Stratagem

Data Warehouse or Business Intelligence?

One of the frustrating and confusing aspects about the Business Intelligence (BI) and Data Warehouse (DW) field is the fact that the industry cannot settle on a single term to describe it. I guess this shouldn’t be surprising given that the industry also can’t agree on a single architecture, data modeling technique, reporting tool or development methodology either!

Back in the days when information technology was called electronic data processing, what is today called business intelligence or data warehousing was called decision support. Beginning in the late 1980s, a number of leading technology consultants began popularizing the term data warehousing to describe an end-to-end solution for extracting, storing and reporting business data.

About ten years later, the term business intelligence emerged to describe a more business focused approach to data warehousing that put the business users’ access and information visualization front and center when it came to deploying an enterprise information repository.

While we no longer use decision support to describe the technology industry, I continue to hear people using both business intelligence and data warehouse – sometimes interchangeably – to describe an enterprise information repository. Because there is no consensus on a name, I’ve been seeing the rather clunky phase “data warehouse/business intelligence” (BI/DW) used more and more in articles and books to describe this industry.

A few years ago, I was working as a consultant on several BI/DW projects simultaneously. As I drove around Wisconsin I had to remind myself to use each client’s preferred BI/DW terms to avoid being misunderstood.

When I was overseeing a financial reporting project for the Department of Workforce Development in Madison, I referred to what we were building as a data warehouse. On the other hand, I was helping build an enterprise business intelligence system when I assisted the Harley-Davidson team in establishing their business intelligence project methodology. When I headed up the road to Appleton, to work for Thrivent Financial, I was building a data warehouse when I was reviewing data models but deploying business intelligence when I spent time with the team charged with establishing a business intelligence center of excellence.

To my mind, the terms business intelligence and data warehouse are not the same thing; although they are linked in that you cannot have business intelligence without some kind of a data warehouse. Just like a physical “house” is required to make a “home,” a data warehouse is the physical foundation upon which an organization builds its business intelligence competency.

Business intelligence is the merging of strategy, people, process and technology for the purpose of transforming business data into business information. This transformation of data to information allows the people who work for the business or interact with the business to make better decisions using reliable, understandable, and timely information.

A data warehouse, on the other hand, is a critical foundation (databases, tools, interfaces, methodologies, etc.) that must be in place to support the organizational development and effective use of business intelligence.

To deploy true business intelligence, an organization must have (in addition to a data warehouse and its associated components) a solid business intelligence strategy aligned with the larger enterprise strategy.

In addition, BI requires that certain organizational structures and teams be in place to build, use, and enhance the organization’s business intelligence competency.

Lastly, because, BI is an ever changing and maturing competency, meaningful business intelligence requires that an organization has processes in place to constantly re-evaluate and enhance its business intelligence strategy, organization structures, and BI processes.

Business Intelligence and the 2008 Recession

In the last few months, the economy of the United States, along with the economies of much of the rest of the world, have fallen into what is most likely a deep and long lasting recession. During such difficult times, it is inevitable that companies attempt to increase revenue while they work to reduce – some might say, slash – operating budgets in an attempt to survive until better economic times return.

Business Intelligence (BI) was made for times like these.

Companies that embrace BI stand a much better chance of surviving a serious recession than those firms that have not integrated powerful analytics capabilities into their day-to-day operations. Companies that understand their customers, suppliers, competition, and marketplace have a significant advantage when it comes to such things as anticipating customer demands, managing their supply chain, exploiting new revenue opportunities, and reducing the cost of service while maintaining quality.

Gartner Inc. and The Data Warehouse Institute (TDWI) predict that organizations will paradoxically spend more on BI during the current downturn relative to the rest of their IT budgets in an effort of maximize the value of their information assets. In particular, companies are expected to spend considerably more on predictive analytics along with increased spending on BI search and discovery applications during 2008 and 2009.

During recessionary times, companies must pay increased attention to their bottom line while at the same time maximizing revenue potential from every available channel. BI helps them do this by giving managers and operations staff improved insight into their costs, their profitability, and their overall operations. Companies that have integrated BI analytics and architectures into their key business processes are better able to see the “big picture” of their operations and thus better optimize their critical business processes.

