This Guide is a workmanlike introduction to the craft of modelling in Excel business cash flow and financial statements, but it could do a lot more in the area of modelling uncertainty.
The strong point of the book is its emphasis on structuring the model and the modelling process. Models tend to take on a life of their own after the initial development, and sometimes even before the first build is complete, and time spent in thinking, planning and design is usually a very worthwhile investment. This is particularly important if there are multiple people involved in the model's design and use, and the book covers how one might organise a modelling project.
The principal modelling task is designing the mathematical model that produces the desired time series of sales, costs, etc, over time, based on a handful of input parameters. The book goes into some detail about the formulae needed to render the different kinds of time series one might encounter in a typical business model, covering areas such as macroeconomics, forecasting revenue (with good discussion of times series analysis and regression techniques), operating costs, capital expenditure, working capital, and funding, along with ancillary issues such as depreciation, tax, financial statements and company valuation methods.
Also welcome are the chapters on the often overlooked topics of testing, debugging and documenting a model. There is an introduction to recording macros and programming simple Visual Basic code, but anyone serious about programming Excel with VBA can find better material elsewhere.
As I said, a good introduction. It assumes some basic understanding of Excel and business terminology, but not a lot else. There are lots of practical spreadsheet examples with the formulae all well explained, although sometimes the text description is a little disconnected from the screenshot. It is a pity the authors did not go further in several key areas.
* The book introduces the concept of Excel range names, but does not do them enough justice. The naming of ranges is the single most important underused feature in Excel, and mastering its use will dramatically reduce the bugs in a model and, if names are chosen sensibly, provide instant documentation and a model that has a fighting chance of being understood by the next modeller who picks it up and has to maintain or update it.
* Sensitivity analysis is covered in just three paragraphs, when this is probably the single most important use of a model. Anyone can create a model that produces an answer (eg, the NPV of a business opportunity) for a given set of assumptions. But we all know these assumptions, if they represent forecasts of the future, are wrong. Regarding the model as a simulation and playing with ranges of assumptions (testing various "what-ifs"), so that you build a full understanding of the drivers of value and risk in the business situation, is where the real value of modelling lies.
* Monte Carlo simulation is again covered only briefly and with a rudimentary DIY approach. There is no mention of the many tools available that can drive a well-designed model through probabilistic analysis.
Finally, I have to comment on the chapter on macroeconomics, which makes use of the sine function to model economic cycles. Given the discussion in the press about the current recession and whether it will be V- or U-shaped (or maybe W- or--preserve us!--L-shaped), this seems hopelessly simplistic for any practical use. This jibes with the rest of the book, which is practical. However, the example does help introduce how to parameterise a time series, in this case with the cycle time and amplitude.