Technological advances such as mobile computing, 3D printing, or cloud computing enable the creation of novel opportunities for entrepreneurs to create and capture value. However, previous studies revealed that around 75 percent of all start-ups fail at an early stage. This is also true for innovation projects and other forms of innovation related endeavour in incumbent firms (Blank 2013).
One main reason for this tremendous failure rate is that entrepreneurs are typically confronted with high levels of uncertainty about the viability of their proposed business idea. One prominent perspective is that opportunities for such novel business ideas cannot be just discovered by entrepreneurs in the market. Rather, they are endogenously created by actions of an entrepreneur who seeks to actively exploit it in a multistage and iterative process of interaction between herself and the environment (Alvarez et al. 2013). This is especially relevant in the age of digital innovation where entrepreneurial efforts become even more dynamic and dependent on the external ecosystem such as platform owners (Dellermann et al. 2016), partners and customers (Kolloch and Dellermann 2017), or other distributed stakeholders (Nambisan 2017).
Following this argumentation, entrepreneurial decision-making can be defined as complex decision-making problem under both risk and uncertainty (Knight, 1921). While risk includes quantifiable probabilities, uncertainty describes situations where neither outcomes nor their probability distribution can be assessed a priori (Diebold et al. 2010). Consequently, the entrepreneurial decision-making context is highly complex and contains lots of “black swan events” that seems to be unpredictable (Russell and Norvig 2016; Simon 1991; Funke 1991).
For this thesis, I identified several gaps in previous research, which I aim to address with my dissertation.
Research Gap 1 – Limited Investigation of the Sources of Risk and Uncertainty in the Entrepreneurial Decision-Making Context.
The first gap in previous research is related to the lack of understanding of the sources of risk and uncertainty in entrepreneurial decision-making. Little is known about the role of the ecosystem of users, suppliers, partners, and other stakeholders in making decisions. Most research in this field is rather descriptive or conceptual at all (e.g. Alvarez and Barney 2007, Alvarez et al. 2014). Consequently, the lack of empirical investigation of the sources of both risk and uncertainty in the entrepreneurial decision-making contexts as well as the role of the ecosystem as source of those, is the first research gap that was identified.
Research Gap 2 – Limited Investigation of Scalable Mechanisms for Decisional Guidance in Entrepreneurial Decision-Making.
The second research gap that I identified is related to the mechanisms applied for providing decisional guidance which supports and offers advice to a person regarding what to do (Silver 1991). To support entrepreneurs in making their decisions, feedback from social interaction with domain experts proved to be a valuable strategy in managerial practice. Consequently, the dominant form of decision support that emerges is human mentoring (Hochberg 2016) . However, human generated decisional guidance holds also various limitations that can be subsumed under two dimensions: cognitive limitations (e.g. limited information processing capabilities, expertise, flexibility, or biases) that prevent individual experts from providing optimal guidance, and resource constraints (e.g. time constraints, financial resources, social capital, and demand side knowledge) (Zhang and Cueto 2017; Shepherd 2015; Shepherd et al. 2015; Dellermann et al. 2018a). Both limitations prevent from providing optimal, scalable, and iterative decisional guidance for entrepreneurial decision-making and limit the integration of stakeholders in this process. Consequently, I identified the lack of investigation of scalable mechanisms that allows iterative integration of stakeholders in guiding entrepreneurial decision-making as the second gap in previous work.
Research Gap 3 – Limited Investigation of IT-supported Decisional Guidance and DSS for Complex Decision-Making under Uncertainty and Risk.
The third gap in the current body of knowledge is related to the design of IT-supported decisional guidance for classes of complex decision-making problems under both risk and uncertainty. Decisional guidance has been proven as a suitable approach in research on decision support systems in various contexts of IS research (Silver 1991; Morana et al. 2017; Parikh et al. 2001; Limayem and DeSanctis 2000).
