Big data has already become a huge business, and it’s only going to get bigger and even much bigger! Gartner expects the market for Big data and Analytics to generate € 2.11 trillion (US$ 2.7 trillion) in products and services, and generate 4.4 million new jobs by 2014. Let’s quickly review the four main areas IBM suggests we should consider when looking at the challenges attached to Big data:
i. Big data volume
Every day we create 2.5 quintillion bytes of data (2.5 x 1018). How do we turn 12 terabytes of daily tweets into improved products sentiment analysis? How do we sort out this massive onslaught of information and filter the relevant Business Intelligence (BI)? How do we use this information and forecast accurately the development of new products and services to an ever more demanding crowd of ubiquitously digitally connected customers? How do we use these 12 terabytes of daily tweeted information to transform our prospects and customers into brand evangelists?
ii. Big data velocityHow do we analyze the 5 million trade events created each day and identify potential fraud in real time and at best before it even happens? iii. Big data variety
Big data can be presented as structured and unstructured data, such as videos, pictures, files, audio and text. Information can be streamed from the internet but also from surveillance cameras to target points of interest.
iv. Big data veracity
As the sources of data grow on a daily basis, it behooves us to learn how to trust and rely upon the right data we ought to filter. As of 2012, about 2.5 exabytes of data are created each day, and that number is doubling every 40 months or so. However one in three business leaders don’t trust the information they use to make decisions.
What are the three main techniques available today for businesses to harness the richness of this present 21st century Big data tidal wave?
1. Descriptive Analytics for Big data = HINDSIGHT Business Intelligence.Descriptive Analytics or descriptive data has almost become the antiquated discipline of quantitatively describing the main features of a collection of data. It can also lead to inductive statistics or the process of drawing a conclusion from data that is subject to random variation. As a whole, and as its name clearly spells out, it remains solely “descriptive”. The traditional corporate jokes about Controlling informing Sales about issues Sales had already known about, can serve here as a humorous example of Descriptive Analytics! Descriptive Analytics has a definite sense of finality, but nonetheless grants us the hindsight to elaborate possible business forecast outcomes.
2. Predictive Analytics for Big data = INSIGHT Business Intelligence.
Predictive Analytics encompasses a variety of techniques from statistics and data mining that analyze current and historical facts to make predictions about future events. It includes modeling, machine learning, data mining and game theory, to analyze current and historical facts and make predictions about future events. Predictive Analytics can be used to help analyze customer patterns and thus predict future behavior! Amazon.com “Wish list” is a very good example of predictive analytics allowing Jeff Bezos to gather likes, wishes and interests, thus helping his company to customize specific customer promotions, discounts as well as future product choices.
3. Prescriptive Analytics for Big data = FORESIGHT Business IntelligencePrescriptive Analytics is the third phase of Business Analytics (BA). It is the discipline of synthesizing Big data to make predictions and then to suggest decision options to take advantage of the predictions. Prescriptive analytics anticipates what will happen, when it will happen and why it will happen. The interesting aspect of prescriptive analytics is the fact that it can help businesses to foresee possible crisis or problems before they even happen, and thus eliminate the possibility of risky business outcomes.
Most companies still rely on the traditional descriptive data analysis process. How are you coming along with predictive and prescriptive data analysis in your enterprise? Looking forward to your insights and comments.