-
-
Marketplace
-
Channel Resources
Articles from this Site
-
One brand, one Web site! DM Review is now the home of all the content you're used to at BIReview.com and much more. If you are registered at BIReview.com, you're already registered at DM Review. If not, take just a moment to sign up for all the free services we have for you at the new DMReview.com.
Decision Process Automation
The benefits of business intelligence (BI) solutions have been well documented in recent years. One widely distributed study from IDC in 2003, The Financial Impact of Business Analytics, concluded that the average ROI from business analytics initiatives is 112 percent. Other case studies and surveys have shown benefits ranging from the highly qualitative "ability to make better decisions" to quantifiable benefits such as increased sales or decreased costs across organizational functions for customer relationship management (CRM), finance and operations.However, the actual benefits of BI solutions remain well below their full potential. BI solutions remain mostly departmental initiatives focused on reporting. When evaluating the broader market for business analytics, which includes everything from data warehousing tools to BI and packaged analytic applications, IDC estimates that for every dollar spent on packaged transaction processing applications, just 25 cents is spent on business analytics. In other words, there is still a lot of room for improvement in implementing, deploying and leveraging BI and related technologies for improving decision support.
BI solutions serve two primary functions: information delivery and decision support. The former has been the primary focus for most organizations that have relied on their own IT departments or external systems integrators for better methods for delivering information. Over the years the reporting tools have improved in their ability to delivery information faster and more accurately. The market has shifted from batch reports from mainframe systems to client/server-based parameterized reports to highly visual and interactive dashboards and alerting mechanisms. In the course of this transformation, the latency between the time data becomes available and is delivered to end users has decreased. We call this focus on reporting technology traditional BI. However, traditional BI only fulfills a portion of the needs of decision-makers from all levels of an organizations.
Figure 1
It is not enough to simply deliver an alert or highlight a KPI; it is not enough to simply state what happened last quarter or last month or last week; and it is not enough to simply send information to individuals. The graphic opposite depicts IDC's model for decision-centric BI - a closed-loop approach to supporting both information delivery and decision-making needs of organizations. In other words it is a model for decision process automation.
Traditional Business Intelligence:
STREAMLINING INFORMATION ACCESS AND DELIVERY
Traditional BI concentrates on information access by and delivery to individuals. Focused primarily on scheduled, batch-based reporting, information access is being supplemented with near real-time monitoring of business transactions in an effort to reduce data latency and establish early warning systems for the emergence of new trends. As shown in Figure 1, traditional BI software supports activities for the following three steps:
- Track. Results from transactional systems are monitored to get a reading on the state of current operations (e.g., the current marketing or recruitment campaign). These results can be compared with targets or goals that have been established.
- Analyze. Time-oriented data from multiple systems is integrated into a data warehouse or mart in order to support an analysis of key trends. Deviations from expected results or targets, such as factors associated with customer (or employee) attrition, can be explored.
- Deliver information. Reports are published and delivered to business users based on the tracking activities and value-added analysis of the data.
The goals of traditional BI are increased speed and higher accuracy in the delivery of information to ever-increasing numbers of users. A large number of companies will still invest in traditional BI to address end-user requirements appropriate to their skill sets. However, as already explained, information delivery tools address only part of an organization's requirements for BI software.
Decision-Centric Business Intelligence:
ADDRESSING COLLABORATIVE DECISION-MAKING PROCESSES
Decision-centric business intelligence (DCBI) focuses on the decision-making steps in a business process. The goal of DCBI is to assess the relevance of information to a decision and to gain insight in seeking and evaluating possible decision alternatives. These are the steps in the lower half of Figure 1, comprising a collaborative, decision-making workflow:
- Hypothesize. The problem is stated and alternative solutions are sought out. There are software aids for searching for new decision alternatives.
- Model. Models are built to predict the likely result of candidate solutions to a problem. The effects of variable factors on business results are explored or simulated. This is the essence of what-if analysis and data mining. A key factor is the ability to deal with large amounts of data, find patterns and insights, and handle uncertainty by showing the likelihood of a desired outcome when a particular alternative is selected.
- Decide. This is the step in a business process where decisions are made. The results of the analysis and modeling work are considered along with business judgment and knowledge to decide on changes or adjustments to business policies or rules. Decision-making is a process that involves a group of people. Therefore, support for team collaboration is vital.
- Adjust/Act. Making the decision is only the beginning. Its results must be communicated to all people and applications that are responsible for execution. This can involve the translation of the decision into the form that a particular application requires.
The primary takeaway from this methodology is the need to rethink and expand that definition of BI. This process requires close collaboration between line of business and IT stakeholders. On the one hand, business users should begin to request decision support functionality that fits their ongoing business processes. On the other hand, the IT group charged with supporting the BI needs of end users should expand their arsenal of tools and integration services to combine traditional reporting and OLAP tools with predictive analytics, workflow management, collaboration support and unstructured data access and analysis.
In many organizations, IT personnel working on data warehousing and BI projects are not the same as those working on content access, management and search or business process and workflow management. To combat the siloed approach to BI solution development and deployment, many organizations are creating internal BI competency centers or informal groups incorporating multiple IT disciplines with business users and consultants.
Just as ERP solutions enabled the automation of operational processes, BI solutions have the opportunity to provide decision process automation that addresses the competitive and compliance needs of organizations. For decision process automation to happen, organizations need to take a closer look at defining their decision-making processes and incorporating all the appropriate technology, not just reporting, into supporting IT solutions.
Dan Vesset, research director for Analytics and Data Warehousing at IDC, has been a user, implementer and analyst of BI and related software for 10 years. He can be reached at dvesset@idc.com.
For more information on related topics, visit the following channels:
