|
|
|
Features
< Back
Sarbanes Oxley : Auditing : Management
Using Software to Improve Analytical Procedures in Audits and Reviews
By
Brian Hamilton
|
|
Brian Hamilton
Chief Executive Officer
ProfitCents
|
Implied in the growing public consciousness around accounting generally and auditing specifically is that these disciplines can never be sciences. One of the larger fall-outs from the scandals of the past several years is that the investing public has begun to believe that accounting can be an exact discipline. It cannot be. The best accounting and auditing still require plenty of judgment and estimation.
As such, the best auditors are the ones who combine a thorough knowledge of auditing with an ability to apply common sense and judgment to the process of an audit or review. Of course, the difference in the quality of a good audit can often be found in the quality of the judgment of the auditors. This article will deal specifically with how technology and software can be adopted in the analytical procedures portion of audits and reviews in order to make more time available for the softer and more difficult task of improving judgment decisions by the auditor. Helping the practitioner to free-up time in the analytical procedures process will provide him/her with more time to look at the overall condition of the company being audited or reviewed. As such, glaring trends that may be troublesome can be spotted earlier and with less effort.
One of the largest challenges that accountants and auditors have is that they must deal with so much data. It can sometimes be very difficult for professionals to ferret out the differences between the important and the unimportant in the process. It is also difficult to recognize trends. In essence, there is a distinct difference between accounting/auditing and financial analysis, yet both are pivotal to the analytical procedures process. There are so many transactions that the auditor?s job is made tough, especially in cases of fraud or near fraud where the client has every incentive to obfuscate or muddy the records being supplied to the auditor. This is a vital point to understand and one that many people may not appreciate. One of the reasons that the profession has had so much strife over the past several years is not just a matter of ethics but a matter of judgment in the face of a mountain of data resident in the audits of even tiny companies.
How can the profession deal with this? Recent legislative and regulation rulings may help in the future. At a minimum, these regulations give the appearance to the public that the profession is dealing with problems. However, it is dubious that these reforms get to the heart of the real challenge, which is to minimize/make easy the technical or rote parts of an audit and allow more time for the judgment components. If the technical parts of an audit can be made easier through the use of technology, then more time can be devoted to the qualitative parts. This is really important to understand. If auditors are spending so much time on the rote parts of an audit like calculating expected values, for example, then little if any time is left-over for a broader and more global picture and analysis of the business.
There are certain aspects to the process of estimation that can definitely be improved and made more efficient through technology. One of these is setting expected values in the analytical procedures portion of an audit or review. ?Expected values? are the account values that might be anticipated, based upon the past performance of a company. Setting expected values takes time and effort and some knowledge of statistics. As such, auditors can sometimes get sloppy around this key area of analytical procedures.
Why is this so important? By identifying what general ledger account values should be, auditors can apply a management by exception basis for looking into values or accounts that are ?out of line? with what they should/might be. Rather, if auditors have to pour through all Income Statement and Balance Sheet accounts with equal energy, then there is an increased chance of missing larger problems. While it is true that setting expected values can never be perfect, the methods of deriving values can and should be improved through the application of technologies.
Up until recently, the task of deriving expected values has been quite time consuming. For example, say you were attempting to derive an expected value for sales/revenue for the following company:
Example 1:
| | 2002 | 2003 | 2004 | 2005 |
| Sales/Revenue | $10M | $20M | $40M | ? |
What might an ?expected value? or estimated value be for 2005 for the sales of this company? In this example, many methods might be applied to determine the expected value of sales. However, logically, it looks like an expected value for sales for 2005 might be approximately $80M. This is derived by recognizing the trend that sales for the company in question seem to be increasing by approximately 100% each year. The calculation to derive the trend from 2003 to 2004, would be $40M (sales in 2004) less $20M (sales in 2003)/$20M (sales in 2003). The raw data in this example would be compelling as a starting point because:
1. It is so clear. It does not really deviate.
2. There is a clear trend that can be applied to 2005.
In other words, in looking at data like this to set values, you would tend to weight the clear financial trend less than the externalities of the firm (the market or industry). The mathematical constructs in determining values for this company are so obvious.
