Organizations everywhere are facing unrelenting pressure to attract and retain customers in an increasingly volatile and fiercely contested marketplace. Together, these factors are placing heavy demands on operational decision making. Under these conditions, business leaders must not only make the right decisions, they must also make them at the right time. Finding ways of bringing together and making sense of the vast amounts of data flowing within and across the extended enterprise is becoming a key business success factor.
CEOs no longer ask, "How well did we do last quarter?" Instead, they ask, "How well are we doing right now?" Unfortunately, the best that many organizations can do today is to answer the question, "How did we do yesterday?" This situation is similar to asking CEOs to drive their cars by looking in the rearview mirror rather than by looking through the front windshield. Yet, many organizations are forced to work in this way because their systems are unable to provide a real-time picture of how well a business is performing.
Taming The Data Deluge
Organizations seeking to drive latency from their operational decision making face a number of issues. Existing enterprise systems have evolved to the point where they can rapidly generate vast quantities of operational data in real time, but not digest it. Additionally, because there is no obvious barrier to the amount of data that an organization can collect about its operations, customers, suppliers and the market in general, the problem of detecting relevant or related information in time becomes increasingly more challenging especially if the data is highly volatile and moving at a high velocity.
Although the advantages of dealing with business events as they occur may be clear, this mode of working has a huge impact on an organization's operational systems and processes. Many of today's organizations rely on a "snapshot in time" approach to extracting their actionable intelligence from their operational data. Unfortunately, this approach does not allow organizations to continuously evaluate and respond to rapidly changing business conditions in a scalable way.
Businesses can achieve a true up-to-the-moment view in which:
The information gleaned is actually current enough to be useful in managing and executing business processes,
Efficiency is optimized by choosing among the best options available given the circumstances at the time, and
The organization is able to respond to its best customers.
However, meeting these objectives and realizing a timely view of business operations requires a new approach to integrating, analyzing and responding to business events as they occur in time, on time and in real time. With this approach, organizations can make sense of this veritable deluge of data in time to stay competitive and, ultimately, pull ahead in their market.
Software For Real-Time Business
Moving to an information architecture that is able to support real-time business is a key emerging requirement for many organizations. The ability to have CRM, SCM or ERP-based processes operate and respond to business events in real time is finally being recognized as a source of real competitive advantage.
A key requirement of any software infrastructure capable of supporting real time business must be to provide an accurate picture of the sum of all of an organization's moving parts. In other words, the appropriate software must have the ability to continuously correlate the actual performance of every business process against key performance indicators derived from organizational objectives. At the same time, a balance must be maintained between resource utilization and revenue and cost targets. Moreover, this needs to be done at the appropriate level of granularity.
Gartner, Inc. refers to this type of capability as business activity monitoring (BAM). Aberdeen Group describes similar concepts as extreme analytic frameworks (EAFs); while other analysts, to emphasize the importance of actively managing as well as monitoring business processes, refer to this capability under the general category of real-time enterprise performance management.
Figure 1: This figure illustrates how patterns across multiple information sources involving data information correlation and temporal constraints can be used to influence program trading strategies in real time.
Certain industry sectors are leading the way in realizing the benefits of doing business in real time.
For example, in financial markets, straight-through-processing initiatives are well underway to reduce transaction settlement times by squeezing latency out of the operational process. Although this large-scale initiative is complex, the financial industry as a whole clearly understands the benefit of moving toward a more real-time way of transacting business.
Businesses looking to implement these types of systems must pay careful attention to their infrastructure software's real-time operational capability and characteristics. Some of the key capabilities include:
- Ability to analyze data that may be changing in real time against an arbitrarily large and complex set of criteria;
- Ensuring the integrity of real-time data being used to form the basis of decisions and coping with corrections to that data after the fact;
- Capacity to correlate any or all of the continuously changing data variables associated with every concurrently executing business process that is relevant, including the capability to create streams of derived data in real time for use in the same analytic processes;
- Handling time explicitly as another dimension when performing queries, including the ability to refine queries dynamically over time based on analytical output; and
- High-performance notification to systems or people, enabling informed decisions to be made rapidly.
Real-Time Business Applications
Financial markets and telecommunications are two industries that already depend on executing real-time business processes. The following examples illustrate real-world scenarios of real-time business at work.
Example 1: Electronic Trading Decision Support
Companies in the financial services industry have always understood the value of time as it relates to gaining competitive advantage. Making use of the most up-to-date information available about the market is essential for profitable trading operations and managing risk.
Financial professionals in this environment are often surrounded by a number of display screens that they use to monitor constantly changing interrelated market data. Typically, these displays consist of the latest information available about commodities and financial instruments being traded across the various markets. Alongside this information will be any relevant historical data and the latest financial news information. Traders understand that a subtle change in any single factor may result in a price change of a related commodity or instrument which, in turn, may significantly impact their trading position. Therefore, in order to maintain a profitable trading position and limit any downside risk, traders must be acutely and continuously aware of how any single change in the market may affect them.
As the amount and volatility of real-time market data continues to grow, it becomes increasingly important to provide automated decision support to trading professionals. Spotting and tracking interrelated market data and correlating this against trading strategies requires a step- change in the sophistication of trading technology.
Example 2: Fraud Detection
Another cross-industry area of financial concern that can immediately benefit from instant insight into business operations is in preventing, detecting and limiting exposure to fraud. As networks become more global and service expectations become more immediate, the risk of financial loss through fraud becomes greater unless fraud management becomes a real-time process.
From the perspective of financial services companies, it is integral that ATM and credit card networks are continually correlating actual patterns of usage against a measure of the average, thereby giving early warning of potentially fraudulent use. As far as the telecom market, mobile network providers must also monitor for abnormal conditions. For example, if separate streams of call data records (CDRs) show the same cellular user making a call in London and Paris within minutes of one another, a provider would be alerted of this event in real time and could take action accordingly.
Example 3: Business Dashboards
A growing requirement in the operational business environment is what is sometimes termed a "business dashboard." This is closely related to the requirement for enterprise portals, but extends the concept by adding the ability to monitor and manage critical business indicators in real time. While this requirement is cross-industry, a simple example from the telecommunications industry serves to illustrate the point.
Using a customer satisfaction "dashboard" consisting of a simple green/yellow/red traffic light system, a senior operations manager responsible for new services provisioning can be alerted if any number of the many concurrently running provisioning processes have stalled under specific circumstances. These circumstances would be specified as business criteria.
For example, the operations manager may specify that he should only be notified if the provisioning processes have been stalled for more than an hour and if they were initiated for organizations recognized as highly profitable customers. Once alerted, the dashboard would present the operations manager with a consolidated view of the offending processes in priority order, including probable cause and cure (e.g., a lack of circuits in a given geographical area and an option to reassign).
In order to meet this kind of requirement, the system must continuously monitor the progress of every provisioning process against key performance criteria and cross-correlate under-performing processes against information held in other systems, such as a CRM system. The scalability requirement of such an analytic system would be substantial.
Organizations are beginning to recognize that a critical component of their business strategy must be time-based competition. If they are to gain and sustain a competitive advantage in the market, they must have faster reaction times than their competitors and base their business decisions on the latest available real-time information.
A key barrier in moving to this way of doing business is coping with the resulting real-time data deluge. Existing technologies based around processing data in a static fashion may prove inadequate as companies strive to squeeze latency out of their business processes.
To eliminate this key barrier, organizations should research and implement unique BI infrastructure software that removes the "time to index" overhead and allows new queries to be added dynamically with virtually no overhead. Companies that keep these principles in mind as they move toward the real-time enterprise are well poised to realize the vision of doing business in real time.
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