As the chief analytics officer of a growing, innovative company, you have bigger fish to fry than to worry about whether “Big Data” is a singular or plural noun. But at the risk of flouting good Latin, we’re going to treat “Big Data” as a singular, collective noun denoting the vast mountains and deep oceans of information being amassed through digital networks and manipulated to create business value.
In a previous blog post, we framed Big Data merely as enormous, complex, fast-changing data sets. But a few years ago, the Ivey Business Journal defined big data as “a capability that allows companies to extract value from large volumes of data” (emphasis added). This is helpful because it reminds us that Big Data should never be mistaken for a business goal.
And as with any business goal, achieving it requires strategy. So what strategies could be applied to capitalize upon Big Data?
Expert organizations like SAS, IBM, McKinsey, and so on, each have their specialized perspectives and proposals. But the strategic framework in that now-old Ivey Business Journal article provides a sensible, adaptable and easy to understand overview of the landscape. It looks like this:
Source: Ivey Business Journal
This framework posits two dimensions, the first focused on business objectives. When you leverage data to attain business objectives, you are either measuring something or experimenting with something. In the case of measuring, you want to learn how well a known asset is performing. When you have some hypotheses you wish to test, you’re experimenting with something unknown; data guide your conclusions.
The authors call the second dimension data type of which they see two: transactional and non-transactional — more often referred to lately as “structured” and “unstructured” data, respectively. Transactional, or structured, data are those captured in the normal course of business: sales data and conversion rates, for example. Non-transactional, or unstructured, data are those that do not conform to any pre-defined model; they can’t be captured in a row-column database. They’re usually text-heavy and are drawn from such sources as social media. Unstructured data may require elaborate algorithms to analyze them and to yield patterns.
The four quadrants generated by the crossing of these two dimensions can help businesses articulate what it is they wish to set their sights upon and approximately how to get there. Putting together a Big Data strategy will require investment in creating an approach that integrates all types of data to generate business value.
Among the great challenges posed by Big Data is recruiting the expertise needed to understand it and put it to use. As SAS Data Management breaks it down, strategic necessities include
- Knowing how to integrate data from two or more sources;
- Rearranging data into formats suitable for reporting and analytics;
- Improving the quality of raw data by removing duplicates and verifying addresses; and
- Bringing the business and IT sides of the house together to govern data assets.
As a result, we’re seeing the rise of the CDO (chief data officer), a rocketing need for data scientists and data translators, analytics moving into the cloud, and data monetization becoming a major source of revenue. With the growth of Big Data and its consumption comes investment in big-data analytics hardware, software, and services, in data scientists and their continuing education. Forecasters consistently project massive increases in VC funding. Forbes, for example, projects the Big Data investment market soon to surpass $200B, up from $130.1B last year.
Wading into the deep ocean that is Big Data isn’t easy or cheap. But the bottom line is this, as one source put it: “If you’re not doing data analytics yet, start.”
As a forward-thinking innovation center, tekMountain specializes in making sense of emerging opportunities and techniques. With our team of seasoned mentors and an established entrepreneurial network, we invite you to explore what Big Data has to offer your enterprise. Let’s get started.