Management processes and systems can be standardized only if they share data and performance indicators. Most organizations are stymied in their standardization efforts because they have trouble agreeing on definitions for the data that underlies the performance indicators. For instance, the term "employees" can be defined in several ways (e.g., only people who have full-time contracts, temporary workers, part-timers, etc.), and the term "revenue" can be interpreted in many ways. Thus, even a seemingly straightforward metric such as revenue per employee may be difficult to establish. Such a lack of standardization in data makes internal benchmarking impossible.

Obviously, another data-management requirement is that the information stored in operational systems must be correct to begin with. Gartner research has shown that the quality of 25 percent of all data used for management information is poor enough to cause problems. It is incomplete, contains factual errors, or is just unfit for the purpose for which it's being used. After the initial -- often painful -- data-cleanup phase of installing a CPM data warehouse, continuous training and monitoring are required to keep corporate information error-free and usable. High-performance organizations have an excellent grasp on information management. Organizations seeking to emulate them should concentrate on three activities: metadata and master data management, data integration, and data quality.

Agility Is Key

An abundance of research suggests that most organizations fail at executing strategies designed to improve their position in the market because the external environment changes faster than strategies can be devised. High-performance organizations achieve a high level of agility so that they can identify change and respond optimally -- or, even better, set the pace for change within their industry.

There are four primary ways to create agility in an organization. One is by centralizing processes, data, and systems companywide. This approach is usually highly IT-centric, but it ensures that changes need to be executed only once. This speeds up change processes and eliminates the chances for error. A second method for improving agility is through smart sourcing. Standardizing as many product components as possible and using subcontractors to produce and deliver those components can lead to a dramatic decrease in new-product-development time, so companies can respond quickly to market trends. A third model for improving agility is mastering the channel, as Wal-Mart is well-known for doing. The concept of "just-in-time inventory" is crucial here. If the organization monitors the complete value chain, it can react instantly to changing buying patterns. Products that sell faster than others can be restocked immediately. This approach also reduces waste of resources on excess capacity. Finally, project-based management can improve an organization's agility. If corporate functions such as HR can be fluidly deployed as needed by strategic initiatives, rather than being housed within rigid departmental structures, teams can be formed and dissolved more rapidly to pounce on opportunities or respond to threats.

Which of these agility strategies works best for a given company depends on that organization's business model. Over the past 30 years, Oracle has successfully navigated several major transformations. CEO Larry Ellison is responsible for the creation and execution of the company's strategy, and his strong personal leadership sets the direction for the company. Oracle can respond quickly to change because its processes and systems are extremely standardized. Once a change is put in place, it is active worldwide. Yet although Oracle is an agile company, not every agility strategy fits its culture. Centralization is the model that works for Oracle because of its structure and leadership

In other organizations, centralization leads to inertia, as the centralized functions become a bottleneck, building up a backlog of changes that are never implemented. Business units relying on centralized systems sometimes have to compromise in terms of functionality, so no one in the organization has the functionality they need for specific processes. And in other enterprises, centralized processes require so many exceptions that the supporting systems' complexity hinders, rather than helps, the organization's ability to change.