As vendors peddle the slickest new dashboards, scorecards, and planning applications to get your attention, they probably are glossing over the behind-the-scenes work that makes these tools perform reliably.

It's not just about a pretty interface anymore. Infrastructure is back in fashion. Companies have realized that a dashboard, planning, budgeting, or forecasting system needs to deliver trustworthy, integrated, and timely information that supports key performance indicators (KPIs) and serves as a foundation for business decisions. Companies need information strategies that are aligned to their business strategies. They're inextricably linked -- two sides of the same coin.

Enterprise information management (EIM) is the key to this success. EIM includes the strategy, practices, and technologies that allow organizations to effectively manage disparate data. The right approach to EIM will help drive individual and organizational performance, enable business intelligence (BI) standardization, and ensure relevant and timely information is delivered to business users in a way they understand. It also will enable both IT and business agility and adaptability; as data volumes and complexity grow, the only thing that is guaranteed is constant change. So whether it's Websphere, Netweaver, Fusion, or something completely different, a comprehensive EIM strategy will help organizations to focus not only on what's in the middle (infrastructure), but also on what's currently at the bottom (disparate data), and what's ultimately going to be most relevant at the top (information that enables people to make better decisions and do their jobs more effectively).

This task is complicated by the fact that today's IT organizations are struggling to deal with growing data volumes and complexity while the people they are trying to serve actually want less information more easily. The need for a pure-play BI and performance management platform that includes robust data integration capabilities has never been more critical. Without such a platform, companies will not be able to fully achieve this vital step of aligning information strategy and business strategy.

Three trends of more effective use of BI have emerged and require that companies adopt such a platform to reap optimal competitive benefit:

• Companies are moving toward closed-loop BI processes that allow them to model their businesses and set plans and goals that align resources, measure success, and adjust plans as business conditions change.

• Companies want to standardize on a single BI platform to reduce total cost of ownership and pull information from a centralized, certified, and trusted system. Deploying BI systems within departments has given the departments the ability to access, analyze, and share information. However, this has created uneven pockets of success across departments. Also, redundant BI applications have driven up the cost of hardware, support, training, and data reconciliation.

• Companies want to make BI more process-aware and real-time-enabled in order to empower people on the front lines to get information where they work, in the context of how they do their jobs. For example, processing a mortgage loan application requires a credit report, outstanding loan balances, payment history, and other information that helps the loan processor assess the risk of a potential customer. Providing this information right at the moment the loan processor needs it allows him or her to turn around the application faster, improving customer satisfaction, reducing the chance for error, and improving efficiency.

Consider, too, the need for solid architecture as it relates to achieving strategic goals. Imagine an information architecture (see exhibit 1 below) that fits so seamlessly with the people it's designed to serve that it's hard to tell where one stops and the other starts. No small task indeed. Whether you're considering a data warehouse, a series of data marts, an operational data store, or something completely different, without the right architectural foundation and design there can be no long-term BI and performance management success. That's why, when it comes to EIM, you should consider the following questions:

Middle-why?

If your goal is to become a high-performance organization, where people are connected to common metrics, goals, and strategies, you cannot underestimate the importance of the underlying data foundation needed to make this possible. Is your organization seeking to attract and retain skilled workers? Are you hoping to acquire new customers? Are you developing new products and processes in order to stay ahead of the competition? Is your company aiming to increase shareholder value while continuing to be able to adapt to rapidly changing market conditions? Do you have a plan to use IT to reduce costs and create more value? For most organizations, the answer to these questions is a resounding "Yes!" And the work done behind the scenes -- the work that end users never see -- will be the key to the success of these goals.

Take, for example, dashboard and scorecard applications, which provide visibility and accountability into the key metrics driving success of the company (see exhibit 2 below). They look great, but if people don't trust the information they're looking at, the systems are worthless. The reason dashboards are in high demand is their simplicity of design -- easy-to-understand indicators of how well the company is doing, in a quick snapshot. What people don't often think about is the amount of work going on behind the scenes from an information management perspective to make the content trustworthy, integrated, and timely.

• End users must trust the information. The key to the success of performance management systems is end user trust in their reports and dashboards. To help them build this trust, businesspeople want to know where the numbers come from. Can you answer questions about how the numbers were calculated and where they originated? Will it meet compliance requirements? More and more, IT is burdened with the task of answering these types of questions to help their business users gain trust in the information they receive. Having report-to-source-system data lineage allows users to actually see what systems the information comes from and what calculations were performed.

• Metrics must be integrated from multiple sources. To provide a broad view of the business, you need to pull KPIs from your finance, sales, inventory, HR, and other systems. Typical organizations have seven or more systems where they house information, even if they already have a "standard" ERP system. (Imagine how complicated this gets if companies grow through acquisition and inherit systems from other companies that they then have to integrate.) A comprehensive EIM strategy allows you to integrate data from disparate sources, provide agile data integration methods for physical (data warehouse) or virtual integration (sourcing information on the fly from transaction systems without replication into another physical data mart or data warehouse), and finally reconcile disparate data forconsistency -- in some cases transforming or cleansing data on the fly to bring back consistent results.

• Information must be timely to be actionable. As performance management moves from being a strategic planning practice useful only to the upper echelons of a company to an operationally focused system usable by all, the need for timely information increases. Getting an alert tomorrow that you failed to answer 100 customer calls in an acceptable time frame doesn't help customer satisfaction today. A performance management system needs to provide information at the right time to meet the needs of each business user. A comprehensive EIM strategy means data can be accessed directly from systems without waiting for batch data loads that can slow and complicate the day-to-day management of the business.

Middle-what?

The key to the middle is to remain neutral from -- but be tightly linked to --the transactional systems used to run the business. Whether it's a database or a data warehouse, a data mart or a data store, there are many tools available to help you physically or virtually move data and make it relevant to business people. For example, extraction, transformation, and loading (ETL) tools are used to move large volumes of data in batch from multiple sources to targets. And enterprise information integration (EII) tools are gaining in popularity for enabling federated querying capabilities across systems, which enables more of the timely information mentioned earlier, accessed in real-time. Either way, a clear data warehousing strategy helps to solve data centralization, reconciliation, and efficiency issues.

But the challenge of data quality must not be underestimated. Without data quality there is no end user trust in the numbers. Many organizations are creating the role of the data steward to help deliver what essentially becomes a "data quality firewall." The mission, of course, is to ensure that data is clean, accurate, and valid. Look to find ways to deliver data lineage capabilities whereby businesspeople can trace the data back to its source and determine not only where it came from but also how it was transformed and when it was last updated. Also look for ways to deliver impact analysis to administrators whereby they can determine the impact of source changes to the reports and dashboards that are accessing the data.

While there are examples of master data management (MDM) tools, MDM is primarily a practice that uses technologies focused on ETL, EII, data semantics, and metadata management. MDM provides a blanket layer of standard definitions, or semantics, about the business (customer, product, people, or sales) that combines disparate metadata together from different tools and data sources. Some organizations are implementing MDM for data quality purposes, helping them to transform data into to a common standard.

1 2 Next