 1991, pp. 71-2).
Symbolically, exploration may be seen as a sort of “expedition” which not only visits
new places and creates new knowledge, but also constructs a basis for reforming the
modes of thinking and operation (Holmqvist, 2004). Problems, however, arise from the
fact that in practice, there is always some amount of tension between efficiency and
creativity. This is due to factors such as the different “yield expectations” between
routines and expeditions. Whereas routines create a feeling of security and continuity,
expeditions have an uncertain basis and involve various risks. This despite the fact
that in the long term, passivity in the exploration of new ideas and alternative modes of
operation may paradoxically result in the iteration of risks and give rise to a
development difficult to handle.
This article suggests that even seemingly random events can be understood by
examining the regularities at work under the “surface level” of the events. The article
advances the slightly pointed argument that even apparently chaotic events contain
some sort of underlying order and logic, which can be revealed through a careful
analysis of the events. Instead of complete control, understanding refers to being
prepared to face events. The blind spot of predictive business may not be in the events
themselves so much as in the attitude of individuals and companies towards them (see
Taleb, 2007, p. 21). In the end, it is the details of company architecture and the mental
capacity of individuals that determine to what extent companies are prepared to face
any predictable or unpredictable events.
Notes
1. Business context here refers to the internal (e.g. management, personnel know-how,
corporate culture, reward systems and information and communication technology) and
external (e.g. field of activity, competitors, suppliers, customers and legislation) factors
influencing the company’s business.
2. One of the best-known representants of the semiotic school was Charles S. Peirce (1839-1914).
Peirce’s classical semiotics is based on a sign referring to an object outside itself. The sign is
understood by someone, i.e. the sign interpellates the mind of the user, the interpretant. The
interpretant is a sense-making influence.
References
Arita, Y., Nakayama, N. and Awata, Y. (2007), “Development process visualization and project
management”, Fujitsu Scientific and Technical Journal, Vol. 43 No. 1, pp. 97-104.
Aula, P. (1999), “Organisaation kaaos vai kaaoksen organisaatio?” (“Chaos in an organisation or
organisation in chaos”), academic dissertation, Loki-Kirjat, Helsinki.
Baraba
´
si, A-L. (2002), Linked. The New Science of Networks, Perseus Publishing, New York, NY.
Bostrom, R.P. and Heinen, J.S. (1977), “MIS problems and failures: a socio-technical perspective”,
MIS Quarterly, Vol. 1 No. 3, pp. 17-32.
Predictive
business
1607
Bystro
¨
m, K. (1999), “Task complexity, information types and information sources. Examination
of relationships”, academic dissertation, Acta Universitatis Tamperensis 688, Tampere.
Daft, R.L. and Lengel, R.H. (1986), “Organizational information requirements, media richness and
structural design”, Management Science, Vol. 32 No. 5, pp. 554-71.
Davenport, T.H. and Harris, J.G. (2007), Competing on Analytics: The New Science of Winning,
Harvard Business School Publishing Corporation, Boston, MA.
Dragstra, P. (2005), Enterprise Architecture, The Selection Process of an Enterprise Architecture
Toolset to Support Understanding and Governing the Enterprise, Technische Universiteit
Eindhoven, Eindhoven.
El Sawy, O. and Majchrzak, A. (2004), “Critical issues in research on real-time knowledge
management in enterprises”, Journal of Knowledge Management, Vol. 8 No. 4, pp. 21-37.
El Sawy, O.A. and Pavlou, P.A. (2008), “IT-enabled business capabilities for turbulent
environments”, MIS Executive, Vol. 8 No. 3, pp. 139-50.
Grigori, D., Casati, F., Castellanos, M., Dayal, U., Sayal, M. and Shan, M. (2003), “Business process
intelligence”, Computers in Industry, Vol. 53 No. 3, pp. 321-43.
Hair, J. (2007), “Knowledge creation in marketing: the role of predictive analytics”, European
Business Review, Vol. 19 No. 4, pp. 303-15.
Holmqvist, M. (2004), “Experiential learning processes of exploitation and exploration within and
between organizations: an empirical study of product development”, Organization Science,
Vol. 15 No. 1, pp. 70-81.
Indart, B. (2005), “Navigating the decision-making process through automation”, Business
Intelligence Journal, Vol. 10 No. 3, pp. 29-33.
Jalonen, H. (2007), “Kompleksisuusteoreettinen tulkinta hallinnollisen tehokkuuden ja luovuuden
yhteensovittamisesta kunnallisen pa
¨
a
¨
to
¨
ksenteon valmistelutyo
¨
ssa
¨
”, (“Complexity-based
interpretation of integrating effectiveness and creativity in the preparation of municipal
decision-making”), academic dissertation, Tampere University of Technology, Publication
693, Tampere.
Kierkegaard, S. (1998), Pa
¨
a
¨
tta
¨
va
¨
epa
¨
tieteellinen ja
¨
lkikirj oitus (Afsluttende uvidenskabelig
Efterskrift til de philosophiske Smuler, original work from 1846), WSOY, Juva.
