Human-Centric Decision Support Tools: Insights from Real-World Design and Implementation

Decision support tools enable improved decision making for challenging decision problems by empowering stakeholders to process, analyze, visualize, and otherwise make sense of a variety of key factors. Their intentional design is a critical component of the value they create. All decision-support tools share in common that there is a complex decision problem to be solved for which decision-support is useful, and moreover that appropriate analytics expertise is available to produce solutions to the problem setting at hand. When well-designed, decision support tools reduce friction and increase efficiency in providing support for the decision-making process, thereby improving the ability of decision-makers to make quality decisions. On the other hand, the presence of overwhelming, superfluous, insufficient, or illfitting information and software features can have an adverse effect on the decision-making process and, consequently, outcomes. We advocate for an innovative, and perhaps overlooked, approach to designing effective decision support tools: genuinely listening to the project stakeholders, to ascertain and appreciate their real needs and perspectives. By prioritizing stakeholder needs, a foundation of mutual trust and understanding is established with the design team. We maintain this trust is critical to eventual tool acceptance and adoption, and its absence jeopardizes the future use of the tool, which would leave its analytical insights for naught. We discuss examples across multiple contexts to underscore our collective experience, highlight lessons learned, and present recommended practices to improve the design and eventual adoption of decision support tools. I. The Increasing Prevalence and Importance of Decision Support Tools Rapid advances in information technology are enabling the collection of increasingly vast amounts of data more quickly and easily than ever before (Hoch and Schkade 1996). At the heart of decision support tools lies analytics, the systematic computational analysis of data and statistics to inform decision making, which can be descriptive, predictive, or prescriptive in nature (Davenport and Harris 2017). While the presence of more and better information can empower better decisions, our elevated access to nearly inconsumable amounts of data does not alone guarantee better decision-making. The human mind is limited in available processing power; a key study found that decision makers are unable to identify nearly half of the attractive options (Siebert and Keeney 2015). Information at hand is often only marginally relevant, and the presence of overwhelming, superfluous, and partial information only complicates the decision-making process. There is an increasing need for tools and systems that effectively analyze available data and inform decision makers with fact-based, data-driven insights. Decision makers not only want to find the best solution – they also want it quickly. Decision support tools are computer-based technologies that facilitate better decision-making by solving complex problems and enabling human interaction (Shim et al. 2002). The main aim of decision support tools is to provide decision makers with technology that enhances their capability of decision-making, resulting in making more informed decisions (Arnott and Pervan 2008). Well-designed decision support tools improve the quality of decisions on important issues by removing friction and increasing efficiency in problem-solving. Such systems alleviate the condition of information overload by presenting the right information at the right time, thereby boosting decision-making effectiveness. Various domain knowledge and associated technologies have been incorporated in decision support tools including Artificial Intelligence, Business Intelligence, Decision Sciences, Machine Learning, Operation Research, Psychology, User Experience, and related fields. Many decision support systems combine knowledge and technologies from multiple domains to form an integrated tool to aid in resolving decision problems specific to a certain set of stakeholders. As technology continues to evolve, data-driven decision support systems have advanced in sophistication and application to new and exciting areas. Throughout this study, we refer to designers as the role primarily involved with the creation of the decision support technology, and stakeholders as the general role representing clients, end-users, decision-makers, and their management – really, anyone who is involved in the decision-making process, recognizing that these roles vary from organization to organization. II. Characteristics of an Effective Decision Support System While decision support tools hold great promise, not all decision support tools have a successful story to tell. Many projects were launched to design and develop a decision support tool for a specific decisionmaking context, but ultimately failed because the final product was not successfully adopted by key stakeholders (Pynoo et al. 2013, Bhattacherjee and Hikmet 2007, Freudenheim 2004, Briggs and Arnott 2001, Rainer and Watson 1995, Hurst et al. 1983). In this chapter we focus on factors that lead to successful tool adoption, the most important of which is to design with the purpose of aligning with the needs of key stakeholders. There exists a tendency – perhaps understandably so – for designers to overly focus on the development of decision-making models and algorithms; in so doing, this may compromise the ability to recognize and satisfy the exact needs of decision makers. While cutting-edge algorithms certainly have their place, only by sufficiently aligning with stakeholder needs does any project have the opportunity to succeed. Regardless of the level of technical sophistication, in the end the effectiveness and value of the tool largely depends on the extent to which it will be adopted and put into practice by practitioners and decision-makers (Gönül, Önkal, and Lawrence 2006) and this is integral to decision support tool success (DeLone and McLean 2003). Many factors influence the acceptance, or adoption, of a decision support tool. For the purposes of this chapter, it will be helpful to assume the context in which decision makers already have sufficient trust in the knowledge base and believe that the underlying theory and technology can actually improve the quality of their decision making. In this regard, decision support tool acceptance and utilization depend on two factors: the usefulness of the tool, and its ease of use (Shibl, Lawley, and Debuse 2013). The first influential factor, usefulness, can be defined as the degree to which the tool is compatible with the real needs of decision makers and their belief that their issues and objectives are effectively addressed by the tool. In other words, how much can the design remedy real operational challenges faced by key stakeholders and remove friction from the decision-making process? The second factor concerns ease of use, that is, whether decision makers are comfortable in using the tool on a regular basis. Do decision makers believe that the decision support tool so captures and addresses their needs, that they are motivated to engage with and derive benefit from the tool? III. The Design of An Effective Decision Support Tool While designers of decision support systems may intend to build a tool with effective characteristics that encourage adoption and sustain use, such achievements are far from automatic. The translation of decision context specifications into a solid tool that embeds advanced analytics can be a daunting task. The gap between theory and practice ensures routine encounters with practical, theoretical and technical limitations in the transformation of real-world problems into a decision support context. In many cases, assumptions and simplifications of the original problem must be weighed and specific techniques used to bridge this gap, so as to ensure the final outcome is as close as possible to the initial specifications of stakeholder needs. At the same time, the tool should have a compelling design that motivates stakeholders to routinely engage with it in their decision-making processes. In short, the benefits of gained analytical insights and ease of use should (far) outweigh the various costs such as opportunity, setup, training, and switching. It is therefore critical to understand the behavioral and technical challenges of designing, developing, and implementing successful and effective decision support tools. There are a variety of approaches for designing and developing decision support tools and experts differ in opinion on what methodology works best. Regardless of the chosen methodology, we believe complementary skills and expertise are inherent for successful implementation of decision support tools, and these skill sets are just as important as the theory, knowledge and methodology used in their creation. We maintain that the key to designing successful decision support tools is having deep understanding of the needs of key stakeholders together with compulsion to address these needs through incorporation, to the extent possible, in the tool. Without this, tool design and development take place from the limited 1 For more information on knowledge-based decision support systems, we refer interested readers to Chung, Boutaba, and Hariri (2016). perspective of the designer and developer, rather than the collective perspective that is inclusive of all stakeholders (Power 2002). This should take place through a comprehensive process of understanding stakeholder needs, which allows for effective collaboration through possibly extensive engagement to share ideas and brainstorm better design options. It is true that tool designers are experts in their respective technologies. At the same time, stakeholders are the experts in their own fields and their views should inform and drive the technology. While there