Chapter 2: BI and Analytics Meet Business Transformation
One question frequently asked by BI vendors as well as industry experts is “What is the difference between BI and analytics?” In fact, when doing a Google search for “difference between BI and analytics,” you will receive almost 2 million results (as of the time of the writing of this book).
BI and analytics continue to be important capabilities in organizations around the world. In fact, BI is enjoying a rebirth, courtesy of the data explosion brought to the enterprise by big data, social media, the Internet of Things (IoT), and other sources. Because of this fact, it is definitely time well spent to review 1) What the difference is between BI and analytics, and 2) How BI has changed over the years.
What is the Difference between BI and Analytics?
From a high level, BI and analytics both have the same purpose: both help organizations tap into their data to improve decision making. However, how they actually achieve that goal is quite different.
Traditionally, BI leverages historical data to learn from past decisions, while analytics draw on different sources to predict future results. In other words, BI answers questions like, “what happened with…,” “when…,” “who…,” and even “how many…,” while analytics answers questions such as “what if…” and “what’s next…” Both apply to static data, while analytics often includes time-series data, which is a series of data points captured at specified periods over time. Examples of time-series data includes up-to-the-second stock prices, measurements from equipment sensors, and GPS readings in a navigation device.
|Focus||past, present||present, future|
|Typical users||line of business (LOB), IT||LOB, IT, data science, business analysts|
|Questions answered|| “What happened?” |
| “What if?” |
“What will happen?”
|Deliverables|| ad-hoc query |
| ad-hoc query |
|Methods|| roll up |
slice and dice
| descriptive modeling |
statistical / quantitative analysis
Further, the methods each take are quite different. BI products provide users with tools to perform ad-hoc queries as well as create reports, dashboards, and scorecards as well as offer APIs to embed these objects into or even build applications. Some also provide monitoring and alert capabilities, notifying essential personnel if a threshold has been reached. On the other hand, analytic applications incorporate various mining and modeling frameworks as well as provide support for statistical and other quantitative analyses.
Since organizations often were interested in gaining the benefits associated with BI and analytics, they had to buy multiple applications from different vendors, requiring time, money, and resources to maintain.
With the increased amount of unstructured data being created from various sources, organizations have begun to rethink how to store, analyze, and exploit this information, whether it’s stored on-premises or even on a private, public, or hybrid cloud.