Whether you call it Big Data, data science, or simply analytics, data is a gold mine in Asia. With the regional Big Data technology and services market projected to grow at a 34.1% compound annual growth rate (CAGR) from US$548.4 million in 2012 to US$2.38 billion in 2017, according to IDC, there are plenty of opportunities in the space to find an effective way to harness data. Few businesses would pass up an opportunity to predict future events, better understand their clients, or otherwise improve their standing, especially when data on the region is relatively scarce. Still, many of these same companies fail to realize they may have rich sources of data, much less understand how to capitalize on them.
Many businesses today are looking for ways to move from the traditional data warehousing model to one that enables them to utilize the data intelligently to improve processes or services. There are various ways to make data profitable, but there are five core data utilization models that businesses at any level can start with. These five models vary according to complexity and outcomes, and are tailored specific to each business requirements and available resources. CIOs can use these proposed models as a starting point before moving on to more sophisticated strategies.
One of the most basic data utilization models, Collect/Supply model essentially involves the process of gathering and selling raw data. This is hardly a new business. In fact, most companies still continue to use this model out of convenience and complacency for more "innovative" approaches. This model involves basic data gathering, often done manually, where the data are being sold to interested parties thereafter. Unlike physical goods, the same dataset can be sold multiple times in the pretext that they are still relevant and current. Take the case of a database provider selling the same dataset to multiple marketers - while the cost of creation might be high (in the case of manual data entry that cannot be automated), there is near-zero marginal cost of distribution, especially if the data is being distributed electronically.
"Bad data" comes in many forms. One common case is data with malformed, missing, duplicate, or incorrect records. Another business idea, then, is to supply a "clean" dataset that removes or corrects these rogue records. While budget-constrained organisations may prefer purchasing raw data and manually cleaning it themselves, many would prefer on more refined sets.
One of the benefits of the Filter/Refine is that the technical grunt work of cleaning the data are done before they are sold, so clients can focus on the real work. For example, someone who runs a mailing list subscription service will gladly pay for someone else to remove duplicates and filter out non-existent or fraudulent addresses. They do not want to be in the business of managing data; they want to use data to drive their business.
Sign up for CIO Asia eNewsletters.