In addition to the traditional deployment of BI tools to analyze what has recently happened in a business, companies should consider deploying data mining and predictive analytics technologies to identify cross-selling and up-selling opportunities, as well as to find new sales prospects. From an expense perspective, analytic applications for financial analysis, customer profitability, and especially corporate performance management should be implemented. Some analytic applications (such as those associated with campaign management) will generate additional revenue while helping to monitor and control expenses.

Fraud attempts tend to become more prevalent as the economy weakens. Business intelligence can be used to track and analyze fraud attempts and data mining can be used to predict common characteristics of those likely to be involved in fraudulent activities.

One of the keys to survival in these turbulent times is enterprise agility. Agile companies use BI to sense changes in their business environment (competition, economic shifts, customer demands, regulatory changes, technical advances, etc.) so they can swiftly and effectively respond to these changes.

Companies that have embraced BI as a component of enterprise agility are ones where:

• Senior management treats BI as a strategic resources
• BI is seen by the business as an enabler that delivers high and sustained value
• BI governance is effective and supported by a robust infrastructure
• Data quality is high and getting higher
• BI is pervasive throughout the organization
• People, tools, data, and methodologies are in place for rapid BI development and support.

Given the near meltdown of the financial and auto industries, any governmental rescue effort will almost certainly result in increased regulatory reporting and oversight to more closely monitor performance, risk, and financial stability. Firms may have little or no latitude when it comes to funding and deploying BI to meet these regulatory reporting demands.

The deployment of business intelligence during difficult economic times should not be considered a luxury. Instead, BI should be seen as a valuable strategic tool that can help agile enterprises identify revenue opportunities, meet regulatory requirements, prevent fraud, and control expenses.

The Value of Business Intelligence Assessments

Most people take time to plan for retirement, put their kids through college, or organize a yearly vacation. So why do so many people responsible for their company’s business intelligence program fail to adequately plan for what is usually a large, complex, and costly effort?

With nearly 50% of all BI projects ending in some degree of failure, it is hard to understand why companies continue to neglect to allocate even a modest amount of resources to this critical business intelligence exercise. An exercise, that when done right, can literally mean the difference between a functioning business intelligence capability and a lot of wasted money.

A properly performed business intelligence assessment provides a low-cost, actionable examination of three areas critical to a BI initiative – business needs analysis, organizational analysis, and technical/methodology analysis.

An assessment forces a company to take the time to examine its BI strengths and weaknesses across these three areas. It also provides various actionable recommendations on how to fix potential weaknesses, as well as exploit strengths. An assessment process is designed to give a “state of the union” when it comes to BI and can save a company time and money before it rushes headlong into a project. In addition, an assessment, as performed by Stratagem, provides clients with a project timeline and budget, a recommended BI organization, various technology solutions, and a risk/benefit analysis.

In short: A Road Map for BI Success.

The best time to conduct an assessment is before project launch; however, it is never too late to do a BI assessment. If BI to some level is already in place at a company, an assessment is a great way to help determine whether BI is providing the company with any value. An assessment can also show how people are currently using this capability or whether additional opportunities exist for BI enhancements. Of course, if a BI effort falls into that 50% failure bucket, an assessment is pretty much of a must do so that the project can get put back on the rails.

There are both long- and short-term benefits to a business intelligence assessment. In the short-term category, conducting a BI assessment is a great way to validate a company’s BI assumptions and proposed BI direction. In many cases, some of the results of the assessment can be acted upon immediately even before the full BI capability is built out, thus providing immediate return on investment. Long-term benefits include helping to establish a cohesive, evolutionary vision for BI that changes and grows as the company does.

Perhaps the greatest benefit of an assessment comes from the discoveries that a company makes about itself. For example, one company that Stratagem recently work with discovered that their marketing, sales, and finance people all used the same basic metrics to run their portion of the business – basically, the same data but with different names. The assessment helped them realize that efficiencies could be gained by developing a closer collaboration between these separate teams.

A typical assessment takes from four to six weeks to complete. However, when done right, it is an investment in time and money well spent. After all, you wouldn’t take a cross country trip without a map would you?