Analyzing MMSD Spreadsheets

The report entitled What We Know About Spreadsheets by Professor Panko of the University of Hawaii analyzing the literature on accuracy of financial spreadsheets shows a truly appalling result -- significant errors, and furthermore, reliance on the accuracy, and denial of problems.

As I look at the spreadsheets for the budget and list of maintenance projects justifying the referenda, I'm overwhelmed by their complexity and my inability to feel comfortable with their calculations.

I'm no expert at using spreadsheets, relying instead on real databases and batch processing to handle such complex tasks. I have no doubt that MMSD Business Services folks are quite good at spreadsheet use, but spreadsheets are incapable of being maintained even if accurately created in the first place.

And spreadsheets are simply incapable to allowing the timely generation of multiple budget scenarios. The delay in pulling together the budget for this year (from multiple sources), a typically manually intensive process when using spreadsheets for financial and budgeting analysis, shows their inadequacy.

To make the budgeting and financial aspects of MMSD transparent, and analyzable by the public, and capable of generating multiple budget scenarios requires a drastically improved system. And my demands that contracts and decisions affecting the budget (such as cuts proposals) be handled only within the confines of a published set of budget scenarios, make the need to move away from spreadsheet use imperative.

Posted by Larry Winkler at May 11, 2005 4:31 PM
Subscribe to this site via RSS/Atom: Newsletter signup | Send us your ideas