A core competency of the Vander Weele Group is the collection and analysis of large data sets.
Data, ranging from accounts payable to credit card expenditures, can be used to identify suspicious patterns suggesting fictitious invoices, ghost payrollers, excessive credit card spending or improper payments. We undertake the process from start to finish. First, we prepare the data by standardizing common entries categorized differently. Then we apply algorithms to identify anomalies, and finally, we investigate to determine the underlying reasons for the problematic outliers. Our work, which includes trend analysis, has saved clients millions of dollars.
A corporation saved $5 million to $7 million a year after VWG analyzed $40 million in credit card expenditures and identified 145 suspect users, some of whom were criminally charged.
Data analysis of procurement records expanded recoveries from a few hundred dollars to $1.3 million after an international manufacturer discovered indicators of a kickback scheme.
A client processing millions of dollars of checks learned that nearly a half-million dollars were stolen. Within three days, VWG identified the culprit, worked with the client to obtain a written confession, and supported the efforts of the U.S. Postal Service and U.S. Attorney’s Office in indicting the employee, who was part of multi-million-dollar Nigerian bank fraud ring. The culprit was identified using data analysis of phone records.
In 2009 and 2010, VWG performed the first-ever extensive analysis of the Native American Student Information System of the Bureau of Indian Education. VWG analyzed 230,000 school safety and misconduct incident reports from 184 BIE schools. It coded data by category and cleansed aberrant reports before providing the BIE with a high-level overview of crucial information. VWG identified factors affecting performance and areas requiring improvement in BIE’s incident reporting system and provided specific recommendations for change.
A federal agency was charged with locating individuals owed more than $60 million in federal accounts. Data analysis indicated that no Social Security numbers were available for 70 percent of the individuals and no dates of birth for 77 percent. Through open source data analysis, our team culled through death records, military records, genealogy records, online phone books, property records, handwritten U.S. Census records, and more to recreate each account holder’s family tree, tracing through adopted names, maiden names, and married names. In the first two years, our team and strategic partner firm, Lamar Associates, found more than 1,300 individuals associated with $16.2 million of the accounts.