The process underlying big data and data analytics

Beyond having access to massive amounts of data, what may fir st come to mind when thinking data analytics is the knowledge of programming or conducting statistical analysis to uncover insights from data. This goal can be achieved within many organisations or outsourced in a relatively straightforward maimer. However, when it comes to analytics, there is a more pressing need/challenge for organisations. Davenport’s (2013) decision-making process map showcases the importance of this step (Figure 1.1). Fu st and foremost, what is the organisation trying to address for its benefits? In other words, what is the question that the organisation is seeking to answer? Before anything else, the organisation must be aware of what issue or challenge they are trying to tap into, and this must be translated into a research question(s) that can be addressed by the data at hand. This research question must be an appropriate question - clear and specific enough for the data to uncover insights (Mayhew, Saleh, & Williams, 2016). As simple as this may seem, it is not an easy exercise to complete. An organisation must have the capacity to pinpoint its priorities, needs and challenges at any given time. When it comes to data particularly, the organisation must also have appropriate data to address the question. If the data are largely irrelevant to the question the organisation is trying to address, then how the data are gathered could be reshaped to address the question, which is an indispensable exercise to complete. If no data have been collected there is a need to implement data collection beforehand (see Batra, Jacobson, & Santhanam, 2015. on advanced analytics for instance).

Another important matter, also highlighted by Davenport, which is often lacking in data analytics processes across disciplines, is how to best interpret and disseminate the findings emerging from the data. This exercise requires a skill set in itself - more specifically, having a good understanding of what the organisation is trying to achieve, a general sense of business administration, and a capacity to be creative with analytics in order to identify patterns that may exist but are

Figure 1.1 Decision-making process for analytics

Source Adapted from Davenport (2013)

difficult to uncover in the first place; this skill set is scarce in reality. Performing data analytics needs to be understood as much more than clicking a few buttons with a program and looking at what comes out the other end of a computer. It requires curiosity, persistence, and a capacity to understand different possibilities the findings may be suggesting. Disseminating the findings to colleagues and clients is not necessarily an easy task either but must be equally emphasised otherwise all these efforts spent on data could essentially be worthless (Davenport, 2018). For analytics to benefit an organisation, the findings must be communicated clearly to the people in the organisation. The findings must be translated in a simple language so that people in the organisation can understand what these mean for them and for the organisation. The decision makers should also be clear on how to use the findings to implement a strategy that will benefit the organisation.

Importantly, being able to ask and answer appropriate questions, and to interpret and disseminate the findings, and to implement an organisational strategy based on data, all require a culture that will facilitate this process in the first place. For instance, the idea of ‘culture catalysts’ or analytics translators and support from decision makers in the organisation appears to be imperative if an organisation is seeking to be successful with an approach using evidence-based knowledge to benefit its needs. Underlying the culture and the challenges, a plan is needed to set up an approach driven by data from the organisation. The plan should match investment priorities with the organisational strategy, seek to balance the need for affordability and speed with business realities, and ensure frontline engagement and capabilities (see Biesdorf, Court, & Wilhuott, 2013).