Although the adaption of these findings to the context of entrepreneurial decision-making is promising, previous research provides little knowledge on both design principles (abstracted design knowledge) and design paradigms (general rational for the decisional guidance provided) for complex decision-making problems under both uncertainty and risk. While DSS that are based on statistical models are consistent (experts are subject to random fluctuations), are potentially less biased by a non-random sample, and optimally weigh information factors, previous work on DSS provides little knowledge on systems that can deal with a such complex class of problems like entrepreneurial decision making. First, despite of advances in deep learning techniques (LeCun et al. 2015), such systems are constrained by a lack of adaptability and are not capable to capture the complex dynamic interactions between elements that are required for providing decisional guidance for situations that require dealing with extreme uncertainty (Slovic and Fischhoff 1988; Zacharakis and Meyer 2000). Second, such methods are having troubles with processing “soft information” (e.g. creativity) or tacit learning experience, which is required to provide decisional guidance for complex problems. Finally, statistical methods struggle with so called “black swan”/” broken leg” events (Dawes et al. 1989) in in which humans are surprisingly good at predicting with a combination of intuitive and analytical reasoning. Consequently, I identified the lack of investigation design knowledge on decisional guidance and DSS for complex decision-making problems under uncertainty and risk such as in the entrepreneurial context as the third major gap in previous work.
Guidance in general proved to be valuable to accelerate entrepreneurial decision-making despite its limitations. Consequently, the idea of this dissertation is to design mechanisms for providing efficient and effective decisional guidance to entrepreneurs that can constraints of human mentoring, integrate stakeholders, and alleviate limitations of recent statistical methods of intelligent decision support systems.
For this thesis, I use the term design paradigm as the general rational for the decisional guidance provided, which is collective intelligence/crowdsourcing (Chapter III) and hybrid intelligence (Chapter IV). Finally, the term guidance design principles (DP) then define the abstract DSR knowledge contribution and learning of the design of Section 5.3, 5.4 and 6.5.
For this purpose, I suggest and discuss two directions to overcome those limitations. First, I propose the design paradigm of collective intelligence (e.g. Malone and Bernstein 2015; Wooley et al. 2010) and IT enabled crowdsourcing (e.g. Leimeister et al. 2009) to overcome cognitive and resource constraints of individual human mentoring and allow the integration of stakeholders, which constitute a main source of uncertainty for entrepreneurs. Second, I suggest the design paradigm of hybrid intelligence that can enhance the limited capability of decision support systems based on machine learning (e.g. Jordan and Mitchell 2015; Goodfellow et al. 2016; LeCun et al. 2015) and leverages the complementary capabilities of humans and machines in making both intuitive and analytical decisions under uncertainty.
As the context of entrepreneurial decision-making is a highly idiosyncratic class of problem, I focus the first part of my thesis on the decision-making context itself and examine how both uncertainty (e.g. Section 4.1) and risk (e.g. Section 4.3) are created as well as the general logic and design of systems that provide decisional guidance (e.g. Section 5.3 and 6.5).
This thesis aims at answering three distinctive RQ related to providing decisional guidance for entrepreneurial decision-making. The general purpose of this dissertation is, therefore, to first examine the decision-making context and then provide design paradigms and design principles for the problem domain.
RQ 1 aims at exploring the sources of risk and uncertainty in the entrepreneurial decision-making context by investigating the role of the ecosystem (i.e. involved stakeholders) in creating such. The general goal of this RQ is to provide a better understanding of the decision-making context in general as well as an in-depth examination of the ecosystem as source of risk and uncertainty. This examination of the problem is required to develop suitable solutions that aid entrepreneurial decision-makers.
RQ 1: What are the sources of risk and uncertainty in the entrepreneurial decision-making context?
Method: Case study research and FsQCA.
Results: Exploration of ecosystem dynamics as source of uncertainty in entrepreneurial actions; examination of the negative effects of uncertainty and dependence on innovation success; investigation of the mechanism of uncertainty and analysis the mechanisms of both uncertainty and stakeholders in the ecosystem in generating risks for entrepreneurs.
Based on the findings from RQ 1, I identified the integration of the ecosystem as generic valuable strategy to manage risk and uncertainty.
RQ 2: How to design for the integration of the ecosystem as guidance in entrepreneurial decision-making?
Following this logic, RQ 2 investigates the design for the integration of the ecosystem as guidance in entrepreneurial decision-making and consists of two parts: First, I conceptually develop a design paradigm for the integration of the ecosystem as guidance in entrepreneurial decision-making.
