- Comprehensive
- Fast
- Accurate
- Scalable
- Easy to use
- Easy to learn
- Supportive of best practices
HEADLINE-WORTHY SPREADSHEET ERRORS
However, along with the sophistication of spreadsheets comes complexity, and complexity can make spreadsheet errors and risks more difficult to find. When undiscovered and uncorrected, spreadsheet errors can be costly. In 2003, Fannie Mae needed to adopt a new accounting standard for recognizing the value of the mortgages.
To implement the new standard, Fannie Mae relied on internal accounting systems along with spreadsheets that made additional calculations. Unfortunately, one of those spreadsheets contained an error, resulting in a $1.2 billion variance in the financial results that Fannie Mae reported in its 10-Q.
After discovering the error and announcing the correction,
Fannie Mae saw its stock price drop $2.25. In addition to the loss of market confidence, Fannie
Mae had to put in the additional work to correct the report filed with the SEC.2
Barclays learned the hard way that hidden columns in spreadsheets are hidden only from view;
their data and its implications remain intact. In 2008, Barclays bought Lehman Brothers
investment and banking and trading business and thought it had cherry-picked just the good
assets.
As it turned out, when a spreadsheet was reformatted and converted to PDF by two junior
lawyers with minimal Excel skills, a “hidden” column became visible. This hidden column
represented $30M of the not-so-good assets that Barclays did not want. After years of litigation, a
judge ruled that the error was not fraud and the deal was legal.
Unfortunately, the errors that affected Fannie Mae and Barclays are all too common. In a recent
survey of finance users in SMBs and enterprises, Ventana Research found that:
- 45% of finance users in enterprises grapple with the problem of multiple, inconsistent spreadsheets.
- 35% of survey respondents said they were aware of data errors in the spreadsheets used for the organization's most important processes.
Here are the requirements for implementing such a solution:
• Completeness
Enterprises need a way of automatically discovering and cataloging all spreadsheets, so that no relevant collection or analysis of data is excluded from error-checking and analysis.
Enterprises need a way of automatically discovering and cataloging all spreadsheets, so that no relevant collection or analysis of data is excluded from error-checking and analysis.
• Speed
Enterprises need to be able to check spreadsheets, even 100 MB spreadsheets, quickly. Manual inspection often takes days or weeks.
Enterprises need to be able to check spreadsheets, even 100 MB spreadsheets, quickly. Manual inspection often takes days or weeks.
• Accuracy
Error-checking should automatically uncover risks and errors and improve the accuracy of spreadsheets.
• Scalability
Analysis tools that might work on a few 10 MB spreadsheets might become impractical when there are 25 spreadsheets averaging 100 MB each. At many large enterprises, the core business may be managed through tens of spreadsheets averaging 80-100 MB each. The solution should be able to accommodate this much data or more.
• Ease of use
Once risks are identified, correcting them should be fast and easy. Finance users should not have to switch back and forth between reports and spreadsheets or databases and spreadsheets in order to assess errors and apply corrections. They should be able to assess risks and make changes directly in the spreadsheet itself.
• Rapid training
Intuitive design and helpful documentation should make it easy for audit teams and finance departments to adopt a spreadsheet risk management solution. By reducing training requirements, the risk management solution boosts spreadsheet users’ productivity.
• Support for best practices
Spreadsheet analysis and risk notification should support industry and organizational best practices for developing and maintaining spreadsheets. For example, if a spreadsheet risk analysis application flags errors by reformatting spreadsheet cells, it might create confusion in organizations that use formatting such as color-coding to identify different types of cells, such as input cells. Risk analysis should highlight risks without altering or overwriting spreadsheet contents.