Solve Administrative Problems With Quality Tools

A Pareto analysis helps pinpoint why employee expense reports are rejected

by Edmund S. Fine

I spent most of my career in various hardware and software product assurance roles, but now find myself administering a department of 80 engineers and program support personnel, who take a total of 10 to 15 business trips per week.

Each trip generates an expense report, which has to be submitted to accounting so the customer can be billed and the traveler reimbursed. These reports must abide by strict government guidelines, so our accounting department developed a Microsoft Excel spreadsheet used to assemble the expense report information.

Unfortunately, I soon found more than 70% of the reports were rejected by accounting on first submittal despite a requirement for the report to be reviewed at the department level. The rules are fairly lengthy, but in most cases, they are clear and make sense.

The reasons for rejection

Obviously, the rejection rate was unacceptable--it made a lot of additional work for the department secretary and accounting, and the travelers were not being reimbursed promptly. Falling back on my product assurance experience, I classified the reasons for rejection and performed a Pareto analysis. From what the analysis indicated were the top four reasons for rejection, I determined the traveler forgot to attach at least one piece of paperwork to the report. Our rules clearly require:

  • Original receipts for items such as airfare, hotel and rental car fees.
  • Travel itinerary and boarding passes.
  • Daily meal and lodging information for the location visited (to be downloaded and printed from a continually updated Web site).
  • A calculation of allowable taxes on room rentals.

I recalculated the Pareto analysis, grouping the missing paperwork in one category, and realized 80% of rejected reports were rejected because they were lacking backup paperwork. In order to ensure this missing material was included when a report was submitted, the department secretary and I developed a checklist for expense reports, which listed all the necessary backup paperwork that had to be attached. I asked the secretary to summarily reject reports that did not have a completed checklist.

The secretary also compared the backup paperwork with the checklist and rejected expense reports at the department level when the checklist and the paperwork did not agree. The use of the checklist effectively eliminated unattached supporting documentation errors, and the rejection at accounting rate dropped dramatically to 20%. This drop, while indicating substantial progress, was insufficient for our purposes. We then started to search for the remaining root cause(s).

One is not enough

In order to help isolate the remaining causes, I did another Pareto analysis, this time on the error data obtained after the rejection rate dropped. I noted the missing backup categories had disappeared and the remaining significant causes were:

  • Missing information (employee number, reason for travel).
  • Entry in wrong column of spreadsheet.
  • Incorrect account number.
  • Mistranscription of amount from receipt.

For several weeks, the error rate held relatively steady, as did the apparent causes. A follow-up investigation traced the root cause of most of these remaining errors to the traveler's practice of updating a previous expense report spreadsheet (which was stored in his or her personal computer) to record the expenses for a subsequent trip. Entries in the wrong column, missing information and incorrect account numbers were thereby perpetuated.

After further investigation, I also learned the secretary continued to correct errors on the printed spreadsheets and submit reports to accounting so travelers would get paid promptly. However, no feedback about the errors went to the travelers.

A new policy

So we established a new policy: The secretary now corrects the paper copy of the expense report and returns the erroneous report to the traveler so he or she can correct the report. The traveler then prints out a clean, correct report and sends it to accounting after receiving departmental approval. There was some grumbling from the staff about delayed reimbursements, but we clearly explained that a traveler was responsible for submitting a correct report--period. This change reduced the rejection rate to less than 2%.

We have not yet found a way to cope with mistranscriptions except to have the secretary compare the receipt and report and return discrepant expense reports to the traveler to have the traveler make the correction. Still, the rejection rate reduction from approximately 70% to less than 2% has made life a lot easier for everyone, and I can thank my quality assurance training for teaching me this methodology.

EDMUND S. FINE is a chief engineer at Jacobs Sverdrup Advanced Systems Group in Quantico, VA. He earned a master's degree in industrial engineering from New York University and is a Senior Member of ASQ.

If you would like to comment on this article, please post your remarks on the Quality Progress Discussion Board, or e-mail them to editor@asq.org.

Average Rating


Out of 0 Ratings
Rate this article

Add Comments

View comments
Comments FAQ

Featured advertisers