Data Quality is a BackOffice Associates Specialty
April 6, 2010 Alex Woodie
You may have heard that old computer axiom “garbage in, garbage out” so many times that it’s become just another meaningless catch phrase. For the folks at BackOffice Associates, however, it reflects a reality they see every day, which is that business computers are filled with data that is wrong or duplicated, and that these errors can cost a company hundred of millions of dollars. With an array of tools and services, BackOffice enforces data quality, and can help customers get the upper hand on their data quality issues.
BackOffice Associates was founded 14 years ago by the husband and wife team of Tom and Patricia Kennedy, who sold their previous company, a development tool maker called Kennedy Associates based in Richmond, Virginia, several years earlier. Since then, the Kennedys and their 300-plus employees have worked to build the Cape Cod-based company’s reputation in the field of data quality, migration, and governance.
Much of the vendor’s accomplishments in the field of data quality and data governance stems from its work with SAP ERP migrations. Before a customer populates a new SAP system with data housed on an older system, BackOffice Associates can identify which data is likely to be incompatible with the new SAP system and cause real-world problems, like late deliveries.
“It takes us about four days to completely analyze all their data and tell them where the issues are going to be if they attempted to operate their new ERP system with the current data they had in the legacy system,” the company’s CEO, Patricia Kennedy, says in an interview with IT Jungle.
That means knowing exactly how data sitting in one ERP system or database management system will look to another ERP system sitting in a different database management system. Aside from system incompatibility (or the lack of “business readiness”), BackOffice Associates can identify duplicate and incomplete data, no matter where it resides.
The company created a suite of about 70 products (all developed from Tom Kennedy’s CranSoft rapid Web application development environment) that tackle various aspects of the data quality issue. Experience also plays a big role in making BackOffice Associates, which has 450 customers including 80 percent of the Fortune 1000, one of the preeminent names in the data quality business.
“We have reverse engineered the entire supply chain and built our own generic data model for all these different ERP systems,” Kennedy continues. “We do a lot of work with Oracle, JD Edwards, Man-Man, Syteline, BPCS, and Infor. You name it. We touch it.
“But we cut our teeth the last 14 years on SAP. In our opinion, SAP is the most complex global ERP environment. If you can do SAP, you can do anything.”
Bad, Bad Data!
Bad data creeps into enterprise applications in a number of ways, and can have a huge impact on an organization’s ability to complete business processes. Trish Kennedy talks about the high level of complexity involved when a customer uses SAP to automate a manufacturing process.
“Any material they want to enter into that ERP system, 11 roles [or people] are going to have to touch those screens in order to create that material” [master data entry], Kennedy says. To create a single finished product in one location, SAP requires the use of 50 different screens, the company says. When you figure a typical global manufacturer is making hundreds or thousands of products in dozens of countries, you can see why it commonly takes over a month and a half just to enter material master data into an ERP system.
While the multiple levels of screens and approvals are built into the system to help reduce business process errors, they sometimes lead people to take shortcuts, which can lead to data errors. This is especially true when you consider that none of the major ERP vendors have workflow-enabled the processes to streamline the necessary approvals, according to Kennedy.
BackOffice Associates’ solution to this problem was to create a program that insulates the people from the underlying complexity of the ERP screens. This software functions as a “firewall to your ERP” by ensuring data is entered fully and correctly while simultaneously blocking out the portions of the screen that users don’t need to see to do their job, Kennedy says.
“The data has to be complete and accurate when you push the button and post it into your ERP system, whether it be JDE or another system,” she says. “When you post it, if it’s incomplete, then you run the risk of business process interruption within your supply chain.”
On the flip side, bad data may lead an organization to overestimate its customer size or customer’s needs. Duplicate entries in the customer master file could lead an organization to believe it has 14 million customers when it really only has 2 million distinct customers. Similarly, an organization may overstock with inventory if it doesn’t have an accurate view of its customer demands, hurting profitability.
The impact of bad data can also be felt by companies that have taken steps to curb the spread of bad data by implementing overly strict or inefficient protocols. For example, some companies have created “centers of excellence” within their organizations, whereby a group of people use spreadsheets to coordinate and validate the data before it’s entered into the ERP system, Kennedy says. This can stifle the spread of bad data, but is expensive and time consuming, and not very scalable.
While BackOffice Associates offers tools and expertise to help organizations deal with data quality problems, the best fix is for an organization’s leadership to take a top-down approach to data quality.
“A lot of customers we deal with believe that somehow IT is going to fix their data quality problems,” Kennedy says. “What we stress to customers is the business has to own the data and has to be empowered to control the data, to affect the data quality. If you’re waiting for IT to come up with some great strategy technically to effect the data quality, then you’re pretty much barking up the wrong tree.”
BackOffice Associates’ solutions range from about $50,000 into the millions of dollars. For more information, see www.boaweb.com.