Menu
Working Smart

Working Smart

At Bank of America Mortgage, employees were drowning in a sea of paperwork. The office, which handles first residential mortgages for more than 750,000 customers a year, waded through over 1 million pieces of paper annually-a combination of bills, declaration pages, cancellations and reinstatements of hazard polices.

"Our problem was, How can we process this paper better, faster and less expensively?" says Ron Marx, assistant vice president and manager of the hazard and tax department at Bank of America Mortgage. It took 45 full-time employees to handle the homeowner insurance process from beginning to end.

The insurance processors had to open the mail that came from insurance companies and agents, find and circle the loan number, and put it in order according to due date. Then they had to count the forms, bundle them with rubber bands and carry them upstairs for the next person to key in the data. "It was very manual," says Marx. Mistakes could happen easily; if a processor read something incorrectly, a form that belonged at the top of a file could go to the bottom. That could mean a long wait for homeowners, who ultimately put the pressure back on mortgage companies to speed the process.

In 1997 the department decided it had had enough. Marx approached Cigna Software Sciences (which later became Ace USA Software Sciences, in Albany, N.Y.), which offers software solutions and outsourcing opportunities for financial institutions. Joseph DeChiro, director of marketing and sales, was happy to help. "When data was manually processed, the mail came in and they'd have stacks and stacks on everybody's desks," he notes. DeChiro introduced Bank of America Mortgage to MaxMilion software, developed by Ace.

On the front end sits OCR for Forms, the flagship product of Tampa, Fla.-based Microsystems Technology OCR for Forms extracts data from scanned or faxed forms, using a combination of optical character recognition, intelligent character recognition and bar-code recognition. It validates the data, presents questionable characteristics for correction and creates an ASCII file of the data.

From there the data heads into the back end of MaxMilion, which uses decision-making rules to make sure that everything is processed in an identical manner. Bank of America Mortgage is in the process of building templates for the most common forms that come through the process-currently, 50 per cent of the forms have templates, and Marx hopes to see that figure increase. Data that comes from templated forms heads right into the MaxMilion system, which uses a series of rules to compare the data and make sure everything goes in the appropriate place. Is the loan paid yet? Does this insurance go with that homeowner? Once everything is verified, it's ready for disbursement or rerouting through the mainframe system.

If the form doesn't have a template, it's scanned and then the data is entered manually by data processors into MaxMilion, where it joins the rest of the process. "We really streamlined the process by stripping it down," says Marx.

There were other solutions out there, says Marx, but many of them took the paper away from the person without tying that into the rest of the process. MaxMilion did both.

After the implementation, Marx was able to reduce the processing staff to 20. "It could take up to three days to get through the mail on a Monday morning," says Marx. "Now the turnaround time is less than 24 hours." The biggest payoff, he says, comes in control. "As long as [a document] is scanned in, you can't lose it."

Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.

Join the newsletter!

Or

Sign up to gain exclusive access to email subscriptions, event invitations, competitions, giveaways, and much more.

Membership is free, and your security and privacy remain protected. View our privacy policy before signing up.

Error: Please check your email address.

More about Cigna Life Insurance New ZealandFinancial Institutions

Show Comments
[]