This is the third year I have been asked by DM Review to write an article on the state of the art of business activity monitoring (BAM). As I began to write this year's article, I struggled to come up with a theme that demonstrates the progress and changes in BAM since last year. On the surface, little has changed, and there is also a perception in some quarters that acceptance of BAM by the industry is waning. Even Gartner Research, the organization that created the BAM concept, admitted at a recent conference that BAM has not taken off as fast as expected. This state of affairs led to the theme for this article, Is BAM Alive and Well? I will review BAM and its business benefits, discuss whether BAM has achieved those benefits and then look at its likely future. As we will see, there is more activity going on in this area than may be apparent - it's just not always called BAM!
What is BAM?
BAM enables organizations to leverage business analytics to gain a real-time insight into daily business operations. This analytical insight helps business users quickly identify operational inefficiencies and predict potential business problems. BAM integrates business intelligence (BI) with business transaction (BTx) processing. It also brings together vendors and IT groups that hitherto have operated in separate IT domains. This can lead to heated debates about the right products to deploy and the use of different terms to describe the same technology concepts.
From a business perspective, it is irrelevant if we use the term BAM or not - it's the business benefits that a technology provides that matter. BAM is often viewed as being associated more with technology infrastructure than business needs. This is why some BAM vendors are now using terms such as operational dashboard builder to describe their products. A dashboard means more to a business manager than an acronym such as BAM. Let's first look, however, at the concepts and objectives behind BAM, and then come back to the terminology issue.
The Emergence of Process-Centric Applications
As I have already stated, BAM is concerned with integrating BI with BTx processing. The BTx applications run business operations, and the BI applications analyze those operations. This combination works particularly well when the applications involved employ a business process-centric view of company operations.
A business process defines the flow of activities that need to be performed to carry out a business operation such as ordering a product, raising a purchase order or processing an insurance claim. Because BTx applications support the main business operations of an organization, we usually associate business processes with BTx applications, rather than BI or collaborative applications. This is why the move toward the use of business process management has seen most growth in the BTx application area.
Business processes and their associated activity workflows, however, can also be used with BI applications. A BI performance management dashboard that displays analytics about business operations could, for example, present a workflow (sometimes called a guided procedure) to business managers to help them investigate specific key performance indicators (KPIs) in more detail. In this case, a business process is added to the BI application.
Another example of adding a process-centric perspective to BI could be when a BTx procurement process detects a delay in a supplier order. The BTx application could invoke a BI service to analyze the financial impact of the delay before sending a message to the procurement manager to notify him/her of the problem. In this case, the BTx process is extended by adding a BI service into the BTx process flow.
We can see from these two examples that a business process can be added to BI, and BI can be added to a business process. Both of these approaches are useful in a BAM environment.
The objective of a BAM project is to monitor, analyze and report on the performance of business operations, and for business users to act on this performance information to improve business operations. Let's examine each of these steps in turn.
Monitoring involves tracking and collecting information about a business operation. Monitoring is easier if the operation being tracked has been implemented in terms of the business activities required to carry out the operation. This granularity enables the monitoring task to track just those activities that are important from a performance perspective. Without this granularity, the monitoring task can only track the complete business operation. A business process and its underlying activities can be implemented using business process management software and an underlying service-oriented architecture (SOA) that defines each business activity as a separate callable service.
The analysis step of BAM processes the information collected by the monitoring step and creates a set of metrics documenting the performance of the monitored business activities. This analysis can be done synchronously and in-line as a part of the main business process, as in the procurement example previously mentioned, or it can be done outside of the main business process. In this latter case, the monitoring information can be routed to an asynchronous in-line process for analysis, or it may be stored in a message queue or persistent store for off-line analysis by an independent application. The method chosen will depend on how quickly the analysis has to be done and acted upon (i.e., on the timing requirements of the business application).
In-line analysis of monitoring information is supported by several business process management and application integration vendors such as IBM and TIBCO. It is often these vendors that still use BAM terminology. Off-line processing of monitoring information is supported by independent BAM vendors, such as Celequest, and a number of BI vendors. Here, vendors use terms such as operational business performance management and operational dashboard builder, rather than BAM. Operational business performance management and BAM are conceptually the same.
The reporting step of BAM involves displaying or delivering the results of the analysis step. This may be done via a portal, dashboard, e-mail, instant message, etc., depending on the preference of the user and how quickly the information must be acted upon. In some situations, a rules-driven alert may also be sent to highlight important information. As already discussed, a workflow or guided procedure could also added to the reporting step.
The main objective of BAM is to be able to monitor and analyze business operations and deliver information to business users so that they can react rapidly to business requirements and problems. Most of the decisions and actions taken using BAM are reactive in nature - the user acts after a business situation has occurred. Using predictive BI techniques in conjunction with BAM, however, makes it possible to detect patterns in business operations and predict business issues before they occur. An example here would be assessing the risk of giving someone a loan or determining that an insurance claim is fraudulent and should be handled manually.
The decision and action-taking step that follows the analysis and delivery of information from a BAM application is the most critical aspect of a BAM project. It is this step that enables business users to optimize business operations and align those operations with the goals of the company. This is why the output from a BAM application must be tied to a business goal. This goal may be a specific target (reducing call center action time, for example) generated by a balanced scorecard, planning or budgeting application. It may also be a more complex goal such as determining the actual cost of each business activity in a business process to provide a complete picture of the costs of doing business. In this case, the results can be used in conjunction with BI software such as activity-based management.
We can see then that BAM exploits the process- and event-driven nature of BTx processing and uses BI to analyze performance monitoring results and drive business process improvement. Other BI techniques such as predictive analysis, activity-based management and balanced scorecards help improve the effectiveness of BAM processing.
One big issue for BAM project developers is that BI products and applications are usually data-centric, rather than process-centric. Many companies have implemented BI business performance applications for producing data-driven metrics, but these applications often do a poor job of tying these metrics back to the business processes being measured. This tie-back is particularly important for operational business performance analysis. To support BAM effectively, BI must be made process-centric. As shown in Figure 1, this can be done by incorporating BI into operational BTx workflows using the in-line and off-line BAM approaches discussed earlier.
Figure 1: A Process-Centric BI and BAM Architecture
What Does All This Mean?
Returning to the theme of this article about BAM usage, terminology and futures, it is important to realize that BAM is just a piece of the technology puzzle required to support the management and optimization of daily business operations. Other key pieces of the puzzle include business process management (including a service-oriented architecture) and the business intelligence techniques outlined earlier in this article. Even BAM itself is often given different names - operational business performance management being an important one.
While there may be the perception in the market that organizations are not rushing to deploy BAM, there is no question that many companies are actively pursuing related technologies such as business process management and business performance management. Organizations are beginning to understand that they need to use these two technologies together to enable BI to become more process centric. Unfortunately, except for the application vendors, most BI vendors have not yet addressed this need. In the BI product area, the main thrust by vendors has been toward BI-driven operational business performance management, and many companies are beginning to gain major business benefits from this approach.
In conclusion, the underlying concepts behind BAM are alive and well and have much to offer organizations. It is debatable if BAM is the right term to use. Given that BI needs to move toward a process-centric architecture and given the confusion over the BPM acronym, perhaps business process and performance management, or BPPM, may be a better term to use than BAM.
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
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access