• 04/05/2011
    11:44 PM
  • Network Computing
  • News
  • Connect Directly
  • Rating: 
    0 votes
    Vote up!
    Vote down!

An SMB Masters Business Intelligence

Tales of data cleanup and normalization, massive integration and major workflow revisions in decision-making make it difficult for small and midsize businesses to compete against larger concerns when it comes to developing business intelligence (BI) for their markets and their operations. But with the right kind of support from upper management, an appropriately trained IT staff and a vision, SMBs can attain great results.

"When we looked back, we had started at a point where we had no Web services and no system integration, and we initially developed our own workflow management system called Project Center as a main integration point," says Worth. "It's a J2EE Web application that we still use and continue to develop. As time went on, however, there were also other systems like accounts receivable, accounts payable, payroll and accounting, and a future CRM [customer relationship management] system that needed to be integrated. We recognized that Project Center alone would not be sustainable as an integration platform for these other systems, and that a data warehouse model was the correct solution."

The logical next step was to build a data warehouse. "We laid out the structure of the warehouse and the reports," says Worth. "This included reports in areas like work in process, where managers could have complete visibility of everything rolling up into a project. We had tried to do this in the past with spreadsheets, but now we wanted an event-driven system where workflows were linked to work in process and managers could see real-time costs as they occurred."

To achieve the level of integration that would facilitate this scenario, TMG used Attunity's Stream EDC software for real-time data integration and event capture. "We have a heterogeneous computing environment that is a blend of Microsoft, Linux and Sun Solaris in a VMware cloud with an open source database," says Worth. "The Attunity event processor filters data and then loads the information that has changed into the data warehouse from these systems. From here, we have constructed key metrics like sales, earned revenues, margins and project backlogs into dashboards that give managers real-time information so they can make changes if needed."

Worth says this real-time visibility gives TMG the ability to keep sales work backlogs full, while maximizing the profit margins of projects that are in process. "We approached system implementation carefully, beginning with a beta project that was used by a couple of our project managers and that is now used company-wide. Feedback has been very positive at all levels of management, especially at the lower management levels, where managers never had access to this kind of information before. It has expanded the leverage potential of our business intelligence."

What were the major technical challenges of the BI project? "Data quality was No. 1," says Worth. "When we did the initial data loads into the warehouse, we spent additional months cleaning it up. In some cases, data was missing. In others, it needed to be rationalized. The second challenge was trying to align the construction project managers and the financial staff so they could agree on a single 'version of truth' that everyone would work from."

We welcome your comments on this topic on our social media channels, or [contact us directly] with questions about the site.

Log in or Register to post comments