CASE

Jim Honerkamp, CIO of Hillman Group, is proud of his new business intelligence (BI) system. Why not? It's much better than what came before. In the bad old days, executives looking for sales information, for example, had to ask one of Honerkamp's programmers to make a manual database query to pull the numbers from the company's legacy systems. The lag time made the charts "stale the minute they came out," according to Honerkamp, whose company is a $380 million manufacturer and distributor of engraving technologies and hardware, such as keys and signs.

With Hillman Group's new BI system, curious business executives can query the system themselves and get instant answers about such critical questions as the number of unfilled customer orders, which is tracked by the system in real time. There's just one problem: The new system hasn't made the business better-at least not yet-only better informed. That's generally the problem with BI, the umbrella term that refers to a variety of software applications used to analyze an organization's raw data (e.g., sales transactions) and extract useful insights from them. Most CIOs still think of it as a reporting and decision support tool. Although the tools haven't changed much recently, there is a small revolution going on in the ways BI tools are being deployed by some CIOs. Done right, BI projects can transform business processes-and the businesses that depend on those processes-into lean, mean machines. It isn't easy to take BI to the next level; it requires a change in thinking about the value of information inside organizations from the CEO down. Information is power, and some people don't like to share it. Yet sharing is vital to this new vision of BI because everyone involved in the process must have full access to information to be able to change the ways that they work. The other major impediment to using BI to transform business processes is that most companies don't understand their business processes well enough to determine how to improve them. Companies also need to be careful about the processes they choose. If the process does not have a direct impact on revenue, or the business isn't behind standardizing the process across the company, the entire BI effort could disintegrate. Companies need to understand all the activities that make up a particular business process, how information and data flow across various processes, how data are passed between business users, and how people use it to execute their particular part of the process. They need to understand all this before they start a BI project-if they hope to improve how people do their jobs. The new, greater scope of these BI projects gives CIOs a strong justification for working with the business to study processes and determine how these tools and the insights they provide can support and improve them.

Companies that use BI to uncover flawed business processes are in a much better position to successfully compete than those companies that use BI merely to monitor what's happening. Indeed, CIOs who don't use BI to transform business operations put their companies at a disadvantage. For CIOs who have carried out this difficult strategy successfully, there is no looking back. Avnet, a computer systems, component, and embedded subsystems manufacturer, took the new process-oriented BI strategy directly to the processes that matter most: selling and serving customers. The company has put together a system from three BI vendors-Informatica, Business Objects, and Info Burst-to generate reports on orders, shipment schedules, and dates by which Avnet will no longer manufacture certain products. Reports, however, were just the beginning. To transform the sales and customer service processes, CIO Steve Phillips rolled out the system to 2,000 salespeople so that they could actively incorporate that information into their day-to-day workflows and interactions with customers. Employees use the information to modify their individual and teamwork practices, which leads to improved performance among the sales teams. When sales executives see a big difference in performance from one team to another, they work to bring the laggard teams up to the level of the leaders.

"We try to identify, using our reporting tools, where best practices exist inside our work teams and then extend those best practices across the company," says Phillips. One of those best practices is to alert customers if a product they have purchased in the past is about to be discontinued. Salespeople can ensure that customers have ordered enough for all of their future needs or identify a new component to replace the one that's being phased out. Those kinds of conversations boost sales and convince customers that Avnet's salespeople are looking out for their needs and interests. It helps that Avnet's sales team is flexible and willing to adapt to the information. "Because our sales team is so flexible, they'll take this information from BI reports and change processes when they see a benefit to it," says Phillips. Sometimes, they don't even realize they are changing the ways they work-a kind of organic reengineering. Indeed, salespeople benefit so directly from better information and have such a big impact on revenue that they can be the best advocates for transformative BI in the company. Yet this kind of effortless link between information and processes doesn't happen by magic. Phillips says his company has been able to use BI effectively because IT and business users have worked closely and steadily. "We needed to know how things really happen day to day, over and above the documented processes so that we could anticipate some of the business's information needs as we built out the warehouse," says Phillips. Now that the BI system matches up with the way the company conducts its business, improving those processes and sharing the improvements are that much easier. "This is not just about reporting," says Phillips. "It's about using BI to make us smarter."

Quaker Chemical used its BI system to change completely the way it manages accounts receivable. In the past, the process of keeping track of whether customers paid their bills, and if they paid them on time, was primarily the purview of employees in the accounting department. Collection managers used the company's accounting system to identify which accounts were overdue, but they had limited information about the details of overdue balances. As a result, they had visibility only into glaring payment problems-customers who hadn't paid their bills at all in 60 days or more-and couldn't proactively identify which customers were at risk for not paying in full. Occasionally, they asked a sales manager to get involved, but the whole process for identifying which customers weren't paying and why they weren't paying and putting salespeople on the case was ad hoc. To improve accounts receivable, Quaker Chemical decided in early 2005 that salespeople needed to play a larger, more formal role in the collections process. After all, they were the ones who had the primary relationship with the customers and had opportunities to speak with them more often, more proactively, and more sympathetically about their outstanding payments.

To get the salespeople involved, the IT department created a data mart that extracted accounts receivable information from transaction systems: It analyzed historical payments and historical balances by customer and by transaction and then loaded it into the data warehouse. By using its BI tools from SAS to analyze factors such as the amount of time it took Quaker Chemical to collect payment from a customer on a given invoice, as well as the number of times a customer paid part, but not all, of what he or she owed, the company was able to identify which customers were consistently paying late and which customers weren't paying at all. The IT department programmed the data warehouse to run reports automatically on which customers still owed money to Quaker Chemical. The system would then send those reports directly to the sales manager in charge of those accounts several times a month so that they could follow up with those customers. Collections managers no longer have to keep tabs on this information manually. Quaker CIO Irving Tyler says this business process change was successful in part because IT was careful to deliver only the most specific, relevant information in these reports to salespeople. "If you don't focus the information and deliver it intelligently, people won't understand how to incorporate it into their workflows," says Tyler. This kind of dramatic change in process needs to be linked to the overall business strategy, according to Tyler. "Information doesn't necessarily change anything. You have to have a strategy to drive any change," he says. Avnet and Quaker Chemical demonstrate that BI is about more than decision support. As a result of improvements in the technology and the way CIOs are implementing it, BI now has the potential to transform organizations. CIOs like Avnet's Phillips and Quaker Chemical's Tyler who successfully use BI to improve business processes contribute to their organizations in more far-reaching ways than by implementing basic reporting tools. "Our BI system provides information that helps us seek out greater efficiency," says Avnet's Phillips.

CASE STUDY QUESTIONS

1. What are the business benefits of BI deployments such as those implemented by Avnet and Quaker Chemical? What roles do data and business processes play in achieving those benefits?

2. What are the main challenges to the change of mindset required to extend BI tools beyond mere reporting? What can companies do to overcome them? Use examples from the case to illustrate your answer.

3. Both Avnet and Quaker Chemical implemented systems and processes that affect the practices of their salespeople. In which ways did the latter benefit from these new implementations? How important was their buy-in to the success of these projects? Discuss alternative strategies for companies to foster adoption of new systems like these.