Dangers of Excel
The electronic spreadsheet (e.g., Microsoft Excel, Lotus 1-2-3, Visicalc) is arguably the highest societal impact software in history as it has enabled “end user development” of a wide variety of useful computational tools in almost every industry worldwide.
However, the spreadsheets have proven to be error-prone and hard to audit: Studies have shown that 86-94% of the spreadsheets being used (not just in existence, but regularly used) contain errors. [Pan2000] [Pow2009] [Pan2015]
It is extremely difficult to audit spreadsheets, and thus rarely done. [Pow2008] And even where auditing is done, error detection rates tend to be low, and even lower as complexity rises. [Pan2015] With weak typing and no support for units of measure, and a design that emphasizes flexibility and ease of authoring, automated error detection is very limited. [Boc2016]
The difficulty in auditing, combined with the lack of any way to control who edits what portions of spreadsheets, make spreadsheets easy targets for fraud and manipulation. [But2002] [Pan2005] [Mit2005]Worse, the spreadsheet comes with a number of other negative impacts:
Many billions of dollars in losses each year can be attributed to erroneous spreadsheets. [Boc2016] And that does not include the significant dollar impacts of reduced collaboration, innovation, continuous improvement, and multi-dimensional understanding caused by the inherent deficiencies in the electronic spreadsheet model.
Frankly, given the inherent risk and weakness, it’s unsettling how many engineers rely on spreadsheets as their go-to computational tool.
The European Spreadsheet Risks Interest Group (EuSpRIG) website has a list of 89 “Horror Stories” that were caused by spreadsheet errors.
If you notice, these are mostly financial in nature from government or public company sources.
However, it is doubtful that spreadsheet errors are limited to the accounting and financial fields. This skewing of the data is more likely attributable to the fact that these entities are required by law to make these errors public when discovered.
Given the insidious reliance on spreadsheets within the engineering profession, its likely that there are similar stories. Those are probably not advertised publicly for obvious reasons. Engineers keep those painful lessons to themselves and teammates.
While powerful and flexible, spreadsheets have certainly proven problematic in the accounting and financial world. In the even more complex world of mission-critical engineering design and analysis, spreadsheets are obviously not the best tool for the job.
Are you an engineer with battle scars and war stories? Would you like to anonymously share with us your story of spreadsheet pain? We’d love to hear it, and eventually share it on this page for other engineers to learn from.
Is there a better way?
Success Assured® is a non-spreadsheet alternative for “end user development” and collaborative visualization and presentation. It is certainly not a replacement for all uses of spreadsheets and presentation software, but rather an alternative that can be far more effective for many uses, particularly the support of complex decision-making, problem-solving, and optimization.
Success Assured® is designed to:
- use superior visualization to enable effective collaboration, review, and audit, even across different disciplines (areas of expertise)
- enable editing of those visuals to be quicker-to-edit than a whiteboard, encouraging them to be modified on-the-fly during collaborative discussions
- enable dynamic "zooming in & out" and linking such that the audience can easily redirect the discussion, rather it needing to follow the pre-determined path
- enable ownership & control by qualified experts while enabling inputs across disciplines of expertise
- use a superior structural organization to enable much easier multi-dimensional analysis and to increase reusability
- enable set-based (rather than single-point) computations and analysis for efficiency and flexibility, increasing reuse and reducing complexity and errors
- enable more natural / visual exploration of the multi-dimensional decision space, increasing innovation and collaboration
- enable human-in-the-loop optimization that corresponds better to how people need to make decisions
- use a structure that exposes the cause-and-effect relationships that can drive better decision-making and problem-solving
- enable set-based data analysis that exposes worst-case limits and safe vs. unsafe regions for making wise decisions, even in the face of uncertainty
- prevent the common mistakes in statistical analysis and risk management, and avoid models that miss key anomalies or catastrophic failure points
- leverage strong typing, units of measure, and dimensional analysis to automate error-checking of the end-user formulations, further reducing mistakes, and simplifying “end user development” by automating some of the most error-prone work (like converting between different units of measure)
- leverage the natural 2D mathematical notation for equations, along with use of visual cues to aid in human reading and auditing of those equations to further reduce errors and improve collaboration and analysis.