PT the Auditor
They Are Claiming Half a Million Dollars ...
Oh, I'll tell you, it doesn't get any better than this. A local community, a rather large local community, has a problem. They had their auditor go in and found that half a million dollars in tickets were missing from a city-owned garage. They have a private operator running the garage.
OK, here's what the auditor did. He took two days and counted the tickets. He found 1,500 missing tickets for the two days. He then extrapolated that for the full two years that the operator had been running the garage. Well ...
Let's look at it. First of all, this method of computing loss is not unusual. We do it all the time. However, we usually take a larger sample.
When I audit a garage, I normally take a sample of two weeks, not necessarily consecutive, and then extrapolate it out over 51 weeks, not 52. That allows for issues such as holidays, weather and the like. The number then becomes much more accurate.
For instance, let's say there is a hotel involved in the process (as in the case above). The hotel has weekend seminars. On Friday, when everyone arrives, there appears to be a lot of missing tickets (because people arrived and stayed over). However, on Sunday, when they left, there would be an excess of tickets (more people left than arrived). These must be taken into consideration when doing an "average." In the case above, if the auditor took two successive Fridays, the numbers would be extremely skewed.
I am a hard-working old dog. If I find what seems to be a severe problem, such as in this case, I would actually track every day for an entire month - counting tickets issued and returned for every day and charting them. Patterns would appear, and the number of tickets missing could be more accurately reported.
In this case, the operator says the revenue control equipment had malfunctioned. Well, that happens; however, I personally like to place that issue at the feet of the operator. If a gate, TD or cashier terminal/POF malfunctions on one day, it should be repaired the next. The service company should be called immediately when there is a malfunction, and the entire incident logged.
The system is the only thing that is going to protect the operator when an audit happens. It should be running flawlessly. If it's not, the operator should be screaming at the owner to get it fixed or indicate that it cannot be responsible for the revenue while it's broken.
It's too late, it seems to me, to come up with a bad audit and then cry that the machine was broken. How long was it broken - for two days, two months, two years?
We do know from the reports I read that the operator appears not to be completely blameless. It seems that some of the cashiers were taking all-day tickets ($10), collecting the money and then validating them for $8. They would keep the $8 and put the $2 in the drawer. Any revenue control system working flawlessly that uses any kind of manual validation program will not report that.
The system to catch this problem should report "cancelled transactions." Typically, the cashier runs a ticket, displays the amount due to the patron, and then cancels the transaction and runs a new validated transaction that would open the gate and be recorded.
A plethora of "cancelled transactions" would set off an auditor's alarm bells.
I don't know what type of validation is done here, but the best in order of auditability are:
Online machine-readable validations, with machines in each place that provides validations.
Off-line machine-readable validation machines.
Off-line validation stamping machines.
Stamp coupon books that are self-adhesive to the tickets and scored so they can't be removed. These must be numbered and auditable.
Rubber stamps - not good, because anybody can have a stamp made.
Signatures - oh, please, I don't have to talk about this, do I?
Of course, pay-on-foot machines that take the cashier out of the loop and systems that take credit cards in real-time greatly reduce the issues of lost tickets and this type of fraud.
Auditing is in the details. If an auditor is lazy and looks at only a small sample, he is going to have a problem when it comes time to back up his claims with facts. This may end up in court, and when it does, the operator will bring in someone like me to tell the other side of the story. There could be a lot of kibble on the floor before this one is over.
Article Abstract from September, 2006