RQ 2a: What are design paradigms for the integration of the ecosystem as guidance in entrepreneurial decision-making?
Method: Interdisciplinary literature review and conceptual development.
Results: Crowdsourcing to access collective intelligence as design paradigm for decisional guidance; identification of requirements to adapt crowdsourcing for providing guidance in entrepreneurial decision-making.
Second, it is necessary to develop design principles for the integration of the ecosystem as guidance in entrepreneurial decision-making to build DSSs.
RQ 2b: What are design principles for the integration of the ecosystem as guidance in entrepreneurial decision-making?
Method: Design science research projects and conceptual development.
Results: Developing conceptual design principles for a CBMV system for in entrepreneurial decision-making; development of mechanisms for providing feedback and expert matching to apply crowdsourcing for decisional guidance in entrepreneurial decision-making.
Based on the design paradigm and design principles identified in RQ2, the aim of RQ3 is to create knowledge on the design of DSS for providing guidance under uncertainty and risk in entrepreneurial decision-making.
RQ 3: How to design DSS for providing guidance under uncertainty and risk in entrepreneurial decision-making?
RQ 3 again consist of two related parts. The first part RQ 3a extends the findings beyond the scope of ecosystem integration through crowdsourcing and has the purpose of developing more generalizable and superior design paradigms for providing guidance under uncertainty and risk in entrepreneurial decision-making.
RQ 3a: What are design paradigms for providing guidance under uncertainty and risk in entrepreneurial decision-making?
Method: Interdisciplinary literature review and taxonomy development.
Results: Hybrid intelligence as superior design paradigm for decisional guidance to deal with uncertainty and risk; identification of design knowledge for providing guidance in entrepreneurial decision-making.
The second part RQ3b then uses this design paradigm of hybrid intelligence to propose design principles for providing guidance under uncertainty and risk in entrepreneurial decision-making.
RQ 3b: What are design principles for providing guidance under uncertainty and risk in entrepreneurial decision-making?
Method: Design science research projects.
Results: Developing a data ontology and examination of successful decision patterns for entrepreneurial decision-making; development of design principles for a HI-DSS for decisional guidance in entrepreneurial decision-making.
The context of entrepreneurial decision-making describes a specific class of managerial decision-making problem. It is inherently complex as it is uncertain in a Knightian definition (Knight 1921).
More recent research has framed such situations of extreme uncertainty as unknowable risks or unknown-unknowns. Those scholars divide between risk with quantifiable probabilities; uncertainty, which describes risks that are known but cannot be quantified; and the most complex form of unknowable risks or unknown-unknowns where neither outcomes nor their probability distribution can be assessed a priori (Diebold et al. 2010). The latter type of unknowable risk is the dominant form of uncertainty in early stage tech start-ups although all forms exist (Dellermann et al. 2017d). For the purpose of this thesis, I rely on this form of unknown-unknowns when referring to uncertainty.
This facet of entrepreneurial decision-making can be explained as entrepreneurs plan their actions on markets that do not even exist yet or developing novel value propositions which technological feasibility is still unknown. Following this argumentation, the data that would be needed to estimate the probability distributions of certain outcomes or to make assumptions about outcomes does not yet exist (Alvarez and Barney 2007).
This means that even if an entrepreneur would have unlimited cognitive capacity and resources to collect data, she would be unable to correctly quantify the risk (which is the quantified form of uncertainty) associated with certain actions such as the design of a business model (Burke and Miller 1999). Consequently, decision makers are confronted with situations of ‘‘unknown-unknowns’’ (Diebold et al. 2010), “[…] that include both uncertainty and noise due to a large amount of unsystematic risk and conditions of evolving certainty around systematic risk […]” (Huang and Pearce 2015): 636).
Making decisions in such context is highly complex for several reasons. First, not all outcomes of a decision cannot be assessed a priori (Huang and Pearce 2015). Second, even if this was the case it would remain impossible to estimate a probability distribution for such outcomes (Knight 1921). Third, as entrepreneurial decisions and the related outcome highly depend on the ecosystem in which entrepreneurs operates, the decision context is extremely dynamic and dependent on complex interactions (Alvarez et al. 2015). Fourth, entrepreneurial decision-making problems are ill-structured, as not one “correct” solution exists (Simon 1991). Finally, the feedback on weather a decision was good or bad is time-delayed, requiring years to uncover (Alvarez et al. 2013).
