Developing large spreadsheets is long, complex and fraught with the possibility of error.  Spreadsheets have been around for twenty five years, but even today the only way of developing them is still by hand, directly into the open spreadsheet. And the bigger the task, the larger the spreadsheet is likely to be, the longer it will take to develop and the longer it is likely to remain in use.   

EuSPRIG - the European Spreadsheet Risks Interest Group - give examples at their website of  some of the horror stories.  For every horror story, though, there are thousands of times as many examples of spreadsheets that were just a little bit wrong.  Many of these are used for real purposes - decision support, budgeting, financial projection, financial reporting, engineering calculation, medical analysis or a multiple of other uses.  If such a high proportion are simply wrong, from a little bit, through quite a lot, all the way up to the horror story just waiting to break, just imagine the result of raising the average proportion of unflawed spreadsheets from say, the 50% it may be now (and many studies claim that that proportion is actually much, much lower) to the 95% + that it should be.

The present way of doing things has other drawbacks too.  A big spreadsheet written by someone else is likely to be totally opaque - even its author would find bits on which the spotlight has not recently been shone difficult to decipher.  There is almost certainly no specification and no documentation, and even if it is quite clear what numbers are going in and what numbers are coming out (a big if) how the output is derived from the input is usually totally opaque - in short, a black box.  Whatever the version number it is given, a spreadsheet contains everything that has ever been put into it, unless explicitly removed.  Thus bits will almost certainly not have been looked at for months if not for years.

And if that wasn't all bad enough, the process of creating the spreadsheet will have been as lonely, time consuming, stressful and as difficult for the author as it could possibly be.

There has to be a better way . . .