gosphotodesign - Fotolia

How are business intelligence and predictive analytics different?

Understanding what sets predictive analytics apart from business intelligence is important in an age when gathering key insights in critical. Here's what you need to know.

A detailed exploration could be written about the differences between business intelligence and predictive analytics. Here's the short answer: BI attempts to make sense of what has already happened, while predictive analytics uses information about what has happened in the past to project what might happen in the future.

Both business intelligence and predictive analytics applications analyze historical data, but predictive analysis uses data to model probable future events. BI focuses on packaging, presenting and exploring the data.

Business intelligence has been around for many years, and it is the evolutionary successor to simple reporting. Generally associated with dashboards and other forms of visual presentation, BI creates value from large amounts of data by grouping, organizing, analyzing and packaging large quantities of individual facts into a form that brings out underlying truths, patterns, trends and relationships -- in a word, intelligence. BI is what draws value from ERP and other information systems for executives and high-level managers.

Predictive analysis uses various analytical and statistical techniques to build a model that describes an aspect of the business based on patterns and relationships. The model can then project what might happen in the future if past conditions remain or, more importantly, what might happen if conditions change. This "what-if" or simulation capability is what gives predictive analytics its real power and value. The predictive model can provide informed insight into likely customer behavior, product supply and demand, fraud detection, risk assessment, sales and margin predictions, and more.

Predictive analytics software, as a natural extension of data mining and business intelligence, is being developed and offered by the same suppliers, including ERP suppliers, often as companion products or product extensions.

The biggest impediment to more widespread adoption of predictive analytics is that it can be complex and often requires specialized or highly trained people, i.e., data scientists, to reap the benefits. The industry is focused on simplifying advanced analytical applications to overcome this problem and to make them more user-friendly, or at least more usable for business users and other non-scientists.

Both business intelligence and predictive analytics are used to gain important insights. However, predictive analytics expands the benefit of a BI application, moving from "what happened?" to "what might happen?"

Industry observers caution that there should be a clearly identified business purpose for the analytics and predictive tools in order to reap real business benefits. It is far too easy to get lost in the data and lose focus on what is of importance and of value to the organization. 

Next Steps

Analytics and the internet of things for the supply chain

Sensors boost insight for manufacturing

The internet of things is changing supply chain management

Dig Deeper on Financial analytics and reporting