Technology changes consistently disrupt key business processes. Embracing change management has become a much appreciated trend in order to prepare the organization for radical changes in sales, service and marketing business processes. By arming individuals with the training, understanding, documentation and incentives to drive adoption of new processes, change management is the icing on the cake to a successful technology implementation.

However, as we have refined mitigating our nontechnical challenges such as organizational change, an old technical nemesis has reared its ugly head. The ability to expose key customer analytics during the business process itself, sometimes called operational customer intelligence, rather than in a different application will drive competitive advantage. But just when most customer intelligence administrators have stopped losing sleep over performance, the synchronization challenges involved with operational customer intelligence resurrects the issue. Solving these issues is the mandate of many customer intelligence administrators and architects and some techniques are emerging.

Closing the distance between the application and the analytics cannot hurt. Many organizations start by having their applications reach through to their data warehouses or operational data store. Though this is the simplest approach and avoids data duplication, several issues quickly arise:

  • Star schema designs are not optimized for transaction processing;
  • Competition with decision support users causes volatility in performance; and
  • Lack of change control procedures cause disconnects as the data warehouse and application enhance their applications.

In reaction, information from the decision support systems start migrating back to operational systems, closing the decision support loop. This technique places the information within the application itself. However, as demand for more information during the course of customer-related business processes begins making the closed-loop batch extract, transform and load (ETL) process unwieldy, it starts to feel as though the whole data warehouse is being moved over to customer service.
Embedding query tools into portals and applications prevents application architects from needing to design a query or reporting tool from scratch. Business Objects, Cognos, Oracle and MicroStrategy all have portal products that allow their engines to be embedded in Web applications.

However, these engines are initially meant for decision support users. Although these products all have references of hundreds or even thousands of users, few have been put to the test in the rapid-fire operational customer intelligence game. In this game, thousands of users may be requesting advanced analytics, possibly in a customer-facing situation where a time lag is not an option.

Architects are starting to bypass these query engines and substitute the query engine for raw SQL and Java. Though this architecture is flexible and possibly faster, application developers now need to spend time with aesthetics. Whereas reusing a business intelligence (BI) tools infrastructure means changing a field on a report, in this scenario Java jockeys are changing the application that needs to be regression tested - not the nimble company the organization was trying to portray when it started down the customer intelligence path.

These days, it is all about ROI, and innovative ideas are unnecessarily bogged down by little things such as technical feasibility and response time. In this light, the BI appliance market starts to make sense to me. For years, the BI community has been trying to invent the right mousetrap to speed up BI operations. RedBrick was a BI-optimized architecture but just did not have the sales and marketing muscle to withstand the onslaught from the big relational database management systems. Teradata is software and hardware all in one and has traditionally been thought of as the high-end performer for the largest data warehouse organizations. Finding a Teradata guru is like finding a needle in a haystack.

Netezza's database appliance boasts big-time performance with little deployment effort and low maintenance. Clients provide ringing endorsements of the architecture and allow organizations to focus on application development instead of performance tuning.

Maybe more interesting is Celequest. Providing ETL, dashboard functionality, high performance, high scalability and portal/application embedding technology, the architecture seems perfect for operational customer intelligence. Delivered as an appliance with a development environment, the blazing performance of the appliance allows organizations to focus on functionality instead of infrastructure.

Getting into the operational customer intelligence game takes an understanding of which business processes can most effectively be impacted by specific types of analytics. But just when you find the right process and the right solution, architectural feasibility and development become a big red stop sign. New BI appliances help us focus on solving the business problem and not the performance problem. Whether these development environments can stand up to the complexity of operational customer intelligence remains to be seen, but it could be a cost-effective alternative to traditional architectures. 

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