What is a Big Data Strategy and Why is it Important?

The following is an edited extract from The Enterprise Big Data Framework.

The pivotal question that every organization needs to answer, before investing any money into Big Data technology and solutions is their definition of value. What constitutes value in an organization, and how can that value be expressed in return on investment? For example, if a company is able to have a 2 per cent better marketing conversion than a competitor, what is that worth? Or what is the finan­cial value of a 5 per cent more accurate inventory prediction?

Although these fi­nancial estimations seem trivial, they form the crux to success in Big Data. Organizations that can determine the financial value of enhanced decision-making are able to select the right optimization projects and can prioritize different analy­sis projects over others. Organizations that don’t have this capability will come to recognize that their investments in Big Data teams and Big Data solutions are fre­quently not paying off.

Data sets in the world have become so incompressible large, it is neces­sary to focus on the data that matters and, more importantly, on the way that data brings value to an organization.

Purpose and structure of a Big Data strategy

Participants of the famous World Economic Forum in Davos, Switzerland already declared during the annual meetup of 2012 that Big Data has become a strategic economic resource, similar in significance and liquidity to currency and gold.  Most organizations have long-term plans and strategies in place to deal with their eco­nomic plans. Enterprises and companies go through yearly budgeting cycles, in which the planning for the subsequent year and their strategic choices, are defined and explained. If we consider Big Data as an economic resource, similar in nature to other assets, defining a Big Data strategy is a logical step.

The Big Data strategy capability provides focus and guidance for everyone in the organization. It helps to determine where organizations should focus their efforts and investments and what the medium and long-term objectives are, through which an organization can outperform competitors. Despite the great growth and popularity of analytics and Big Data solutions over the last decade and the realization that data contains value that can provide a way to obtain a competitive advantage, multiple research reports show most fail dra­matically, with reported failure rates as high as 85 per cent. They have, in other words, not defined how success in Big Data should be measured and lack a Big Data strategy.

To interpret these high failure rates in the correct way, and to provide guidance to others on how to avoid this in future (i.e. building a capability), it is important to note that an effective Big Data strategy is more than just a written document. Although a formulated Big Data strategy document is a great first (and necessary) step towards effectiveness, it is also important that the formulated Big Data strategy is meaningful, executable and innovative.

Because an effective Big Data strategy is more than just a document, the Big Data strategy capability has been subdivided into several micro-capabilities, as depicted in Figure 5.1[LM1] . Not surprisingly, strategy formulation is an important micro-capability. However, for a robust and competitive strategy, organizations also need to consider innovation management, leadership and governance, sustainability and communica­tion. If, and only if, these micro-capabilities are effectively combined can an organi­zation claim that it has a future-proof Big Data strategy.

Barriers to an effective Big Data strategy

Many organizations are struggling with the implementation and execution of a Big Data strategy and, as a result, many Big Data initiatives fail. Be­fore it is possible to introduce capabilities that help to overcome this problem, it is advisable to take a step back and consider the main reasons why Big Data strategies have failed. What is the reason that large and successful companies have difficulties with making Big Data (and hence data-driven decision-making) a successful enter­prise function? And what are common denominators that can be used in the micro-capabilities?

Over the years, many organizations have tried to understand common inhibitors and barriers to Big Data success. A common classification is to subdivide these bar­riers into technological barriers, human barriers and cultural barriers, as depicted in Figure 5.2. [LM1] Whether this classification is fully complete or whether there are some additional inhibitors is, from a strategy perspective, not a very interesting discussion. The main point is to learn from these barriers to determine which capabilities can help organizations overcome failure in their Big Data strategy. By considering barri­ers to success closely, we can identify critical capabilities that are necessary to over­come any inhibitions.

The difficulty with each of these barriers is that they cannot be placed in isolation. For example, hosting training programmes to address a lack of skills (a human bar­rier) will elevate skill levels for a certain group but is by no means a guarantee that a big data programme will subsequently be successful. In a similar way, a successful Big Data strategy cannot be seen in isolation. It is the result of several different com­ponents (which we labelled micro-capabilities) which will need to work together in a coherent way.

Big Data strategy: a leadership responsibility

The implementation of a business strategy is a complicated process, and a Big Data strategy is no exception to this rule. It can even be stated that the execution of a Big Data strategy is more complicated, given the technical complexities inherent in Big Data. The formulation of a strategy alone is not enough for successful implementation; the additional capabilities are required.

When it comes to the implementation of Big Data strategies, leadership and deci­sion-making play a critical role. Because of the technological, human and cultural barriers, the leadership of an organization is ultimately responsible for the success or failure of the Big Data strategy. Individual leaders are required to create and sustain the vision of the organization and to subsequently build the required capabilities.