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

The top five data management mishaps of 2011

Some of last year's biggest information management mess-ups could have been avoided through closer adherence to best practices, according to experts.

The staff at searched the globe for some of the most notable data management mess-ups of 2011 and came up with some very interesting fare. There is a case of mistaken identity, a rich couple feeding from the government trough and even the late Apple chief Steve Jobs was in on the action. Experts say that all of the stories in this list could have been avoided by adhering to generally accepted data management best practices. In no particular order, here are our picks for the top five data management mishaps of 2011.

And the rich get richer

A married couple living in a $1.2 million ocean-front home in Seattle, Wash. has spent a good portion of the last eight years vacationing in exotic locales like Moscow, the Dominican Republic, Mexico, France, Israel and Turkey.

But now the U.S. attorney’s office is accusing the couple – David Silverstein and Lyudmila Shimonova – of fraud for receiving more than $135,000 in federal housing and disability assistance during the same time period.

Federal prosecutors allege that Silverstein, a successful chiropractor, received $1,272 a month in housing assistance payments from the government for more than eight years after Shimonova claimed that she lived alone with her two children and that Silverstein was her landlord. The government says Shimonova also received financial assistance under the Temporary Assistance for Needy Families program. The government is now seeking repayment plus tens of thousands of dollars in fines.

Data management experts say the improper payments might have been spotted sooner with simple de-duplication or matching software that can help users find irregularities by looking for duplicate customer names, addresses and other attributes.

“This is the kind of thing that gets people’s emotions going,” said Tony Fisher, president of DataFlux, a master data management (MDM), data integration and data quality software vendor. “Many organizations have duplicate entries of their customers and if they don’t link it together they don’t get the full understanding of who the customer is.”

That ex-felon isn’t aging well

Last December, an unfortunate case of mistaken identity prevented a Boston woman from getting a much-needed job. The woman was offered a job as a pharmacy technician pending a clean background check. But much to her surprise, the check turned up 14 criminal indictments.

Experts say poor data management techniques prevented the background check company from recognizing the difference between the job seeker and an ex-felon with the same name who was 18 years younger. The case is still under investigation.

The case is another example of a matching issue that could have been avoided with de-duplication software that looks beyond names and addresses to other attributes like age, gender or city of birth, according to Fisher.

“What happened here was that there were two confined sets of attributes that went into making the match,” he said. “That’s a good example of why you might want to look beyond just the most basic of attributes to determine whether two things are the same.”

Indian exports rise $15 billion – give or take $9.4 billion

Economics experts were quite surprised to see that exports from India surged to $15 billion during a seven-month window last year. After some inquiries, the Indian government acknowledged that a data entry mistake caused the value of exports to be overestimated by about $9.4 billion.

Fisher said the mishap should serve as an example to all organizations that data entry problems --while common -- can lead to all kinds of issues and concerns. He added that in addition to encouraging employees to enter data with care, one way to lower the incidence of data entry errors is through reference data management.

“Reference data management basically defines the valid values for particular entities. They’re all defined in one place and they’re all going to be defined by some corporate policy board,” Fisher said. “When data is entered, it’s banged up against those rules for validation.”

I didn’t know I owned that much property

About 20,000 residents of Brockton, Mass. got a bit miffed at the tail end of 2011 when they received third-quarter real estate tax bills that overstated the acreage they own by a multiple of 10. City officials say the mistake by Kelly & Ryan Associates – the company that prints Brockton tax bills – was the result of a wayward decimal point. The incident follows a two-year dispute where angry residents took issue with water bills that reached as high as $100,000.

Apple admits 'mistakes' amid data privacy concerns

Now-deceased Apple chief executive Steve Jobs admitted to several data management “mistakes” last year amid concerns over how Apple’s iPhone and iPad devices were handling users’ location data.

In early April, two researchers found an unencrypted file in Apple’s devices that contained information about locations visited by users during the previous 12 months. The find sparked fears that Apple was tracking users and prompted several investigations. 

Apple denied tracking users and attributed the issues to programming errors and bad decisions – such as the decision to keep the file unencrypted and to store the data even when users had turned off location services. Apple eventually issued a patch designed to resolve those problems.

Experts say the story should serve as a reminder that any organization that collects user data must handle that information with care, according to Jon Oltsik, senior principal analyst with the Enterprise Strategy Group in Milford, Mass.

“If you are collecting data you should err on the side of over-communication with people to make sure that they understand what you’re collecting and what you’re using that data for,” he said. “If that information comes out covertly, people really get mad.”

Dig Deeper on Data management tutorials

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.