News Stay informed about the latest enterprise technology news and product updates.

Data quality buoys U.S. Naval Institute

Data quality software helped the U.S. Naval Institute clean up customer data, improve member services, reduce wasted time and save money.

In November 2003, the 135-year-old U.S. Naval Institute (U.S.N.I.) received a returned membership renewal letter in the mail. It didn't contain a check for membership dues, however.

Instead, at the bottom of the form, was a handwritten note. It was from the intended recipient's widow, reminding the Institute that the member had died more than a year earlier and that this wasn't the first time the Institute had been notified of that fact.

The error sent a message that "we don't really care" about our members, said Rick Woodcock, U.S.N.I.'s chief technology officer. Based in Annapolis, Md., the civilian nonprofit agency publishes books and magazines and runs conferences on naval-related issues, and it relies on its 50,000 plus members, mostly retired Navy personnel, for much of its funding.

"This is not the way we want to treat our members," Woodcock said. Poor data quality was sinking U.S.N.I.'s reputation.

Case study: Customer data quality
Organization: U.S. Naval Institute  

Industry: Non-profit professional organization  

Project: Reduce duplicate member and customer data, ensure valid addresses, share clean data organization-wide  

Technology: Experian QAS Pro, QAS Pro Web, QAS MatchIT, and QAS Batch  

Months in use: 10

ROI/results: Reduced duplicate member profiles from 10,000 to 10 to 15 per day, improved operational efficiency, streamlined data sharing
The problem was that each department at U.S.N.I. – including the member services, fundraising and publishing departments -- used different databases, including Microsoft FoxPro, 4D and SQL 7 databases, and kept its own, isolated membership and customer records. As a result, each member had a different profile depending on whom you asked, and data – like the death of a member -- was not shared among departments.

But embarrassing mishaps like the one with the deceased member's widow weren't the only problems disparate data sources and duplicate membership records caused. The institute was also wasting plenty of time verifying names and addresses, and money, on postage and production costs for magazines, books and membership renewals that never reached their destinations.

To fix U.S.N.I.'s data-sharing problem, Woodcock decided to standardize all departments on Microsoft Dynamics NAV, an enterprise resource planning and database system for small and medium-sized businesses. He knew the investment would prove worthless, however, unless the Institute first cleaned up its data quality problems.

Fixing data quality at the root

After several frustrating months evaluating data-quality vendors, Woodcock eventually discovered Experian QAS. The London-based vendor specializes in name and address verification technology but also makes software that reconciles duplicate data records. It proved to be the only vendor that promised to fix U.S.N.I.'s data-quality problems at the source, Woodcock said, rather than just "put a Band-Aid" on it.

Woodcock invited Experian QAS on-site for a demonstration and liked what he saw. "I listened to a lot of vendors that were all talk," he said, declining to name the other vendors he considered. Experian QAS "was the first company that really delivered what they described they could do for the price they said."

"Duplicate data was a huge problem. We really didn't have any way to weed through it."
Rick Woodcock
chief technology officerUnited States Naval Institute
After a successful proof-of-concept, U.S.N.I. purchased and implemented four Experian QAS products in October 2007: address verification software QAS Pro and QAS Pro Web, data de-duplication tool QAS MatchIT, and QAS Batch for data integration.

QAS Pro and QAS Pro Web are designed to ensure that only valid addresses can be entered into a database. When a U.S.N.I. representative enters a new member's zip code into the system, for example, QAS Pro prompts him or her to choose the member's city or town from a verified list of municipalities in that zip code, then a valid street, then building number, and so on. The result is fewer keystrokes by U.S.N.I. member services workers and fewer magazines, books and other correspondences being sent to the wrong place.

QAS Pro Web does the same thing for members, creating their own profile on the U.S.N.I. website, reducing errors and preventing members from entering false addresses.

The MatchIT product lets the Institute's various departments compare membership rolls. The software identifies and ranks the likelihood of possible duplicate member profiles in, say, the magazine and conference departments' respective databases, allowing Woodcock and his staff to manually reconcile the results.

Before implementing MatchIT, on any given day the Institute was sitting on 10,000 duplicate records, Woodcock estimated. Now, he said, the number of duplicate records is down to a remarkable 10 to 15 per day. It's such a small figure that U.S.N.I.'s two part-time receptionists can run the MatchIT software and reconcile duplicates during their spare time.

Shared data solves and creates problems

Ironically, Woodcock realized, once implemented, Microsoft Dynamics NAV could create problems of its own. When data is silo'ed, each department can keep data in the format it wants. To share data across departments via a single data source, however, data must be entered in a consistent format. But, he said, different U.S.N.I. departments have different needs.

More on data quality tools and best practices
Learn why Gartner says data quality and data integration software are converging

Find out how to avoid three common data quality mistakes
For example, the member services department, which sends magazine and book shipment orders to a warehouse for fulfillment, uses the standard U.S. Postal Service format for member profiles: all uppercase letters, no punctuation marks and everything abbreviated. The Institute's fundraising department, on the other hand, prefers "wedding invitation" style – names and addresses spelled out – for a more personal touch when soliciting large donations.

"[The fundraising department doesn't] want anything to have even the slightest hint of cookie-cutter, boilerplate [language]," Woodcock said. "Everything that they do, they want it to look personalized and addressed specifically for that person … but [member services] cannot send data in that 'wedding invitation' format to our warehouse because it'll break the system and nothing will ship."

The solution? U.S.N.I. employees enter data in the more personalized, "wedding invitation" format so the fundraising department can send customized letters to donors, but the QAS Batch product transforms the data into standard postal service format for the member services department so it can easily ship orders.

"Batch allows us to choose which way we put the data in and adjust it as it comes out," Woodcock said.

Once its data quality problems were ironed out, the U.S.N.I. began implementing Microsoft Dynamics NAV, a three-year process it expects to complete in the next 18 months. Then, cleansed data can truly be shared across the organization.

"It's very important for us to maintain a personal touch, not to alienate our members, so [they] feel that they're a part of something that cares about them. And that was the big goal," Woodcock said. "Beyond that, the money saving that came out of doing business with QAS was icing on the cake."

Dig Deeper on Data quality techniques and best practices

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.