Travel & Expense

The Power of Multi-Dimensional Analysis

on June 10, 2014

Consider this scenario: You are a travel program administrator at a Fortune 500 company. You are tasked with managing the travel program, making sure that corporate travel policy is adhered to, and protecting the company from loss or brand damage due to bad behavior.  You decide the best way to accomplish these three things is to both identify T&E Spend that is out of policy and catch transactions with merchants that are considered “risky”. But how? The most common method is to use a manual audit of 10% of the expense reports, but this is labor intensive and misses what is happening in the other 90% of transactions.  You could aim for 100% audit, which is a lot more work, but even then, how can you identify those merchants who disguise their identity? Here are a few additional ways to ensure program compliance and investigate potentially fraudulent transactions: 

  • One approach is to run reports on credit card transactions based on “risky” Merchant Category Codes.
  • Purchase a “strip club” list and try matching it to merchants in expense reports. 
  • Cross check risky transactions for activity that occurred on weekends. 
  • You might search for keywords in expense report descriptions, comments or merchant names.
  • Spend time analyzing behavior patterns of an individual. Watch for trends and look for abnormal spend amounts in various categories. How does this individual compare to the rest of the company? 

All of these are good ideas. Many can be partially effective, but all of these approaches result in a lengthy manual review of long lists. Not to mention that cross checking numerous reports can be hard to correlate.  Finding those transactions that were on a weekend, at a risky MCC code, and with an MCC rarely uses is, quite frankly, like finding a needle in a haystack. Wouldn’t it be great if you could do all of these analyses in parallel and apply a weighted risk to each of them?  Now, what if this was automatically done for you? 

That is exactly the way that Oversight’s Insights On Demand™ works!  It applies many analytic techniques at the same time: MCC code analysis, statistical analysis of amounts, behavioral patterns of employees, corporate spend patterns, weekend/holiday analysis, external data such as keywords and address lists, and then combines these tests using artificial intelligence techniques that give a weighted risk to each analytic test. This multi-dimensional layering quickly identifies the most risky expense items – the needles in the haystack. 

Now that you have such information easily at your fingertips, what are you going to do about it?  In the next blog I’ll discuss the integrated case management tools that Oversight provides that allow for easy review, email follow-up and an audit log of all actions taken.

Nathanael L’Heureux

Nathanael L'Heureux is Vice President of Solution Consulting at Oversight Systems. He holds a Bachelor of Science degree in electrical engineering and computer science from the University of California, Berkeley.

spend-analysis-vol3