Take another company example:
Example 2:
| | 2002 | 2003 | 2004 | 2005 |
| Sales/Revenue | $10M | $20M | $30M | ? |
What might be the expected value for sales in this case? It?s less clear because the sales of the company are increasing but seem to be increasing at a reduced rate, at least for the years being evaluated.
Or, how about the following case?:
Example 3:
| | 2002 | 2003 | 2004 | 2005 |
| Sales/Revenue | $10M | $10M | $20M | ? |
Again, it would seem logical that the expected value of the company?s sales would increase but the calculation would be more difficult, given the uneven data for the periods under consideration.
It?s rare when data is as smooth and clear as it in Example 1. Where a simple trend analysis might be a good starting point for Example 1, it seems like a different mathematical technique like regression analysis might be more appropriate for Examples 2 and 3. Even though regression analysis cannot be a perfect one for all parts of the process, it would seem to be a good methodology as a starting point for auditors.
Regression analysis is cryptic and time consuming, especially if it is applied to all accounts. For example, while it is possible and probable that you might apply regression analysis to a particular set of company data, think of how time consuming it would be to perform calculations manually against all accounts. Technology makes possible a quicker and more reliable method for calculating expected values in cases where auditors are either using both trend (two periods of data) or regression analysis (three or more periods of data). Mathematical models can be easily applied to this part of analytical procedures to give accountants time to look at overall conditions and values out of line rather than having to devote time to rote calculation.
If setting expected values is achieved more quickly, how will this specifically help increase the work done in analytical procedures? Using good statistical modeling does not eliminate the need for high quality analysis of the data and other methods to determine expected values. This is where the process of determining expected values becomes more ?art? and less ?science?. We want to minimize or lesson the science by automating the calculations and increase the amount of time available for qualitative review/judgment decisions. A reduction in the time it takes to calculate the statistical rote part of the process will allow for more time and research into the other components that can make the difference between a good audit/review and a bad one:
1. Interviewing management. Understanding how financial and economic conditions are changing in the company.
2. Performing industry research. The company?s financial performance is driven by both factors inside the company (good management, good products and service, etc) and factors outside the company (industry changes, employment, industry averages/benchmarks).
3. Evaluating trends and looking more deeply at accounts that are unusual.
There are several software applications on the market today that speed-up the process of setting expected values. Many auditors also create their own analytical frameworks using ?Excel?. Macros can be developed that use statistical models that are generally accepted (An example would be the ?times squared? calculation in regression techniques). The general idea in any good system is to have a set of algorithms that automatically calculate expected values using historical values as a basis.
In summary, there are several specific advantages to using technology to set expected values to ensure better analytical procedures, which will lead to better reviews and audits:
1. Calculations are quicker and more reliable. On balance, less chance for error is introduced through automation.
2. Data that is rough or uneven can be predicted with more accuracy.
3. More time can be devoted by auditors to reviewing data and less time to calculating it. As such, auditors can delve more deeply into accounts that are ?material? and possibly more troublesome.
Integral to getting better audit practices is helping the auditors with the real challenges they have on a day-to-day basis. It?s likely that auditors involved in recent, well publicized fraud cases could have more readily detected problems if they had devoted additional time to basic overall management and financial analysis. Using software and technology to reduce the time needed in analytical procedures will give accountants time to devote to the full scope of an audit and review.
Brian Hamilton
Chief Executive Officer
ProfitCents
Brian Hamilton is the chief executive officer and leader of the management team for Sageworks, Inc., which develops ProfitCents, an application that aids accountants in communicating with clients.
Brian can be reached at brian.hamilton@sageworksinc.com or 919.851.7474 (x501).
|
|
|
|
|
