REDMOND, Wash., Feb. 7, 2001 — It may be one of the biggest business clich
é
s of our time, but it is also a fundamental truth: we are overwhelmed by information.
Powerful digital technologies have made it easier than ever to collect nearly unlimited data on every aspect of the business process. Companies today amass information on customers and what they buy, how promptly they pay for it, how many referrals they generate and how much customer support they require. They accumulate performance data about operations, competitors and partners, and click-stream data on the way visitors move through Web sites. There are forecasts on everything from the state of the economy to the price of energy to digest, and demographic and market trend reports to keep track of, and countless email messages to read.
For the average knowledge worker, accumulating data has become far too easy. The problem is that their ability to amass data far exceeds their ability to use it. Most business managers are swamped by the sheer volume of information that they need to sort through to make good business decisions. Most data sit unused in huge databases. Ironically, the end result of all this data collection is that many companies understand less about how their business works than they ever did.
“Ten years ago there was no such thing as ERP, or supply chain management, or CRM,”
says Clay Young, vice president of marketing for Knosys. Based in Boise, Idaho, Knosys is one of the worlds leading providers of business analysis and decision-making applications for large corporations.
“Most of us didnt sell through catalogs, e-commerce, and brick-and-mortar stores all at the same time. Now we do, and weve got all this data. But we have no better visibility into what it all means. The truth is its probably worse.”
However, there is good news for every business decision maker who is wondering how to make sense of the gigabytes of data at his or her fingertips. A new set of technologies is delivering the tools needed to take that data and shape it into a useful picture of a company, its customers, and its competitors. The result is called
“business intelligence,”
and its arrival signals the beginning of a new era in the way companies do business.
“Weve been through the information revolution; now were at the dawn of the knowledge revolution,”
explains Rupert Bonham Carter of Cognos, which provides business intelligence solutions to enterprise customers around the globe.
“Weve been collecting data since the punch clock, because we knew someday wed be able to make use of it. Today, the systems needed to get value out of that data are becoming widely available. Business intelligence means that the promise of knowledge at your fingertips is finally becoming a reality”
Business Intelligence for Everyone
“Business intelligence”
is the ability to integrate all of the data that a company and its partners collects and organize it through a computing platform that gives knowledge workers quick, intuitive access that allows them to make and implement decisions quickly, based on a detailed strategic understanding of the business landscape. Business intelligence provides decision makers with tools for analyzing trends, discerning patterns, and evaluating relationships. It delivers the knowledge needed to make critical decisions based on facts rather than on intuition.
“Business intelligence is about leveraging all of your data assets to improve your decision-making capabilities and gain a competitive advantage,”
says Steve Murchie, Microsoft group product manager for SQL Server.
“It means gaining a holistic view of your customers so you can truly begin to understand who they are, where they are coming from, why they buy the things they buy, and what you can do to keep them coming back for more. With business intelligence, you can determine whether your Web site is increasing sales or just taking business away from your retail channel. Youll know with certainty which customers bring lifelong value to your company and deserve the highest levels of service and attention, and which actually cost you money.”
The enabling technologies are complex. Business intelligence is built on distributed computing systems that aggregate and organize data from multiple operational systems into data warehouses and
“data marts.”
It requires services that cleanse and extract those data, and tools that facilitate sophisticated analysis. Multidimensional databases and OLAP (on-line analytical processing) cubes are essential components.
The concepts and the technologies behind business intelligence are not entirely new. Data warehouses date back to the late 1980s and business intelligence systems have been in place at some of the worlds largest corporations for more than a decade. But they were prohibitively expensive to build and maintain, and extremely complicated to use. Moreover, extracting real business intelligence required a team of data analysis experts. As a result, access to the information was restricted and the value of such systems was sharply limited.
That is all changing dramatically. Rapid advances in the technologies that make business intelligence possible are taking it out of the realm of experts and delivering it into the hands of almost anyone who manages any aspect of a business.
“Ten years ago business intelligence was a field for artists,”
says Dr. Claudia Imhoff, author of the book The Corporate Information Factory and president of the data-management consulting firm Intelligent Solutions.
“Those systems cost tens of millions of dollars to build, and you had to be a real specialist to run one. Then, about six years ago there was an order of magnitude change. Better tools for accessing and monitoring data helped make it more of a craft than an art and it lowered the cost from tens of millions to just millions.”
Another dramatic change came about two years ago, when the cost of a creating a data warehouse and implementing business analysis applications fell again by a factor of ten.
“Now weve made the next drop, and its no longer just for artists or craftsmen,”
says Imhoff.
“Weve reached the age of mass customization, where we finally have business intelligence for the masses. Now middle-tier and even lower-tier companies can look seriously at taking advantage of these business analysis tools.”
A Watershed in the Evolution of Business Intelligence
The catalyst for this age of mass customization was the release of Microsoft SQL Server 7.0 in 1999, which included an OLAP server and Data Transformation Services, features that no other database product could match. It was the first database system that contained all of the tools needed to build a data warehouse and business intelligence system.
SQL Server 7.0 was the culmination of five years of hard work, according to Microsofts Steve Murchie. During that time, Microsoft completely re-architected the product, rewriting 90 percent of the code from earlier versions of its SQL database. In 1996, the company also acquired Panaroma Software Systems, giving Microsoft access to the worlds best OLAP technologies.
