Print this article | Return to Article | Return to

Designs of Intelligence

As companies extend business-intelligence software to the masses, they should think before they deploy.
Scott Leibs, CFO Magazine
November 1, 2006

Companies go to enormous trouble to capture and store data, and then go to even more trouble (and expense) when they attempt to extract and use it. The systems for accomplishing the latter task go by various names — business intelligence (BI), performance management, analytics — and while a technologist could make a case as to why one category differs from another, for most corporate employees the distinctions are largely moot.

A current spate of mergers will further blur those boundaries, as BI vendors rush to add performance-management applications to their "product footprints" and performance-management vendors expand into analytics. A few companies already cut across all three categories, and the world's largest software vendors, including IBM, Oracle, and SAP, are rumored to be exploring acquisition targets.

That urge to merge is motivated at least in part by the relative success of the (broadly defined) BI market, which is booming at an annual rate of 11.5 percent, and the business-performance- management category, with growth of 12.8 percent, compared with overall growth in the software market of just 8 percent. In turn, that growth is being driven by the spread of the technology to more desktops in more parts of the organization. Traditional BI software has typically been the province of technologically astute analysts, who use it to generate reports for employees who need to know what the numbers say without learning how to make them talk. But newer performance management and analytics applications are being provided to employees throughout a company, from the executive suite to the call center.

Companies have tried to provide those capabilities to the masses before, with mixed success. Business intelligence is sometimes branded as a classic example of "shelfware," software that is bought but not necessarily used or even deployed. There are a number of signs, however, that the situation is changing.

For one, mergers take some pressure off companies to stitch together products from multiple vendors in order to address both the back-end data needs and the front-end presentation and interface issues that BI projects entail. While a genuine merging of disparate technologies typically takes a year or more following an acquisition, at the least a vendor may now be able to offer a broader family of products and help clients piece them together. Over time, analysts say, that integration will exist out of the box and may be largely invisible to end users, who may find themselves tapping the powers of BI and related technologies without even realizing it.

That, says IDC analyst Dan Vesset, is at the heart of what he terms the "third wave" of BI. In this latest wave, "a vast population of employees whose BI requirements have not been met" by the initial waves of mainframe and client-server-based technologies will find that new Web-based software improves information flow and decision-making. Expanding the reach of BI, says Vesset, will depend not only on technological advancements, but also on how companies learn from past mistakes and match their investments in software to their business practices.

Competency Centers
At De Lage Landen, the asset and vendor finance arm of Dutch Rabobank Group, the creation of a BI "competency center" has played an important role in bringing BI to more workers. Launched two years ago, the center aimed to push a range of BI capabilities, including desktop scorecards that track key performance indicators, out to employees as quickly as possible.

"If we had left that role to IT," says Daan Greven, the firm's corporate manager for process, controls, and data management, "things would have taken longer" because BI would be fighting for attention amid a score of other projects. Instead, a team of five people now fields all company requests for anything pertaining to (in this case) BI software from Hyperion Solutions. "The team members are a mix of the very technical and those with more of a finance/economics background," says Greven. They address problems and requests for changes and enhancements to the system and are part of a larger "information infrastructure" group headed by Greven that ultimately reports to the chief financial and risk officer.

Competency centers have been championed by Gartner and other consulting groups and are popular in the IT world for everything from ERP implementations to integration projects to exploring new technologies such as open-source software. When the concept is applied to BI, however, success usually hinges on strong finance-department involvement and close communication with business units that might benefit from an infusion of BI. "Ideally [a competency center] should report to finance," says Hyperion chief strategy officer Howard Dresner, "because finance is by its nature a cross-functional group, and the currency of business is money."

Dresner notes that while there were no BI competency centers six years ago, today it's estimated that 20 percent of large companies have one, and another 20 percent plan to add one within a year.

While competency centers play a useful role in helping employees actually use whatever BI software the company rolls out, Greven says the centers do more. "They have a very important advisory role in helping us realize our long-term vision of turning data into useful information that can help us drive strategy," he says. For example, while the center responds to employee requests for help or added capabilities, it isn't merely passive: it also helps push the technology further into the company by understanding how the software's capabilities complement a given department's needs. In a sense, it acts as a collective tutor and coach.

And Greven says the center has proven to be an excellent launching pad for new finance staffers, because it gives them a broad introduction to all corners of the company, and a look at how technology intersects with everything from basic financial-reporting needs to more sophisticated analytics projects.

Culture Counts
Whether a company establishes a formal competency center or not, senior finance executives have had enough exposure to BI software in its many forms to understand that a company's culture will shape just how quickly and how far the technology can expand. When Dan McGowan, vice president of financial reporting and analysis at Southeast Corporate Federal Credit Union, arrived at the company three and a half years ago, "we had very antiquated systems — in fact, I'm being generous to even refer to them as systems." He found that it was relatively easy to get the finance team to handle internal and statutory reporting on a new performance-management system (from Applix), but when he then tried to roll out certain BI capabilities to the investment and human-resources groups, he realized that "people differ greatly in basic PC proficiency. Some people aren't even sure how to open a spreadsheet file."

While the company did offer some training in the form of one-on-one sessions McGowan conducted, he says that "successful organizations are marked by people who are naturally curious and willing to try new things." He agrees with a fundamental tenet of the competency center approach, however, in that "leadership in BI and performance management has to come from people who understand both the business entity and the capabilities of the latest tools." McGowan got a look at that firsthand when his company's IT department took an early stab at building a data warehouse. "Their intentions were good," he says, "but few employees understood how to use the system or even why they would want to."

At mortgage insurer The PMI Group, CFO Donald Lofe says that to improve financial planning and analysis throughout the company, "we needed to get better IT capabilities to the troops" in the form of BI software (from Cognos) to drive the firm's strategic planning "from the ground up." Lofe says the firm offered various levels of training depending on an employee's needs, and, more important, "we made sure we had constant phone support from either a PMI staffer or a vendor employee so that absolutely no question went unanswered." The goal, he says, was to keep post-training momentum high by addressing any concerns or questions immediately, so that employees developed an affinity for the new software.

At the same time, customers are pushing vendors to make such software simpler. "It has to be easy," says John Kennedy, controller for the locomotive and rail-car services unit of Progress Rail Services, "or people won't use it." In his view, corporate software should follow a consumer model. "I can look online at the stocks I own," he says, "and get a very clear snapshot. We want people to be able to look at scorecards and similar systems from anywhere and get that same quick take." The company's executives are now using BI software from SAS Institute and exploring the option of rolling it out to managers at nearly 100 facilities around the country.

Vendors are rushing to develop (or buy) visualization technologies to make that possible. But Crispin Read, chief marketing officer for Cartesis, cautions that the software must jibe with a company's culture. "Things like key performance indicators are a management issue, not a technology issue," he says. "If KPIs determine how people are measured, people will use the software that helps track KPIs. If there is no particular push, then users may exercise their discretion to ignore it."

"People don't beat down the door for new software," agrees McGowan of Southeast Corporate. "They have to be nudged and pushed. But today there is so much good software out there, with good price tags, that the efficiencies you gain are worth what it takes to provide it to as many people as you can."

Scott Leibs is a senior editor at CFO.

Top Players, for Now
Company Rank by market share Rank by growth*
Business Objects 1 7
SAS Institute 2 6
Cognos 3 10
Microsoft 4 1
Hyperion 5 9
Oracle 6 4
MicroStrategy 7 5
SAP 8 3
SPSS 9 8
Information Builders 10 2
* Based on gains in revenue, 2004–05
Source: IDC