In my August column, I began discussing some changes that have been creeping into information management because they are practical yet lack acceptance and codification. Specifically, I spoke of the role of the modern data warehouse and the sharing of its analytic function with the operational environment through the mechanisms of master data management (MDM), enterprise information integration (EII) and other operational forms of business intelligence (BI).

Keep in mind that neither EII nor anything else is a substitute for data warehousing. EII will have limited complex transformation and integration capabilities. Frequent federated-style processing with single queries integrating data across a large number of data sources can be restricted or have unacceptable performance. Likewise, transaction processing - when integrity constraints across data sources are complex - could keep EII off some selective mission-critical paths for a few users.

The data warehouse is the place for long-term historical storage of clean information. Regulatory and compliance requirements for corporations to keep historical data continue to increase. The data warehouse will fill the numerous gaps in the operational strategies mentioned herein. It will need to perform when called upon for queries, mining, strategic dashboarding and reporting. However, concurrency requirements may not continue at as hectic of a pace as the analytics are spread out. Programmed and highly facilitated decision- making will compress rather than expand the need for data for interactive decision-making. Incidentally, this world lends itself very well to the data warehouse appliance, which is gaining traction in the market.

However, there are many environments and pockets where EII - or other methods of integration among operational components - could automate some of the decisioning that now only occurs post-operationally in the physically integrated environment. The lack of physical integration does not have to mean the inability to use integrated information for a purpose. As well, the operations to be performed on integrated information do not always require manual intervention.

Operational BI enables responsiveness and true knowledge. Some of the most critical decisions are made in the operations of the business, and decisions should be made as soon as the opportunity exists. In some cases, this is programmatic change, and in other cases, it means facilitating the information flow to decision-makers. Sometimes both are needed.

Real-time decisions can be automated where possible. Dr. Richard Hackathorn has illustrated the time-value curve demonstrating the various latencies that can occur from event to action. Knowingly or unknowingly, these latencies have been accepted and programmed into information management environments. With each latency, value is obviously lost. There's latency in capturing the data into a usable form, in preparing the data for analysis and, finally, in delivering the information to cause a business result. Any latency removal can be worth it.

In terms of information delivery, while reporting is usually requested, it can often be avoided with some investigative effort. I'm not talking about those dormant reports that nobody looks at - although those are fair game. I mean understanding the actions produced by the analysis of report output and re-engineering those actions back into the operations environment. In database management system terms, this is the object type of trigger often used to generate the action. Other forms of transactional monitoring, operational workflow and rules engines are equally important.

Likewise, dashboarding could be subject to review, but as an already summarized form of corporate data more closely tied to exception management and specific business action, these are expected to remain prominent. Dashboards are produced at various levels, from strategic to operational, and the underlying data is generated from extract, transform and load (ETL)/data warehouse or EII/operational data. The line between operations and BI will not be blurred, but removed. I have not been in an environment that was not using operational BI. It is just seldom acknowledged as such. Greater success is possible where selective real-time capabilities and event- and trigger-driven BI are added technically, and a willingness to rethink the placement of BI in the architecture occurs.

Operational BI makes BI accede to its higher calling of actually doing business, rather than reporting on it. As an information consultant, I continually find myself architecting and managing programs that span both the operational environment and the data warehouse environment - because both places are where information can have high impact on company goals.

In many shops, IT will need to become more integrated with the business to correctly interpret the logic used and to update systems appropriately. Resources, processes and corporate priorities will shift. These are steps that should be taken immediately to facilitate the needed growth in information management environments.

To summarize, as businesses evolve, so does BI. It is high volume and real time, and decision cycles are decreasing. Information's importance is sinking into corporate culture, and information is often cited as a main company asset. Consider adding MDM, an EII server and operational BI to the operational environment to take advantage of earlier-stage information availability. 

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