With increased (read, "regulatory oversight") importance being placed on the accuracy of corporate information, it is more important than ever to remove the latency from corporate information and have real-time data for analysis and decision making.
Enterprise performance dashboards are indispensable tools within the business intelligence framework in collecting, presenting and enabling analysis of critical organizational information. How do you make sure that this indispensable tool is built for speed and accuracy? It all starts with a real-time operational data store (ODS) that pulls together applicable transactional and operational data from relevant enterprise systems and enables analysis using consistent business rules and data definitions.
Data warehouses and ODSs differ in many respects, but the most significant to this discussion are age and type of information contained and functionality within the decision-making environment.
Data warehouses provide data captured at a fixed point in time. Information from various sources is collected, cleansed and loaded into the data warehouse. The information is historical and aggregated. With warehouse data, you can run different analytics on the information, and you can perform data mining and other types of long and short-term analysis. However, the data is static a snapshot of a specific timeframe. There is very little information in most data warehouses that can be considered to be "real time."
On the contrary, ODSs contain transactional and operational data as well as aggregated, historical data. ODSs are like hybrids of typical transactional processing systems and data warehouses. ODSs typically have the ability to support high-volume data inserts, while simultaneously supporting update, query and delete functionality via the organization's executive information and transaction processing systems. ODSs also enable all this functionality to be carried out in real time.
Data warehouses hold very high-quality data that is great for strategic decision making, but it's not very fresh. Warehouse data is not sufficient for real-time decision making.
In contrast, the data contained in most ODSs built using powerful databases such as Oracle Corporation's Oracle9i, IBM's DB2 and NCR Corporation's Teradata is also of moderately high quality, but it's completely up to date which is absolutely critical for real-time analysis and operational and tactical decisions. Be aware, however, that even with these powerful databases, data quality problems can still arise in some analytical situations.
Data warehouses and ODSs are vital components of the overall real-time enterprise framework. However, for real-time analysis, ODSs are indispensable. While both enable users to recognize patterns in data, ODSs allow users to recognize patterns in real time. This real-time analysis capability facilitates quick and time-appropriate decision making.
For example, someone steals a credit card in Chicago and begins to use it over the Internet before the card is reported stolen. That could be disastrous for the issuing bank. A real-time ODS would activate a flag indicating a suspicious pattern. The bank's security analysts could then react immediately and freeze the card. Many headaches for both the issuer and cardholder could be avoided.
However, the key long-term benefit of a good ODS is its ability to provide intelligent, profitable interactions with every customer at every point of contact. Consider the case of a customer calling the local telecommunications service provider to request technical support with his/her DSL service for the third time in a week. The customer also happens to have local and long distance phone service, as well as cable TV service with the provider which makes that customer very valuable.
If the service provider has a real-time ODS in its business intelligence framework, the customer service representative (CSR) should be able to recognize that the customer has had previous problems, and perhaps recognize a pattern to the problem and solve it permanently. The CSR should also be able to see that the customer has multiple accounts, and thus work very hard to ensure that the very valuable customer is satisfied. It's a win/win situation. The provider keeps a valuable customer and the customer gets knowledgeable, personal customer service.
As stated previously, any business intelligence infrastructure needs a data warehouse. There is no substitute for the ability to recognize, interpret and make strategic decisions based on long-term data patterns and establish triggers in the ODS that may otherwise go unnoticed. However, contrary to what many data warehouse vendors would have you believe, warehouses are not panaceas to data analysis problems. Today, when the competition is a mouse-click away, it's absolutely critical to have real-time, tactical decision-making capability. That capability is best provided by a real-time ODS.
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