SAN MATEO (03/20/2000) - Business decision-makers analyzing risk frequently work with spreadsheet models of business situations. However, the need to manually enter multiple variables quickly consumes too much time to make it practical to test a wide range of variables. Using a smaller set of variables results in decisions that are based on the review of a very small portion of the data model's possible outcomes -- sometimes just the best or worst case scenarios.
Palisade Software Corp.'s @Risk Professional 4.0, beta, automatically runs hundreds or even thousands of possible variable combinations, based on ranges defined by the user. The results are displayed in tables or a wide range of flexible graphs, all within the familiar interface of Microsoft Excel.
Palisade Software touts @Risk as the most powerful solution available in the risk-analysis software market. It supports models with more than 50,000 probability distributions; a competing product, Crystal Ball from Decisioneering, supports about 1,500.
However, casual risk-analysis users may find @Risk's myriad controls and functions a little overwhelming, making a more basic package more appropriate.
I tested @Risk with Microsoft Excel 97. I installed the software in minutes and found the online tutorial to be a useful overview of the software and risk analysis in general.
Launching Excel in the standard way, I was glad to see no sign of the added @Risk toolbars; the @Risk toolbars and special functions appear only when the @Risk application is used to launch Excel, making it far less intrusive for frequent spreadsheet users.
Palisade's @Risk comprises three main components: the @Risk Model window, the @Risk Excel add-in, and the @Risk Results window. The Model window allowed me to designate inputs and outputs for my simulation, view input distributions, and define correlations.
The Results window provided me with a variety of output options, including interactive graphs and scenario reports. The Excel add-in provided special @Risk functions that could be typed directly into cells in my existing spreadsheets with just a few keystrokes.
As I worked with some of @Risk's sample spreadsheet models, I found that the product uses colored text in the spreadsheets to make it easy to identify cell values that have special @Risk functions running behind them. This colored text shows users exactly where @Risk functions exist in any spreadsheet in which the @Risk functions are used, and makes it far easier to modify the spreadsheets.
I found Palisade's claims of high-speed simulations well-founded. My modest test platform, a Pentium II, 166MHz laptop, proved more than adequate to run simulations in a matter of just seconds.
Palisade @Risk's extensive and flexible graphing capabilities surpassed those available with Crystal Ball. I viewed my result sets as histograms, cumulative area graphs, or tornado graphs. Graphs viewed while running simulations are updated in real time, which provides a moving picture of the risk-analysis model results.
Furthermore, the graphs themselves contain a variety of controls for adjusting variables such as probability distribution. Advanced risk-analysis users and trained statisticians will find this level of control appealing.
Palisade's @Risk 4.0 Professional includes the BestFit application, which sold separately in previous versions. BestFit allows for distribution fitting or insertion of sample data from an external source into a simulation. For instance, in one model I worked with, I used historical pricing data for a product to provide possible prices in my risk-analysis model. This capability provides a powerful tool for forecasting future events based on past outcomes.
@Risk 4.0 comes in three versions: Standard, Professional, and Industrial.
@Risk 4.0 Standard does not include the BestFit application, whereas the @Risk 4.0 Industrial version includes both the BestFit application and an additional component: the RiskOptimizer, which allows for optimizing simulations with genetic algorithms. Past versions of @Risk are also available for the Lotus 1-2-3 spreadsheet application.
Some minor display glitches appeared when I viewed multiple windows within the @Risk Results window. However, when I minimized and expanded the windows, these problems disappeared.
Overall, the application performed well, without any major crashes or errors during testing. We tested a very late beta version of the product, so we were happy, but not surprised, to find it in good condition.
Newcomers to the world of risk-analysis software will appreciate the @Risk online manual's thorough treatment of the subject, which included detailed descriptions of the included risk-analysis spreadsheet models.
Data preparation is key
Simply purchasing @Risk and installing it on your PC will not propel you into the world of risk analysis. @Risk requires well thought out and constructed spreadsheet models of real-world events to provide useful data. Without an accurate spreadsheet model, @Risk can return nonsensical results.
If you have no experience creating risk-analysis spreadsheet models, you will need to start with basic models, then increase the complexity as you become more comfortable with @Risk. You will also need some time to become comfortable with the dozen or so buttons @Risk adds into the Excel toolbar.
Business decision-makers looking for spreadsheet-based risk analysis that goes beyond simple best case/worst case scenario models should take a look at @Risk 4.0. This relatively inexpensive piece of software can provide any organization with invaluable insight into the potential risks and rewards of complex business decisions.
Chad Hammond is a professional services manager at LeadTrack Systems, a San Mateo, Calif.-based customer relationship management consulting firm. He can be reached at firstname.lastname@example.org.
THE BOTTOM LINE: BETA
@Risk 4.0 Professional, beta
Business Case: @Risk performs risk analysis on complex spreadsheet models, which yields highly relevant output. Its integration with existing Excel spreadsheets means that current modeling spreadsheets do not have to be re-created.
Technology Case: @Risk's Excel add-in architecture means a small footprint on the desktop and little or no end-user support requirements. Simulations run quickly even on lower-end clients.
+ Tight integration with Excel
+ Extensive list of risk-analysis functions+ High-quality output graphs+ Inexpensive client with no server componentCons:
- Confusing user interface for novice users- Some display glitches when viewing multiple windows in Results screenCost: Professional edition, $695; Standard edition, $495; Industrial edition, $1,195Platform(s): Microsoft Windows 95/98, Windows NT 4.0, Microsoft Excel 7 or laterShipping: Late March 2000Palisade Software Corp., Newfield, N.Y.; (800) 432-7475; www.palisade.com.