Following this argumentation, I define entrepreneurial decision-making as complex decision-making task that requires to deal with both, uncertainty (unknown-unknowns) and risk.
Dealing with such complex decision-making tasks is particularly difficult, as decision makers are not perfectly rational, but bounded rational (Cyert and March 1963; Newell and Simon 1972; Simon 1955). Such bound rationality typically has two dimensions that result in human deviations from optimal action: cognitive bounds and cognitive biases. The first dimension, covers limitations such as basic computational constraints of the human brain such as working memory, information processing etc. The second dimension is related to idiosyncratic human errors that lead to systematic deviations from rationality in judgment and choice (Kahneman 2011). This bound rationality prevents decision makers from optimizing their actions and is the most basic rational for the need of decisional guidance in general (e.g. Silver 1991). Nevertheless, human decision makers use various strategies to solve such problems.
To understand how individual entrepreneurs, deal with such contexts and make decisions, one must zoom into the individual cognitive strategies of decision-making under uncertainty and risk (Tversky and Kahneman 1983; Dane and Pratt 2007). For this study, individual cognitive properties entrepreneurs (Mitchell et al. 2002) will not be integrated in this discussion as this is beyond the scope of this thesis. Rather I will focus on the generic cognitive processes that are applied for making decisions under extreme uncertainty.
The most dominant streams of cognitive psychology assumes that individual decision-making is influenced by two different systems of decision processing (Glöckner and Witteman 2010; Evans 2008). The first mode of reasoning is rather unconscious, rapid, and holistic, more popular under the term of “system 1” thinking. The second type is conscious, slow, and deliberative better known as “system 2” thinking (Kahneman and Frederick 2002; Stanovich 1999). The first mode of thinking is also frequently termed as intuition, which describes a “non-rational” and “non-logical” mode of thinking based on simple heuristics, and mental shortcuts (Epstein 1994; Kahneman and Tversky 1982). The second mode of thinking can be defined as analytical reasoning, which should follow strict rules of probabilistic statistics (Griffiths et al. 2010).
There is a long-standing discourse on which mode of thinking is superior. For instance, intuition is frequently associated with inaccurate or suboptimal choices (Kahneman and Egan 2011; Bazerman and Moore 2008). In contrast, other scholars argue that intuition is often superior as analytical reasoning is limited by working memory, which is especially relevant when decision complexity increases (Gigerenzer 2007).
For the context of entrepreneurial decision-making, previous research argues that the most valuable approach is a combination of analysing and quantifying all available data on the one hand and dealing with unknown-unknowns through intuitive decision-making at the same time (Huang 2017; Huang and Pearce 2015).
Decision makers in the context of entrepreneurship, such as angel investors rely on “algorithm-based” factors to integrate objective and quantifiable information such as financial statements, risk analysis, return on investment calculation, market information, and other forms of “hard” data (Zacharakis and Meyer 2000; MacMillan et al. 1987).
This strategy is typically complemented with a subjective and affective judgement of an entrepreneurial opportunity that is based on intuition and prior experience (Hisrich and Jankowicz 1990). The integration of soft and cognitive factors such as human intuition is a valuable strategy for making decisions under extreme uncertainty (Huang and Pearce 2015).
Consequently, on the individual level of entrepreneurial decision makers a combination of intuitive and analytical reasoning is most valuable for making decisions under extreme uncertainty (Huang and Pearce 2015; Huang 2016).
To address both modes of reasoning and making assumptions about certain actions, entrepreneurs must collect empirical evidence. Using decisional guidance in this vein can support decision makers in situations that consist of both uncertainty and risk (e.g. Silver 1990).
