What is Analytics?

What is Business Analytics?

Defining Analytics

As may be expected with much business terminology, particularly one which has generated so much hype, the term ‘analytics’ is arguably more widely used than it is understood. This gap has not gone unnoticed and at the time of writing there are over 83 books available on the subject at Amazon, countless blog articles, and a Google search for the term returns over 4.85million results. Whilst almost any of these will give an adequate understanding of the subject area as whole, the definitions they offer are inconsistent, diverse and often contradictory.

Perhaps the most cited definition of analytics, and seemingly the basis of Wikepidia’s entry on the subject, is provided by Davenport and Harris (2007, p7):

[T]he extensive use of data, statistical and quantitative analysis, explanatory and predictive models, and fact-based management to drive decisions and actions. The analytics may be input for human decisions or may drive fully automated decisions.

[Note: Davenport (in Davenport et al, 2013) later revised this definition to remove the requirement that it drove any decision or action].

This definition infers that there are three key aspects to analytics: data, statistical/quantitative analyses and the use of this as an influence on decision-making. The authors specifically exclude the technological and architectural aspects of delivering analytical insights to businesses, which they instead categorise as “business intelligence” (BI). This view is supported by Randy Bartlett who argues “Business Intelligence = Business Analytics + Information Technology” (Bartlett, 2013, p 4).

However, this distinction between BI and analytics is less widely supported in the literature, with other authors describing BI as a subset of analytics or that the two are in fact exactly the same (a full discussion of this issue will be presented in a forthcoming post). Chen et al (2012) sidestep the issue of separating analytics from BI by instead proposing the two as a composite “BI&A” (an approach also taken in Lin et al (2012)) which they describe as “the techniques, technologies, systems, practices, methodologies, and applications, that analyse critical business data to help an enterprise better understand its business and market and make timely business decisions”. Instead of considering analytics to be a separate development they describe three time periods/stages.

BI&A 1.0: Described as a “data-centric approach”: this stage bore close relation to what would normally be described as business intelligence; incorporating data warehousing, structured data and ad-hoc querying which incorporated well established statistical and data mining techniques (such as clustering or regression analyses) only.

BI&A 2.0: This period, in the authors’ argument, changes were motivated by developments in web technologies and social media. ‘BI&A’ adapted by incorporating web and text analytics and increasingly began to incorporate unstructured data into its analyses.

BI&A 3.0: The author’s consider this stage to be characterised by the use of data collected from multiple devices and objects, as opposed to the 2.0 period which was focused on the emergence of the web on predominantly dedicated devices (e.g. computers and laptops). In particular they highlight devices such as tablets and smartphones, as well as objects or items with RFID codes and tags (i.e. ‘The Internet of Things’).

Such an approach has much appeal; not least because it collapses a distinction between BI and analytics which removes a necessity to characterise each of these related areas. However, some issues remain. Firstly, "BI&A 1.0" would predate discussion of analytics and would effectively rename an existing area that arguably doesn’t need renaming. Secondly, the focus is moreover on data-collection and types than on the processes that constitute each period. Whilst this can be seen as a valid approach, by focusing solely on ‘inputs’ this can undermine the importance of the 'output' (e.g. decision support) or the analytical methods that create this; arguably the more important in terms of usage, development and teaching.

Other definitions in the literature are similarly data focused. Stubbs describes analytics as “any data driven process that provides insight” (Stubbs, 2011, p 10), whilst Isson and Harriott (2013, p 3) suggest “the integration of disparate data sources from inside and outside the enterprise that are required to answer and act on forward-looking business questions tied to key business objectives”.

An alternative definition is offered by Laursen and Thorlund (2010, pXII): “delivering the right decision support to the right people at the right time”. The authors state that their motivation to use such a definition stem from a desire to frame the discussion from the perspective of the business users who will be more concerned with what an investment in analytics may return for the company not how it functions. Similarly, Emblemsvåg (2005) argues that the essence of business analytics is supporting decision making, which he believes only effective when there is understanding: “we must truly understand an issue to be able to act upon it wisely”. He suggests a four stage process whereby data can be used to help create understanding, as shown below in Figure 1.

From Data to Understanding

Figure 1 – The Process from Data to Understanding (adapted from: Emblemsvåg, 2005)

However it is his belief that the analytical practices of BA cannot produce this understanding alone. The quantitative processes that may be used in BA can be used to translate data into information. However, in order for this to become knowledge in the mind of the decision makers he argues that further, non-analytical work is required and points to analytics as moreover “attention generating instruments than knowledge generating instruments” (Emblemsvåg, 2005). To translate this into the final stage, understanding, the author argues the information gained from the analytical insights needs to be framed by the decision-maker in the full context of the situation; in other words combining this with domain or situational knowledge.