The inclusion of OLAP services in SQL 7.0 marked a real watershed in the evolution of business intelligence, says Rupert Bonham-Carter of Cognos.
“By providing this powerful and inexpensive platform for business intelligence, Microsoft has done something that probably no other company could have done,”
he explains.
“They have made business intelligence a mainstream technology.”
As a result, business intelligence is one of the fastest growing segments of the enterprise software industry. According to Survey.com, the market for data warehousing and business intelligence software will reach $26 billion in 2001 and is growing at a rate of 28 percent a year. SQL Server itself has seen revenues jump more than 50 percent over the last year, compared with a growth rate of just 17 percent for Oracle, SQLs highest-profile competitor.
Those figures probably underestimate the impact of SQL Server on the business intelligence marketplace, says Nigel Pendse, co-author of The OLAP Report.
“The number of seats is expanding at about a 40 percent rate,”
he says.
“If you just look at revenue, it might appear that the market is slowing. But thats because SQL Server costs so much less than any other product. If you look at users, its clear that there is rapid expansion. And as more people wake up to quality of SQL Server, that growth will accelerate because cost will no longer be a reason not to proceed.”
How good is SQL Server? According to Steve Murchie, it is the only product that offers comprehensive business intelligence functionality in a single box. He also says it offers the fastest time to market, the best ease of deployment, the greatest scalability, and the best reliability of any database product on the market.
He is not alone in his assessment. A study by the American Institute for Research found that database administrators using SQL Server accomplished tasks three times more easily than those using Oracle. The Aberdeen Group reports that the cost of SQL Server is just 27 percent of the cost of Oracle when you include purchase price and maintenance, and that SQL delivers TCO that is 3.7 times better when administrative expenses are factored in. In addition, SQL Server holds five of the top 10 results on the latest TPC-C benchmark tests, along with the many of the most important industry benchmarks, including those created by SAP and PeopleSoft.
One of the best people to turn to for an assessment is Nigel Pendse. His OLAP Report is widely acknowledged as an outstanding source for up-to-the-minute information on the business intelligence world, and he has a reputation for being brutally frank in his analysis.
“This is remarkably good technology,”
Pendse says.
“It scales well. It performs well. It integrates well with relational sources. It is remarkably bug-free, and it is cheaper than anything else on the market. It is genuinely the best product on the market. The one place where it has any weaknesses is the front end, but in many ways thats a good thing because it has left the market open for third parties to create applications that sit on top of SQL Server, and there are many excellent choices out there today.”
The availability of great third-party applications is a key part of Microsofts business intelligence strategy, and it is helping companies around the globe to achieve remarkable results through the implementation of business intelligence solutions built on Microsoft technologies.
Hewlett-Packard is just one example. In 1999, the companys Consumer Products Group built a business analysis system using Microsoft SQL Server 7.0 and the ProClarity Analytical Platform from Knosys to help it examine the sales and distribution of printers and other imaging products through large resellers like Best Buy and Office Depot. At the time, the company had collected vast stores of data but was struggling to find a cost-effective way to utilize the information. The goal was to find patterns that would help them understand what effect advertising and distribution had on sales. Once the platform was in place, Hewlett-Packard was able to quickly begin making fundamental changes in its advertising program and distribution system.
“In the past, HP treated all distribution channels and all regions as equal,”
says Clay Young of Knosys.
“But that meant that they might have an item understocked in a region where it was selling well and overstocked where it wasnt moving at all. ProClarity has helped them analyze sales data and focus their marketing and distribution efforts so that advertising targets customers who are most likely to buy and inventory is distributed to stores where it is most likely to be purchased.”
The Next Step
In 1999, the release of SQL Server 7.0 dropped the price of business intelligence systems from millions of dollars to hundreds of thousands. New Internet technologies and the release of SQL Server 2000 take that one step farther, putting business intelligence within reach at prices that were unimaginable even two years ago.
One company taking advantage of these new possibilities is Seattle-based digiMine. Founded last year by a group of Microsoft veterans, digiMine delivers business intelligence over the Web. Utilizing an applications service provider (ASP) business model, the company builds, hosts, and maintains databases for client companies and provides the front-end reporting and analysis tools needed to make sense of the data. The cost savings are significant. Many of digiMines customers were spending millions of dollars per year trying to understand the data they were collecting and getting limited results. Today, they are finally beginning to see the full benefits of the business intelligence revolution, and it is costing them less than $10,000 a month.
“We left Microsoft to start digiMine because we saw the technology was there, but companies were having a hard time creating solutions that really worked,”
says digiMine President and CEO Usama Fayyad, who did pioneering data mining work for NASA in the early 1990s before joining Microsoft.
“What digiMine does is take best-of-breed technologies, which includes Microsoft SQL Server, Microsoft Windows, the OLAP services in SQL Server and more, and put together compelling solutions that really deliver value and are relatively inexpensive. Thats democratizing what was strictly a tool for high-end, Fortune 500 companies.”
According to Fayyad, the very existence of digiMine is proof of the power of Microsofts business intelligence technologies.
“We launched last year in March with three founders and by July we had a beta product with 10 customers,”
he says.
“These are very complex and sophisticated technologies and the fact that we could get up and running so quickly is a testament to just how easy it is to put these solutions together and build enterprise-class solutions in a short time.”