With various ways to apply BI, there is no one right way to analyze information. Organizations can choose to look at historical data, identify trends, measure current performance against set targets, develop statistical analyses based on what-if scenarios, etc. The possibilities are endless, which is one reason why business intelligence is widely applied within organizations at the individual, departmental and enterprise level.
Although traditionally used for historical analysis, the trend towards operational intelligence and the need to maintain a competitive market advantage means that organizations are required to use a forward-looking approach regarding how they analyze information. Consequently, the use of predictive analytics within BI is gaining momentum. This article highlights why the shift towards predictive analytics is occurring and the benefits of taking a proactive approach to data analysis.
The Shift Of BI From Historical To Present To Predictive Data Analysis
In general, organizations apply business intelligence in one of three ways. They analyze historical trends and develop reports, they use analytics on top of operational systems to get a constant view of performance and key metrics, or they apply statistical reasoning to the information captured to develop what if analyses. Each of these applications offers benefits to organizations but adopting predictive analytics within a BI framework enables organizations to plan for future growth and avoid potential risks by looking at historical and current performance on top of defined factors. Overall, BI maturity helps organizations move from one end of the spectrum to the other to get the most out of their business intelligence solutions.
Many business intelligence applications aim to help organizations get a better picture of what is happening within their company and in their respective competitive markets. This provides the basis for traditional BI by collecting data over time and identifying potential trends. By looking at these trends as well as at an organization’s performance, organizations can stay current by identifying what has occurred within various business units and how this performance compares to overall strategic goals.
With traditional BI, organizations generally capture monthly, yearly or (in some cases) daily data to see whether they achieve their goals and increase their market share as desired. The easiest example is the identification of sales performance over time. How is product X performing in region A in comparison to the last three quarters or versus product X performance in region C? And why do these discrepancies exist? By combining this information with external factors including location intelligence, market trends, weather, competitive product releases, etc., organizations can start to identify why they did not meet their targets - or alternatively, the factors that contributed to their success. This application of business intelligence offers organizations great benefit, but does not extend beyond looking at information that may be considered stale.
Alternatively, organizations develop an operational approach to business intelligence by using BI as a gateway to understand what is happening within the organization on a daily or intra-day basis. This operational view of BI is more closely related to proactive BI use, but still lacks the predictive or forward-looking approach to data analysis. Sales analytics and dashboard solutions that sit on top of Salesforce.com offer a good example of how organizations use business intelligence as an extension of their sales, CRM or ERP solutions to keep on top of what is happening within their organization.
Predictive analytics is most often applied within organizations as part of an overall BI framework. Historical and current information is used to develop predictions about what might happen in the future. Generally called what-if analysis, organizations can identify the type of scenarios they want to look at and use data mining and statistics to develop answers to questions about future sales, product positioning, market trends, etc. With the expansion of the types of data being analyzed, organizations can develop broader questions and applications to expand BI within the organization by being more proactive and forward looking. Forecasting provides a good example of how predictive analytics can be applied within the organization.
Why Be Predictive?
Business intelligence is continuing to shift away from historical analysis and towards analyzing operational data and using predictive models to help plan for the future. The increase in data volumes, expansion of supported data sources, and the ability to load data at regular intervals daily without affecting the production environment changes the way organizations can apply BI. Traditional BI was appropriate when loading millions of rows or more could only occur as part of daily or weekly batch jobs. With data warehouses that easily store TBs of data and with the ability to look at social networking sites and analyze Web traffic, looking at historical information is no longer enough to stay ahead of the competition.
Now many organizations apply BI on an operational level through the use of dashboards on top of their operational applications, which can help call centers identify employee performance, customer churn rates, etc. Moving beyond the historical and operational applications of BI enables organizations to take advantage of the historical and current data that has already been captured. That information is then used to help identify scenarios related to sales, customer experience programs, company expansion and the like. Because data that is both internal and external to the organization can be captured and analyzed, organizations can also look at market data, how this data compares to their sales figures or other information and use that to develop what-if analyses.
Overall, the application of predictive modeling and data mining by looking at the data already collected in a data warehouse can help organizations go beyond financial forecasting and towards identifying future product success and upcoming industry trends. Predictive analytics gives organizations the ability to pool the information they have and develop flexible scenarios. This provides analysis capabilities that includes the ability to change variables within scenarios to identify how third-party and market changes might affect overall performance. The real advantage of using predictive analytics is the ability to take out the guesswork from long-term planning and the ability to better align strategic goals with project success.
About the Author
Lyndsay Wise is an industry analyst for business intelligence. For over seven years, she has assisted clients in business systems analysis, software selection and implementation of enterprise applications. Lyndsay is the channel expert for BI for the Mid-Market at B-eye-Network and conducts research of leading technologies, products and vendors in business intelligence, marketing performance management, master data management, and unstructured data. She can be reached at firstname.lastname@example.org. And please visit Lyndsay's blog at myblog.wiseanalytics.com.
(Copyright 2009 - Dashboard Insight - All rights reserved.)