For making analytically supported decisions this means gathering information such as financial data, or market reports (Maxwell et al. 2011; MacMillan et al. 1987). Statistical models that use large amount of data as input are, thus, capable of predicting parts of the outcome and value of certain decisions. Such “[..] actuarial (statistical) models refer to the use of any formal quantitative techniques or formulas, such as regression analysis, for . . . [supporting] clinical tasks […]” (Elstein and Bordage 1988). Therefore, they proved to be a valuable form of decisional guidance in the context of early stage ventures (Zacharakis and Meyer 1998). The use of actuarial models as an analytic for of decisional guidance is valuable as its guidance is consistent, not biased by a non-random sample of prior experience and its “optimal” information factors ( (Fischhoff et al. 1977; Fischhoff 1988; Slovic 1972). Therefore, I focus on ways to integrate such form of decisional guidance in entrepreneurial decision-making through the mechanisms of AI and ML in Chapter IV.
Additionally, for dealing with situations of uncertainty the interaction with an entrepreneur’s external environment (ecosystem) proved to be the most valuable strategy for decisional guidance (Alvarez et al. 2013; Alvarez and Barney 2007).Therefore, I identify the form of guidance that emerges from social interaction with the ecosystem as a proven complementary strategy to improve decision-making through analytical decisional guidance.
This form of dealing with uncertainty are gathering feedback from peers, family members, or friends or validating one’s idea by consultants and mentors (Tocher et al. 2015). Thereby, entrepreneurs test their assumptions against their ecosystem to receive feedback on the viability of their actions. This allows entrepreneurs to cognitively objectify their idea in situations of unknown-unknowns (Alvarez and Barney 2010; Ojala 2016) and persuade a reasonable number of stakeholders of the viability of the opportunity to gain access to further valuable resources that support the entrepreneur in enacting the opportunity (Alvarez et al. 2013). Therefore, I focus on ways to integrate such form of decisional guidance in entrepreneurial decision-making through the mechanisms of crowdsourcing in Chapter III.
For this thesis, I used the business model is as core object when studying entrepreneurial actions and decision-making. Therefore, I will start by defining this term and provide an understanding of the interpretations of the concept that are used for this thesis.
Although lots of different definitions regarding the concept of a business model exist, it provides a holistic framework for the economic model of a firm (Morris et al. 2005; Zott et al. 2011). In general, this model is focused on how value is created and capture (Gassmann et al. 2014). Thus, the business model describes the logic “[…] by which the enterprise delivers value to customers, entices customers to pay for value, and converts those payments to profit [...]” (Teece 2010:172). The business model can, thus, be characterized as organizational design choices that define the “[…] an architecture for product, service and information flows, including a description of the various business actors and their roles […]” (Timmers 1998) and examines “[…] the content, structure, and governance of transactions designed so as to create value through the exploitation of business opportunities[…]“ (Amit and Zott 2001): 511).
Therefore, the business model is “[…] a statement of how a firm will make money and sustain its profit stream over time […]” (Stewart and Zhao 2000). Thereby, it is arranging the operational logic such us internal processes of a firm and its strategy (Casadesus-Masanell and Ricart 2010) and requires decisions on service delivery methods, administrative processes, resource flows, knowledge management, and logistical streams (Afuah 2014).
First, the business model can therefore be used for classifying certain types of firms (Zott et al. 2011; Magretta 2002), which allows to classify new ventures and define similarity among them. This application of the concept is relevant for the expertise requirements and matching of this thesis (Section 5.3; Section 5.4).
Second, the configuration of design choices can be used as antecedent of heterogeneity in firm performance. Therefore, we use the business model to examine its design choices as an important factor contributing to firm performance (Zott et al. 2011). This application of the business model is relevant for this thesis in Section 6.1, where I examine the effect of design choices in defining entrepreneurial success and in Section 6.5, where I use ML techniques for providing guidance on design choices that lead to start-up success.
The business model is core of entrepreneurial actions and related decision-making (Demil et al. 2015). Previous work in entrepreneurship heavily focused on how entrepreneurs create novel opportunities to create value (Shane and Venkataraman 2000). The business model is, thus, applied to provide an explanation and structuring framework for examining entrepreneurial actions by adding “[…] a more holistic, fit-based view of strategic management […]” (Priem et al. 2013). Therefore, it explicitly focuses on the role of users and the ecosystem in explaining entrepreneurial actions by discussing the value proposition (e.g. Chesbrough and Rosenbloom, 2002) or by including the firm’s ecosystem in the process of creating and capturing value from an entrepreneurial opportunity (Amit and Zott 2001; Zott et al. 2011; Zott and Huy 2007; Plé et al. 2010).
