The democratization of business intelligence
From data repositories to collaboration platforms
The need for competitive analysis and other high-level information has spread throughout the organization.
By Kim Ann Zimmermann
There was a time when if you wanted a report that analyzed information from a variety of sources—a marketing database, a sales forecast and a customer database, for example—you were one of a chosen few in an organization who could have access to that bit of business intelligence. To actually get the information you needed, however, and to share it with colleagues was difficult at best.
For example, a marketing manager at a bank might look at customer information that was stored in a number of databases—checking, savings, credit cards and loans—when trying to determine which customers were the most profitable to pursue for a new product or service. Prior to business intelligence systems, that marketing manager would have had to enlist programmers to develop software to gather and analyze all of the information. The rest of the knowledge workers at the bank were relegated to using spreadsheets and other crude methods of data comparison.
"We've really graduated from the spreadsheet," says Dave Henry, VP of product marketing for Sagent. "Today's organizations need to respond much more quickly to market conditions. That means putting more information in the hands of more people in a organization. It's not just an issue of getting information to a few data analysts who are going to interact with the data warehouse."
According to Henry, the democratization of business intelligence is part of a larger trend of using business intelligence to collaborate among departments within an organization as well as with trading partners and others outside of the company."What we're seeing emerging is business intelligence Web services. After you've done all the analysis, you want to share it with other people," Henry says. "Take a manufacturing setting, for example. There are several people in an extended supply chain who need access to information that is being developed within the company. With Web-based business intelligence, it is easier for people from multiple companies to share information."
Traditional business intelligence systems haven't been appropriate for collaboration. Henry says, but "now you can have a chart or a graph that is updated every hour. Everyone in the supply chain can collaborate looking at the up-to-date metrics for immediate business activity monitoring."
Data standardization and data integrity are especially important for accurate business analysis, so a number of safeguards have been built into many business intelligence systems.
"We've added some data standardization tools to ensure data quality and make sure that there are no missing values or incorrect data getting into the data warehouse," says Henry.
As the trend toward standardization takes hold, BI is becoming a true Web-based technology, according to Neil Patil, senior director of product marketing for Brio. "As we deliver business intelligence to the mainstream, the Web is the platform of choice. This is no longer a technology that is locked in by the financial analysts and IT guys. It is for executives as well as people on the front lines," he says.
As more people in an organization use BI, the need grows for pre-packaged metrics. "We're seeing more and more companies asking for standard packages that look at key performance indicators," Patil says. "They need to get metrics that might relate to the sales pipeline or supply chain or fulfillment."
That trend has also led to more personalization of business intelligence systems as well as modules tailored toward specific job functions, such as sales or marketing. Patil says, "If you're in sales, the system knows your profile and which business metrics relate to your business functions."
As more people in an organization use business intelligence tools, it becomes more important that those tools be adept at Melding together data from a variety of sources, according to Louise Wannier, president and CEO of Enfish.
"It is not just enough to pull together financial data, marketing data and information from the Web, a business intelligence system needs to be able to pull that information together in a fashion that is relevant to the user," she says.
Wannier gives the example of a salesperson preparing to call on a new client. He or she might want to pull in information from an internal database, a report that a colleague wrote, information from the potential client’s Web site as well as the latest news items written about the company. "They need that information in front of them on one screen in an easily understood format. They not only need the document, the system has to point them to the right section of the document," Wannier says.
Business intelligence tools are evolving from repositories of data to platforms for collaboration. "Whereas before information that had been produced by a workgroup was stored on the company intranet, it would never be found unless the person who was doing some business intelligence work knew where to look," says Wannier. "Now, that information is distributed and available to everyone."
The collaborative nature of new business intelligence systems means extending availability beyond the organization, according to Donald MacCormick, global product communication director for Crystal Decisions. "It is important to be able to share this information with selected partners outside of the organization," he says. "Companies are finding that they need to deliver these capabilities to their partners, customers and suppliers."
Business intelligence systems also need to be flexible to meet the needs of various users and departments. MacCormick says, "You might have one person who wants a report on the top 10 customers, another one is looking at analyzing phone bill costs." Key to the success of any business intelligence systems is scalability. "Over the past few years, there has been significant change in the amount of data and the number of people who will have access to that data for business intelligence purposes," says MacCormick. "You're getting this information out to thousands of people, instead of a select group of people, so you really do need to scale up quickly and efficiently."
Kim Ann Zimmermann is a free-lance writer, 732-636-3612, e-mail email@example.com