My final review with the RCA
Corporate Audit staff involved the Accounts Receivable sub-ledger at the global
telecommunications subsidiary. By the time I was asked to help, the job was
already over the budgeted time and everyone was under the gun to get the report
to management. To keep costs at a minimum, I requested interns from Rutgers
Graduate School of Business.
|
During training
conducted by the Arthur Young CPA firm, I became the statistical sampling
"expert" when I corrected the instructor on the final day. Arthur Young had an
excellent sampling tool and I asked my boss if we could try stratified random
sampling to help with the project deadline.
Basically, a stratified
sample focuses more attention on higher impact accounts (in this case customer
records). We examined the printout of sample selections and found we were to
examine the top 5 customers and a representative sample from each of the other
strata. I won't discuss our results but I was promoted based on the report
findings and the relative speed in which we concluded the review.
Many
of the benefits of stratified sampling may be utilized in operations where a
strong reliance is placed on perpetual inventory calculations. People using
perpetual counts often need to make spot checks to verify the results of the
ideal usage formula. So what items should you spot check?
I would
recommend checking every item in your top 25 purchased goods list. Since you
won't be scrutinized by Arthur Young accountants, it's up to you to decide how
many additional items need checking between physical counts. To give you an idea
of the power of sampling, we used around 350 accounts in a universe with tens of
thousands. Our limited review located critical control issues and we brought
them to management's attention in a timely manner.
If you have 1,000
items in stock, check 5 or less shelf stable dry goods. With your top 25, this
will make 30 items. Try to randomly check another 25 to 50 (depending on time
required). Now the fun begins as you try to reconcile the perpetual inventory
level calculated vs. the actual amount on the shelf. The first five or six spot
checks will point out obvious recipe errors and yield issues. Since you're
sampling you should expect problems found for a particular class of items (for
example produce) may require more intensive work on the entire
class.
When you find a problem with portion control items and the recipes
are solid, note the dates of the last physical count and the spot check date.
Mark any of these problem items for increased scrutiny. You may consider control
sheets located near the walkin coolers and freezers.
The use of spot
checks and sampling techniques will greatly improve results in operations where
physical counts are required monthly. If you count everything each week, your
mid-week spot counts may be very brief and focused. Some operators count the top
25 daily. This should suffice for most weekly count operators.
An
excellent additional benefit of perpetual inventory spot checks is the
refinement of recipes used in the calculations. Accurate recipe costs and
ingredient yields are requirements in more sophisticated menu engineering
calculations.
| |
|