Both these definition make particular emphasis on analytics as most significantly a support mechanism for decision makers. However, such definitions would seem to exclude decision automation, an application of analytics highlighted in Davenport and Harris’ description as well as in many other papers (e.g. Panian, 2008; Niedermann et al, 2011). As such these definitions would seem to suggest that analytics exist solely for the purpose of supporting higher-level decision-making and strategy. Whilst this may be one use of analytics, its role in automation would suggest that it also has value in making ‘smaller’, day-to-day decisions and those that cannot be practically made in the appropriate timeframe by human interaction alone.

INFORMS, the US society for operations research and management science, define analytics as: “the scientific process of transforming data into insight for making better decisions” (Boyd, 2012). With such a concise definition there is less to find objections to, and would seemingly extend to incorporate decision automation. However, there is little to distinguish this from BI which would presumably propose similar aims and it is unclear whether a “scientific process” would necessarily exclude the processes of computer science to transform raw data into information to be used by businesses without any quantitative analytical processes.

Whilst these definitions are varied, four central themes are clear: the inclusion of technological aspects (i.e. computing), quantitative methods, decisions making resources (GUIs, data visualisation, communication methods, etc.), and data (this point is further expanded upon in a separate post). Consequently it is these elements (i.e. the full lifecycle of data from its collection to its dispersion amongst decision makers) that will be the basis of the definition offered here:

"Business analytics is the process of transforming data, from a variety of sources and of a variety of types, into insights that support, improve and/or automate business decisions, using technological, quantitative and presentation techniques."

In keeping with the literature this definition demonstrates three essential aspects:

1. The variety of data that can be analysed in the process.

2. The critical aspect of analytics remains the output: that is the support, improvement and/or automation of business decisions, resulting in improved performance or the generation of business value.

3. The application of the process involves transforming the data using a mixture of technological, quantitative and presentation techniques (such as data visualisation or GUIs).

Whilst this definition seems in keeping with the literature discussed and conceptually fitting, a field such as analytics is fast-moving and contentious, and so this cannot really be seen as any more than a "working defintion". However, such discussion does forward awareness and the debate about the topic, and provide a framework for the other articles on this site.

Have your say by leaving a comment below!

 

REFERENCES

Bartlett R (2013). A Practitioner’s Guide to Business Analytics: Using Data Analysis Tools to Improve Your Organization's Decision Making and Strategy. McGraw-Hill: New York.

Boyd E (2012). Revisiting ‘What is Analytics. Analytics Magazine, [Online Version], July/August 2012, p6. Available from: http://viewer.zmags.com/publication/40f74a63#/40f74a63/7, [accessed March 2013].

Chen H, Chiang RHL and Storey VC (2012). Business Intelligence and Analytics: From Big Data to Big Impact. MIS Quarterly, 36:1165-1188.

Davenport TH and Harris J (2007). Competing on Analytics: The New Science of Winning. Harvard Business School Publishing Corporation: Boston, MA.

Davenport TH, Bensoussan BE and Fleisher CS (2013). The Complete Guide to Business Analytics (Collection). FT Press: Upper Saddle River, NJ.

Emblemsvåg J (2005). Business Analytics: Getting Behind the Numbers. International Journal of Productivity and Performance Management, 54: 47-58.

Isson JP and Harriott J (2013). Win with Advanced Business Analytics: Creating Business Value from Your Data. John Wiley & Sons: Hoboken, NJ.

Laursen GHN and Thorlund J (2010). Business Analytics for Managers: Taking Business Intelligence beyond Reporting. John Wiley & Sons: Hoboken, NJ.

Lim EP, Chen H and Chen G (2012). Business Intelligence and Analytics: Research Directions. ACM Transactions on Management Information Systems, 3: 17-27.

Niedermann F, Radeschütz S and Mitschang B (2011). Business Process Optimization Using Formalized Optimization Patterns. In: Abramowicz W et al (eds.). Business Information Systems: Lecture Notes in Business Information Processing, 76: 123-135. Springer: Berlin.

Panian Z (2008). A Break-Through Approach to Real-Time Decisioning in Business Management. In: Proceedings of the 2nd WSEAS International Conference on Management, Marketing and Finances, WSEAS, Harvard, MA, pp 25-30.

Stubbs E (2011). The Value of Business Analytics: Identifying the Path to Profitability. John Wiley & Sons: Hoboken, NJ.

You are here: Home Analytics Articles What is Analytics? What is Business Analytics?

Contact us

  • This email address is being protected from spambots. You need JavaScript enabled to view it.     Connect via LinkedIn    |    In assosciation with:    The OR Society    Loughborough University    |    About the Project

  • Address:

    The ORATER Project, C/O MJ Mortenson, School of Business & Economics, Loughborough University, Leicestershire, LE11 3TU