Each week, DataWorks will define a data-related term marketers need to know. In this first edition of Data Defined, we're starting off with the obvious: Big Data.
What is Big Data?
Big Data refers to relatively large amounts of structured and unstructured data that require machine-based systems and technologies in order to be fully analyzed. The much-hyped term has inspired a slew of definitions, many of which involve the concepts of massive volume, velocity and variety of information. In other words, what turn data into Big Data is the amount of information, and the speed at which it can be created, collected and analyzed.
Another very important factor in distinguishing Big Data is storage. For many entities -- be they huge industrial tech firms or boutique retailers -- the ability to store large amounts of data in the cloud as opposed to storing data in physical servers that require space and maintenance is key to the emergence of Big Data as business reality. Also important to the emergence of Big Data as a real concept for businesses are recently-developed software systems such as Hadoop which allow companies to gather insights from the data onslaught.
The kinds of data and metrics employed to learn from it vary wildly depending on the type of entity collecting and using them. For marketers, Big Data can encompass website log files, consumer-uploaded images, retail transactions, loyalty card information, email communications, mobile location information, and social media commentary about brands.
Srinidhi Melkote, director of analytics, Wyndham Exchange and Rentals, on Big Data:
Any data that contains non-obvious information that businesses can discover and exploit to improve their outcomes is valuable data, even if it's not necessarily "big." For instance, a typical large e-commerce site would collect data on things like product and page views, customer/membership information, product reviews, page shares on social networks, photo and video uploads, product pricing and promotions, customer service phone logs, etc.
This kind of data contains valuable information that helps answer all sorts of relevant questions for which answers are not obvious. Examples of such questions include, "How should I staff my customer service department based on time of day and day of week?" "What kind of promotions spur transactions not just on promoted products but other products as well?" "What kind of product recommendations generate the best profit margins?" "Who are the high value customers?" and "Who are the most engaged customers?"