Moreover, the business model concept provides a perspective on the relevance and role of implementation when entrepreneurs try to< benefit from an opportunity (Demil et al. 2015). Consequently, the business model can be used to as a kind of action plan for entrepreneurs. The design of a business model is, thus, one of the most pivotal tasks when entrepreneurs aim at capitalizing from an opportunity (Ojala 2016). Therefore, I define the highly uncertain process of iteratively making design choices, testing it against the market and other stakeholder, and reassess the proposed design as core of entrepreneurial action in early stage ventures. Following this logic, I use the design of a business model as phenomenon of interest for examining entrepreneurial decisions and suggesting guidance mechanisms to support such decisions.
Previous research gives a wide array of different interpretations of the concept of business model (Massa et al. 2017). I will therefore provide a discussion on how the concept is used for this thesis.
First, the business model concept defines attributes of a real firm (Casadesus-Masanell and Ricart 2010; Casadesus-Masanell and Zhu 2010; Markides 2013). This interpretation leverages the business model concept as schema for classifying real-world manifestations of ventures and allows the identification of business model archetypes (Johnson 2010; McGrath 2010; Rappa 2001; Gassmann et al. 2014). For this thesis, this interpretation has a dual role. On the one hand, it is used for connecting concrete design choices to firm performance. On the other hand, I apply this interpretation for providing decisional guidance on real-world manifestations of a start-up.
Second, the business model is interpreted as cognitive schema (Magretta 2002; Martins et al. 2015; Chesbrough and Rosenbloom 2002). Previous research argues that entrepreneurial decision makers have an image or a mental model of the firm, not the firm itself (Eggers and Kaplan 2009; Eggers and Kaplan 2013; March and Simon 1958). Consequently, Martins et al. (2015: 105) conceptualize business models as “[…] cognitive structures that consists of concepts and relations among them that organize managerial understanding about the design of activities and exchanges that reflect the critical interdependencies and value-creation relations in their firms’ exchange networks [..]”. For this thesis, I use the cognitive schema interpretation of business models to communicate an entrepreneurs mental model of a start-up to its ecosystem. The business model is for instance used to communicate the mechanisms of value creation and value capture to the crowd (e.g. Section 5.3; Section 6.5).
Finally, the business model has an important role as formal conceptual representation (Osterwalder 2004; Osterwalder and Pigneur 2010). This interpretation connects both the attributes of a firm and the cognitive schema interpretation and highlights the role of the concept in providing a simplified representation of reality (Massa et al. 2017). Thus, it defines an explicit formalization of the firm, written down in pictorial, mathematical, or symbolic form. In the context of this thesis, this interpretation is applied to use ML techniques for examining business model design choices and bringing a human mental model in data representation for the ML part of providing decisional guidance (e.g. Section 6.4; Section 6.5). The use of such formal problem representations that allow to structure knowledge comparable as used in the human mind (Ha and Schmidhuber 2018; Stuhlmüller 2015) is especially relevant for solving AI-complete problems and create a shared understanding between humans and machines (Evans et al. 2018).
Decision support systems (DSS) have a long tradition in IS research and is one of the most pivotal systems that were explored in this field (Todd and Benbasat 1999; Gregor and Benbasat 1999; Alter 2013; Alter 1980; Benbasat and Schroeder 1977).
DSS are a special type of IS that are focused on supporting and improving managerial decision-making (Arnott 2004). Such systems use decision rules, decision models, and knowledge bases to support managerial decision makers in solving semi- and unstructured problems ((McCosh and Morton 1978). Therefore, DSS design an environment in which human decision makers and IT-based systems interactively collaborate. This is especially relevant for providing cognitive aids that assist managers in complex tasks that still require human judgement (Keen 1980). In this collaborative problem solving, human focus on the unstructured part of the problem, while the IT artefact provides an automatic structuring of the decision context (Arnott and Pervan 2005).
