In order to function effectively in today's business environment and maintain competitiveness and profitability, companies must have visibility at all times into activities at all levels of the business. They must also be able to respond quickly to business events as they occur.

The need for real-time visibility and immediate response capabilities has given rise to business activity monitoring (BAM), a process by which key operational business events are monitored for changes or trends indicating opportunities or problems, and enabling business managers to take corrective action.

While business intelligence has come a long way in terms of providing insight into various aspects of a business, business intelligence today still presents mostly a historical view of what "has happened" rather than "what is happening" in the business.

As is the case with many solutions whose time has come, the concept of BAM is rapidly gaining steam. A May 2003 Gartner Group report indicated that, "By 2004, in enterprises where faster reaction is key to operational efficiency, BAM will be one of the top four initiatives driving IT [information technology] investment and strategy."

What BAM Is

A term coined by Gartner, BAM is a way to gain meaningful, instant visibility into critical business operations. It works by capturing events from operational systems. Depending on the business and the BAM application, events can range from scanning bar codes to receiving orders or taking customer service calls ­– and then correlating these events with relevant contextual data. A record of a customer's previous sales activity is an example of contextual data.

To be effective, BAM should permit deep visibility into operations, but BAM should also perform the event-context correlation extremely quickly. For example, a BAM system in a customer service environment might quickly alert a manager to a top customer's call. Such fast action makes it possible to quickly resolve any issues –­ and keep the customer's business. By contrast, a non-BAM (or batch) system requires extra, time-consuming steps to perform the correlation, and the manager might not be alerted to the issue until hours later, when the customer has been lost.

BAM systems won't replace data warehouses. By definition, a data warehouse is a place to accumulate or aggregate data for subsequent analysis. Typically, users look to data warehouses for historical analysis and planning functions within a business. As a result, the data warehouse generally contains fixed data models and permanent data storage, available for user-specified analytical "roll-ups." However, for operational functions (e.g., supply chain, customer interaction, logistics) within a company, the need for up-to-date information and business rules governing event-specific actions becomes very important, requiring a BAM solution that is designed to serve operational needs.

The BAM system must capture and process events with minimum latency. BAM systems are, therefore, tightly integrated with their operational sources, and they are optimized for event processing and for correlating event information with historical or other contextual information.

In fact, the two are highly complementary: the data warehouse serves as a historical source for the BAM system, and the BAM system puts the data warehouse's historical knowledge into an event-based business intelligence framework.

What BAM Isn't

Some organizations attempt to solve the "real-time visibility" problem by speeding up the loading of a data warehouse. Also referred to as real-time data warehousing, this approach uses message broker or EAI (enterprise application integration) software to load data, directly or via extract, transform and load (ETL) software, into the data warehouse.

While real-time data warehousing does provide near real-time performance, it does not enable the event-driven decision making that BAM is known for. True BAM systems, or what Gartner calls "pure-play" BAM systems, are characterized by architectural approaches that are fundamentally different from the EAI, business performance management (BPM) and ETL systems of the last decade.

Figure 1: Comparison of Real-Time Data Warehousing and BAM

BAM in Action

BAM offers unique benefits to those businesses and industries for which real-time operational information is critical.

Financial Services: BAM can help with portfolio management, programmatic trading, fraud detection, risk management/compliance, real-time or interactive customer marketing, and Patriot Act compliance. For example, a BAM system might assist with Patriot Act compliance when it is used to dynamically model scenarios where transaction patterns show suspicious activity that could indicate illegal activity.

Manufacturing: BAM can help with order management and shipping forecasts, sales forecasting/pipeline monitoring, product recalls and quality control. In quality control, for instance, the BAM system can monitor data feeds from manufacturing data collection systems. Based on these feeds, the BAM system can generate alerts as necessary and calculate trends by using time-series computations.

Retail: BAM can help with real- time inventory analysis, real-time marketing and promotions, and product recalls. In inventory analysis, for instance, BAM would help match ongoing sales trends with in-store and in-shipment inventory to alert management to inventory anomalies, and to help managers adjust to seasonal or sudden demand spikes.

In other industries, BAM can help solve a range of problems in several areas by providing real-time dashboards that provide continuously updated key performance indicators and exception reports for events that require action.

BAM: Architectural Challenges

While the benefits –­ and the basic workings –­ of BAM are relatively easy to understand, the software architectures required to drive BAM are more complex, and thus more challenging. BAM systems must be able to:

  • Capture event-driven real-time data from a wide variety of sources –­ from message queues, Web services and databases to context-oriented sources such as data warehouses and traditional databases.
  • Calculate temporal information by collating time-series data sets from event streams –­ such information can then be used to trigger time-based thresholds.
  • Perform dynamic modeling by integrating event and contextual information on the fly, and produce a stream of analytical models –­ models that can automatically update themselves based on input from subsequent event or contextual information.
  • Execute business rules to set thresholds according to key performance indicators and other business-specific triggers.
  • Encourage business user access by providing ergonomic, user- friendly interfaces for specifying business rules and dashboards for viewing continuously updated metrics.

Beyond these, the greatest challenge lies in the need for real-time event-context correlation and high- quality analytical output. In order to create higher-quality output – ­such as reports that include time-based dynamics ­– the BAM system must add extra cycles and extra intelligence to the core event-context correlation processing. This challenges the BAM architecture's ability to combine high- speed streaming of events with the dynamic modeling needed for event-context correlation.
Data quality can also pose a challenge. According to a recent report by Gartner: "Incorrect data in events can cause false alarms or missed problems that should have been caught ... Enterprises that do not address the data and event quality issues when deploying BAM systems will risk failure and face a difficult time achieving a return on the investment." Data quality for context data is easily addressed by using a reliable source of cleansed data (e.g., a data warehouse where data has been cleansed before loading). For event data, BAM products will provide basic capabilities to cleanse the event data using translations, decoding and matching keys against context data to ensure valid events.

Today, business intelligence is more than a vision of the world's largest enterprises –­ it is a fundamental, critically important strategic resource. Businesses succeed or fail on the strength and the speed of their business intelligence because the world of business moves faster –­ and is less forgiving ­– than ever before.

No longer can businesses, large or small, ignore the need for capturing and analyzing information in real time. And no longer do enterprises have to piece together existing business intelligence architectures in order to gain the advantages of real-time analysis. BAM is here, able to serve critical information at the touch of a key and able to get working without a mammoth corporate financial commitment.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access