 |
Hidden
Secrets: Reliance
on a SAS 70
for SOX Testing (Part 2 of 2)
By: Mitchell H. Levine, CISA
Audit Serve, Inc.

HP NonStop Server
Security and Encryption Solutions
www.xypro.com
Reliance on a
SAS 70 review is not an approach that most companies would prefer in
ensure that SOX controls are effective within an outsourced service.
However, there are no alternatives for most companies and therefore
they have to make the best of the situation which at times requires an
adjustment of the controls they expect to be in place.
This is the second part of the article which focuses on the
testing of the controls which are presented by the service organization to
the service auditor to test. In
order to determine whether the testing within the SAS 70 review can be
relied upon for a SOX project, one must determine whether the sample size
and sample period used within the SAS 70 meets the requirements for SOX.
The sample size requirements for SOX are based on the control
frequency. Guidance has been
provided by the CPA firms of the sample size requirements which need to be
compared to the sample sizes used within the SAS 70 report (i.e., 20 for
daily, 10 for weekly and 2 for quarterly controls).
Some of the traditional testing methods used in SAS 70 reviews such
as collaborative inquiry and observation cannot be used since these test
approaches do not meet the types of required SOX testing in order to meet
sample size requirements. In
addition, most SAS 70 reports do not identify the control frequency and
therefore the service auditor would need to be contacted to obtain this
information.
The second question is whether subsequent SAS 70 reviews have
been linked together to cover the sample period for the entire year.
SAS 70 reviews are structured to cover a specific period in which
testing was performed to ensure that controls were effective for the
entire period. However, many
service organizations do not schedule continuous SAS 70 reports to cover
the each year and therefore they may be gap which relates to the period in
which controls need to be effective for SOX.
It should be noted that SOX only requires control to be effective
as of the end of the fiscal year. Therefore,
as long as the SAS 70 test period covers the end of the fiscal year of the
user organization, reliance can be placed on the SAS 70 review from a test
period standpoint.
If an organization outsources their computer operations to a
vendor, it should be understood that the systems included in the testing
of controls within the SAS 70 review may not be the actual systems in
which the user organization’s data resides on.
However, SOX has been emphasizing the concept of management level
controls in which assurance is obtained that all units have adopted and
follow the same controls structure. Therefore,
testing of sample number of units would provide assurance that all units
follow the same control structure.
Mitchell
Levine is the founder of Audit Serve, Inc. Audit Serve performs al types
of integrated and IT Audits, SOX testing and SAS 70 reviews.
Email Mr. Levine at Levinemh@auditserve.com
if you would like to discuss alternate project approaches for your next
SAS 70 review.
Copyright 2006, Audit Serve, Inc. All rights reserved.
Reproduction, which includes links from other Web sites, is prohibited except by
permission in writing.
This article appeared in a past issue of the Audit Vision
E-Mail Newsletter.
|