The BPO Problem: Data Integration
- By John K. Waters
Remember Y2K, when it seemed as though everyone was jumping on the ERP bandwagon and doing big data migrations? You never heard much from the big vendors about how difficult that job was.
Five years later, as business process outsourcing gains momentum as a way for companies to cut costs by hiring firms to handle functions in human resources, finance and customer service, David Lyle is experiencing deja vu.
“It’s the same problem, only in some ways magnified,” Lyle says. “Now even the system integrators don’t want to talk about it. But the data isn’t any cleaner, and the job of migrating and integrating that data isn’t any easier.”
Lyle is VP of product strategy at Informatica. The 12-year-old firm is known as a leader in an area of data integration historically referred to as Extract, Transform, and Load, the processes that enable companies to move data from multiple sources, reformat and cleanse it, then load it into another database.
Today, Informatica is embracing a broader definition of data integration that includes operational data integration—things like data migration and consolidation, building operational data stores, synchronization among different databases and master data management.
One of the drivers of this change, Lyle says, is the accelerating shift among companies of every stripe toward outsourcing. BPO in particular is forcing companies to rethink their data integration strategies, says Matthew Polly, Informatica’s VP of product marketing. In almost every case, BPO requires the migration of in-house data to BPO providers.
“More and more of our customers are leveraging our technology and applying it to that particular use case,” he says. “They are recognizing that there is an important data integration and migration component to BPO.”
BPO keeps clients’ staffing requirements to a minimum and allows businesses to focus on their core competencies. But as more and more functions are outsourced, explains Lyle, the IT infrastructure becomes disaggregated, and the data becomes fragmented. Data silos that once stood inviolate behind the company firewall are suddenly stretched across organizational boundaries, adding a new dimension of complexity and increasing risks.
And there’s an important data integration and migration component to BPO that people don’t often consider, Lyle says. Most organizations won’t simply hand over their data and then kill their legacy systems. “They typically want to keep control over some of that information within the four calls of their enterprise for a while,” he says, “and then slowly wean themselves from the legacy applications they’re replacing with a BPO provider. We fit nicely into that use case, too.”
Informatica customers are also extracting data from the BPO provider systems for business intelligence and reporting so they can maintain visibility and auditability of the processes under way outside the enterprise firewalls.
Outsourcing is not a passing fad, but an industry-wide phenomenon that has been growing by leaps and bounds since about 1990. Informatica’s research shows that BPO in particular is spreading quickly, and that between 5 and 15 percent of BPO management is data integration.
John K. Waters is a freelance writer based in Silicon Valley. He can be reached