Kupi, E., Iloma
¨
ki, S-K., Talja, H., Lo
¨
nnqvist, A. and Sillanpa
¨
a
¨
, V. (2008), Aineettoman pa
¨
a
¨
oman
riskienhallinta – Riskit ja riskienhallinnan ka
¨
yta
¨
nno
¨
t yrityksissa
¨
(Risk Management of
Intangible Assets – Risks and Risk Management Practices in Companies), VTT Working
Papers 104, ESPOO.
Lillrank, P., Kujala, J. and Parvinen, P. (2004), Keskenkera
¨
inen potilas. Terveydenhuollon
tuotannonohjaus (Unaccomplished Patient. Production Management in Health Care),
Talentum, Helsinki.
Lo
¨
nnqvist, A. (2004), “Measurement of intangible success factors: case studies on the design,
implementation and use of measures”, academic dissertation, Tampere University of
Technology, Publication 475, Tampere.
Luckham, D. (2007), The Power of Events. An Introduction to Complex Event Processing in
Distributed Enterprise Systems, Addison-Wesley, Boston, MA.
Lundberg, A. (2006), “Leverage complex event processing to improve operational performance”,
Business Intelligence Journal, Vol. 11 No. 1, pp. 55-65.
Maier, R. (2004), Knowledge Management Systems. Information and Communication
Technologies for Knowledge Management, Springer-Verlag, Berlin.
March, J.G. (1991), “Exploration and exploitation in organizational learning”, Organization
Science, Vol. 2 No. 1, pp. 71-87.
Markus, M.L. (1983), “Power, politics, and MIS implementation”, Communications of the ACM,
Vol. 26 No. 6, pp. 430-44.
MD
47,10
1608
Maula, M. (2006), Organizations as Learning Systems. “Living Composition” as an Enabling
Infrastructure, Advanced Series in Management, Elsevier, Amsterdam.
Mitleton-Kelly, E. (1997), “Organisations as co-evolving complex adaptive systems”, a paper
presented at the BAM97 British Academy of Management Annual Conference, London,
September 8-10.
Moensted, M. (2006), “High-tech, uncertainty, and innovation. The opportunity for high-tech
entrepreneurship”, in Teoksessa Bernasconi, M., Harris, S. and Moensted, M. (Eds),
High-tech Entrepreneurship – Managing Innovation, Variety and Uncertainty, Routledge,
London, pp. 15-32.
Peipert, G. (2005), “Real-time visibility: instant insight into business events”, Business Intelligence
Journal, Vol. 10 No. 4, pp. 24-7.
Pirttima
¨
ki, V. (2007), “Business intelligence as a managerial tool in large Finnish companies”,
academic dissertation, Tampere University of Technology, Publication 464, Tampere.
Pollack, H.N. (2003), Uncertain Science ... Uncertain World, Cambridge University Press,
Cambridge, MA.
Ranadive
´
, V. (2006), The Power to Predict. How Real-time Businesses Anticipate Customer Needs,
Create Opportunities and Beat the Competition, McGraw-Hill, New York, NY.
Sahay, B.S. and Ranjan, J. (2008), “Real time business intelligence in supply chain analytics”,
Information Management & Computer Security, Vol. 16 No. 1, pp. 28-48.
Snowden, D.J. and Boone, M.E. (2007), “A leader’s framework for decision making”, Harvard
Business Review, November, pp. 1-9.
Taleb, N.N. (2007), The Black Swan – The Impact of the Highly Improbable, Penguin Books,
London.
Teece, D., Pisano, G. and Shuen, A. (1997), “Dynamic capabilities and strategic management”,
Strategic Management Journal, Vol. 18 No. 7, pp. 509-33.
Tyman, J. and Huang, E.M. (2003), “Intuitive visualizations for presence and recency information
for ambient displays”, Conference on Human Factors in ComputingSystems,
Ft Lauderdale, Florida, USA, April 5-10, 2003.
Weick, K.E. (1995), Sensemaking in Organizations, Sage, Thousand Oaks, CA.
Weick, K.E. (1998), “Improvisation as a mindset for organizational analysis”, Organization
Science, Vol. 9 No. 5, pp. 543-55.
Welke, R., Cavalheiro, G. and Dahanayake, A. (2007), “Improving IT-enabled sense and respond
capabilities: an application of business activity monitoring at Southern International
Airlines”, Journal of Cases on Information Technology, Vol. 9 No. 4, pp. 40-57.
Zeng, L., Lei, H., Koyanagi, T., Ohsaki, H. and Chang, H. (2007), “Model analysis for business
event processing”, IBM Systems Journal, Vol. 46 No. 4, pp. 817-31.
About the authors
Harri Jalonen is working as Senior Lecturer at Turku University of Applied Sciences, Finland.
His research interests focus on knowledge management and complexity theories. Harri Jalonen is
the corresponding author and can be contacted at: harri.jalonen@turkuamk.fi
Antti Lo
¨
nnqvist is working as Senior Researcher at the Department of Business Information
Management and Logistics, Tampere University of Technology, Finland. He also holds the
position of Adjunct Professor at Helsinki University of Technology. His research interests focus on
intellectual capital and business performance management, especially in service organisations.
Predictive
business
1609
To purchase reprints of this article please e-mail: reprints@emeraldinsight.com
Or visit our web site for further details: www.emeraldinsight.com/reprints
