A Formulaic Approach for Selecting Big Data Analytics Tools for Organizational Purposes

Introduction

Big data has become one of the most essential tools in organizations and the analytics aspect of it makes it more viable. Big data is characterised by volume, velocity, variety, and veracity. Currently, there are four most common types of BDA, which are diagnostic, descriptive, predictive, and prescriptive. Based on this premise, there is a need to investigate the impact which characteristic (volume, velocity, variety, and veracity) of BDA (descriptive insight, predictive insight, and prescriptive) have on innovation competency in an organization, and collaborative positive insights (Ghasemaghaei & Calic, 2019).

Big data Analytics (BDA) tools are increasingly used by different organizations across sectors, ranging from e-commerce, government administration, education to science, technology, and healthcare (Vassakis, Petrakis & Kopanakis, 2018; Wang et al., 2018). Grable and Lyons (2018) explain how BDA tools (software) are used for marketing, insurance, telecommunications, retail and fraud detection purposes. E-commerce organizations use analytics tools for activities such as examining website traffic, purchases, and to determine customers’ interests and types and frequencies of transactions (Picciano, 2012). Some institutions of higher learning employ BDA tools to assess and address institutional performance, students’ performances, and overall progress, to predict the future in teaching and learning. In healthcare, BDA is used to analyse patients’ profiles and patterns of health conditions (Raghupathi & Raghupathi, 2014; Sagiroglu & Sinanc, 2013). From marketing perspective, many organizations employ BDA tools to gain deeper understanding about their customers’ behaviours, needs and preferences (Watson, 2014). In a nutshell, BDA enables and supports organizations towards sustainability and competitiveness (Kuoppakangas et al., 2019; Grover et al., 2018).

Despite the coverage and premise, there are challenges in selecting the most appropriate analytics tools, particularly by organizations in developing countries (Iyamu, 2018). This can be attributed to primarily two factors: the practice of the concept is slow (Kuoppakangas et al., 2019); and availability of skilled personnel are scarce. For an organization to gain useful insights and value from big data, it requires the application of the most appropriate BDA tools for the analysis (Al-htaybat & Alberti-alhtaybat, 2017), which has been a challenge for many organizations. There is emphasis on the ‘most appropriate’ because no one tool fits all or can fulfils every requirement. Organizations apply BDA differently as their businesses and environments dictate. Some organizations integrate the tools with other software within various platforms such as the data warehouses. According to Nakashololo and Iyamu (2018), other organizations apply BDA tools in isolation, to test their potential value. Hernandez and Zhang (2017) argue from a different angle, that healthcare organizations primarily apply BDA tools to construct patients’ unified datasets. This is different from how other organizations employ the tools, where the focus is mainly on business strategy and organizational processes for effectiveness and competitiveness (Sharma, 2015). Irrespective of the focus and approach, BDA tools remain a challenge in many organizations, which perhaps triggered a multilevel analysis approach (Iyamu, 2018). This seems to have contributed to solving only the analysis aspect of the challenges. The challenges of selecting most appropriate BDA tools remain.