Business Intelligence Solution Goals

For decades, the computing industry has focused on creating applications to gather input from human users and to collect data from any sources. Now, nearly all businesses have, or have access to, huge repositories of transactions: records of information such as product sales, survey results, phone calls, and the like. In databases, all of these things are stored as simple values in fields and tables. Before these numbers and characters were collected from millions of keyboards and application users, the data was useful information about people, products, and other things that real people cared about. Somewhere along the line, it all got transformed into data and stored in huge databases. One of the most important goals of a BI solution is to retrieve pertinent data out of these enterprise databases. It's to transform all of the right numbers and characters into useful information in such a way that it tells a meaningful story about the information that business leaders need to know to make decisions.

Think about the words business and intelligence and how they should be used together to describe an ideal business environment. Just about everyone can tell sad stories about the inefficiencies and blind direction of many businesses. If business leaders can make informed decisions, completely aware of product demands, customers' needs, competitive forces, sales trends, manufacturing costs, and surpluses and shortages, running a stable and profitable business would be much easier. Customers would want to buy your products and services if they thought your business was intelligent-that the people who ran your company were smart and made intelligent decisions.

A functional business intelligence solution makes this dream a reality by putting relevant, reliable information in front of business information workers and leaders in the right form, at the right level of detail, and at the right time. Relevant data means determining the kinds and detail level or granularity of data that is required to support the business analysis.

To design such a solution, it's important to understand the goals of a business intelligence solution. Let's review them.

Combining Relevant Data from Multiple Sources

Source data is transformed into valid and trusted information. Because data has been reconciled and validated, report data becomes a single version of information. Business users know that this is the point of reference for all other data.

Users understand that the time frames of the information can be current or historic. This enables information users to gain a perspective of the present and the past, so they can understand and respond to trends and variances.

Information is structured according to pre-defined relationships and rules. By conforming to business and industry standards specific to the type of data being presented, users already know how to use and report on information using familiar patterns. They can also make comparisons to information from other sources because it conforms to industry rules and forms.

Providing Fast and Easy Access

How fast is fast enough? Getting data quickly is a very subjective goal. For someone accustomed to waiting all day for an old report to run, a 30-minute query might be just fine from the user's perspective, but the network or database administrator might not be thrilled with the impact it has on servers and the network environment whenever the report runs.

Getting information quickly might not be challenging if this is the only goal. However, getting consistent and reliable information quickly and easily might be another issue. To be useful and trustworthy, information must be of consistent quality. Users are loyal to a system that they can trust and, to earn this status in the minds of users, it is helpful to assure them that data is always accurate and consistent. Business users certainly appreciate a simple, user-friendly interface, but if the system isn't effective enough to deliver the right information quickly-and in a form that is easy to understand-it might as well print hieroglyphics.

There are a lot of different business reporting tools in use today, but many go unused because they are cumbersome for report designers and inconvenient for users. They can also be inefficient or not integrated with other solution components. The kind of solution that users will consistently go to for critical business decisions is obedient to their simple request to provide an answer to a specific question. After posing a seemingly difficult question about the uncertain seasonal sales trends of a particular product commodity, the CEO will be more than cheerful when handed a report that clearly shows the trends for the past ten years based on solid facts and a promising future.

He'll be even more ecstatic to learn that the system, which supports simple, ad hoc queries and elegant reporting, was a thrifty investment. It required only a few brave IT workers who were willing to step outside of their comfort zone and learn to use a new set of tools to build a true data warehouse, automated ETL process, and OLAP reporting cubes.

A solid BI system provides direction. It's fast, reliable, and easy to use, and the data is always clean. In time, business leaders and information users will trust the system with a sense of reverence and respect because it helps executive leaders keep the ship on course and to make course changes only when the compass on the executive dashboard tells them to.