Article Preview
TopIntroduction
In today’s business milieu, visual analytics has emerged as a technological innovation offering analytical capabilities to provide users at different organizational levels with information and insights for their decision-making activities (Sharda, Durus, & Turban, 2015). Enterprises continuously accumulate an abundance of raw data, and their ability to produce smart decisions is in part based on their perspicacity and efficiency in analysing and interpreting data (Clarke, Tyrrell, & Nagle, 2016). If critical data are not presented and their insights are uncovered, it becomes increasingly likely that enterprises will make weak decisions as a result and will fall behind their competitors with greater proficiency in data analysis and management (Stodder, 2015). As enterprises are engaged with data-driven decision-making, this also requires that business users become comfortable with, and sophisticated in, using data to guide and inform their decisions. In this context, visual analytics role to support business decisions becomes paramount, as it enables enterprises to extract raw data and present them in a meaningful way that generates the most value (Maheshwari & Janssen, 2014). Moreover, visual analytics is prompting a cultural shift toward analytics-driven enterprises by empowering a broader range of business users to explore more data and discover new insights that they can employ readily to improve business strategies, processes, and customer engagement (Stodder, 2015).
Visual analytics is defined as a process of analytical reasoning facilitated by interactive visual interfaces (Thomas & Cook, 2006). It is emerged from the need to more deeply analyze data, and it strives to support the analytical reasoning and decision-making process by creating systems that maximize the human capacity to perceive, understand, and reason about complex, dynamic data and situations (Adagha, Levy, & Carpendale, 2015). In addition to their inviting visual interfaces, visual analytics systems often incorporate forecasting, modeling, statistical, what-if, and predictive analytics (Keim et al., 2008). These capabilities give users more power and control over analytic discovery, enabling them to progress further on their own, in a self-service fashion, rather than depending on IT developers’ intervention (Chinchor, Cook, & Scholtz, 2012; Stodder, 2015). Consequently, nontechnical business users across the enterprise are enabled to analyze and drive insights from datasets of increasing size and complexity. This can be important in larger enterprises, where there are often considerable IT application backlogs, as well as in small and midsize enterprises, where there is a lack of extensive IT support for data analysis (Stodder, 2015). These capabilities are expected to enhance the flexibility and adaptability, and subsequently an enterprise’s long-term competitiveness and survival.
While business enterprises are clearly motivated to deploy visual analytics technologies, what is less clear, and also of central importance, for an enterprise introducing visual analytics technology is to be cognizant with the factors driving its users to accept and adopt visual analytics systems. Despite the numerous cognitive and communicative advantages of visual analytics, enterprises have implemented visual analytics systems only to find that the systems are not used or do not contribute value to the enterprise, according to a recent survey by The Data Warehousing Institute (TDWI) (Stodder, 2015). The same survey revealed that business users still spend almost two-third (65%) of their time analyzing data in tables and text, while only 12% of respondents ranked tables as “highly” useful in helping them to glean insights and making informed decisions. It has been suggested that such failures are caused by over-reliance on IT and data scientists in the enterprise (Stodder, 2015). However, the challenges of adopting visual analytics do not merely depend on the enterprise’s technological abilities, but on how well visual analytics systems and applications support or ‘fit’ the needs of the enterprise and the work-related tasks of its users (Chinchor et al., 2012).