More recently, IS research has focused on the application of AI for the purpose of DSS, thus, starting a sub-domain of intelligent DSS (Remus and Kottemann 1986; Bidgoli 1998). These intelligent DSS are for instance rule-based expert systems and more previously ML supported systems that apply for instance ANN, genetic programming and fuzzy logic (Turban et al. 2005). Contrary to the general application of AI in automating tasks and replacing human judgment, DSS aims to supporting the human decision-maker rather than replace her (Arnott and Pervan 2005).
While research on intelligent DSS is a steadily evolving field, knowledge on DSS that are capable to solve highly unstructured and complex problems (i.e. wicked problems) is still nascent (Meyer et al. 2014). Therefore, there is a clear gap in previous work in solving tasks such as providing guidance to entrepreneurial decision makers.
In general, decisional guidance is a concept that describes any aids or advice that tells a human decision-maker what to do (Morana et al. 2017). This is not limited to technological aids, but also other forms of advice such as mentoring etc. For instance, in the context of entrepreneurial decision-making, so far, guidance is provided as face-to-face mentoring in institutions such as business incubators (Dellermann et al. 2018b).
In the context of IS research, decisional guidance Information describes design features of a DSS that provides such advice to the user (Silver 1991, 1990; Arnold et al. 2006). (Silver 2006) defines decisional guidance as “[…] the design features of an interactive computer-based system that have, or are intended to have, the effect of enlightening, swaying or directing its users as those users exercise the discretion the system grants them to choose among and use its functional capabilities […]”. Such advice (e.g. explanations or suggestions) then helps users to achieve a certain goal ( (Benbasat and Wang 2005; Wang and Benbasat 2007)Gregor and Benbasat 1999) and allow to “[…] integrate the expertise of one or more experts in a given decision domain […]” (Arnold et al. 2006:2). Decisional guidance can be described as both a “[…] decision aid as technological intervention [that] should assist in the implementation of normative decision-making strategies; or [a] decision aid as a behavioural approach with the aim of extending the capabilities and overcoming the limitations of decision-makers […]” (Todd and Benbasat 1999:11). Consequently, decisional guidance provides recommendations for solving problems or supports the user in making decisions (Silver 1991).
Decisional guidance can be characterized along ten distinctive dimensions (Morana et al. 2017). Each of the dimensions relevant for this thesis will be discussed in the related Section.
When provided to human decision makers, decisional guidance can influence certain aspects of the context, thereby leading to measurable outcomes (Parikh et al. 2001). First, it influences the decision itself, thus, increasing the quality of a decision (Meth et al. 2015) . Second, it can affect the decision maker, thereby, increasing user satisfaction and learning (Gönül et al. 2006). Finally, decisional guidance can also have an impact on the decision-making process. Consequently, it alters the efficiency of the process itself (e.g. Parikh et al. 2001).
This thesis uses decisional guidance as central phenomenon of interest. For the purpose of this thesis, I use the aggregated definition of Morana et al. (2017: 33), who define decisional guidance as “[…] the concept of supporting users with their decision-making, problem solving, and task execution during system use by providing suggestions and information […]” while “[g]uidance design features refer to the actual implementation of the guidance concept […]”.
As the context of entrepreneurial decision-making is a highly idiosyncratic class of problem, I focus my thesis on the decision-making context itself and examine how both uncertainty and risk are created as well as the general logic and design of systems that provide decisional guidance.
I use the term design paradigm as the general rational for the decisional guidance provided, which is collective intelligence/crowdsourcing and hybrid intelligence. Finally, the term guidance design principles (DP) then define the abstract DSR knowledge contribution and learning of the design.
The effects of decisional guidance on measurable outcomes, however, is beyond of the scope of this thesis. I made these decisions for two distinctive reasons, which I will discuss at the end of the thesis in more detail. First, a lot of previous work examined the effect of decisional guidance in various contexts (e.g. Arnold et al. 2004; Parikh et al. 2001; (Limayem and DeSanctis 2000); Gönül et al. 2006) and non-IT-based decisional guidance is a common form of supporting entrepreneurs (Dellermann et al. 2017c), which leads me to make the assumption that those effects might be similar in this context. Second, measuring the effects of decisional guidance in entrepreneurial decision-making is extremely complicated as the outcome of such decisions are typically several years time-delayed (Maxwell et al. 2011).
Original published on Kindle Direct Publishing