30 years after they were introduced, spreadsheets remain heavily used in the business world. This dependency on spreadsheets is almost never a good thing. Every company Triple Helix has worked with has convoluted, error-filled, opaque, human-dependent Excel processes in place. Talk to any organization about their dependency on spreadsheets, and comments like these will surface:
- “I am totally dependent on 11 spreadsheets to get my job done, which I know is way too many. And the worst part is that I only have a solid understanding of the calculations and logic for 3 of them.”
- “I spend more time making sure my spreadsheets are accurate than I do analyzing the data in them. As the CFO for this company, that is just ridiculous.”
- “When it comes to making important decisions, I am really uncomfortable relying on figures spit out of a huge spreadsheet that 8 different people have worked on.”
- “It’s frightening that all these Excel spreadsheets we use are determining forecasts, monthly profitability and other things, instead of the systems we have which are supposed to be doing that.”
An over dependency on spreadsheets occurs for one main reason: Excel is available on just about every employee’s desktop, and it can be used with little or no training. This is great for ease of use, and for many purposes Excel is a powerful, flexible, accessible tool. It can also lead to a lot of problems.
The bigger problems occur when Excel is used to do things it wasn’t designed for, such as become everyone’s favorite workaround solution for systems already in place (ERP, CRM, etc.). Many businesses don’t take Excel’s limitations into account and stretch the software beyond what it was built to do. When this happens, considerable risk can be introduced. For example:
- A spreadsheet is the brainchild of one person. It was not designed as a collaboration tool, which is what commonly happens. As the number of authors increases, so does the number of invisible errors that creep in, and so does the difficulty in maintaining an accurate spreadsheet.
- The bigger and more complex the spreadsheet, the more fragile Excel becomes. It can simply break at certain points. When this happens, the risk of data corruption and lost data goes way up.
- Spreadsheets promote silos of information in a company and decentralized data management. It makes automating links between spreadsheets and applications tough to do and is a primary reason why so much costly, error-prone manual handling of data occurs in companies.
Triple Helix recommends the following approach to organizations whose dependency on spreadsheet software is hurting instead of helping productivity and introducing too much risk:
- Assess spreadsheet use in the organization, at an individual level, group/collaborative level and business process level. Gain a clear picture of how spreadsheets are being used company-wide.
- With this inventory of spreadsheet usage in hand, determine where your larger exposures lie:
- Large/complex spreadsheets
- Spreadsheets whose originator is no longer at the company
- Business-critical spreadsheets that impact daily operations
- Spreadsheets where important financial or other data is stored
- Spreadsheets tied to manually intensive processes
- Spreadsheets whose data is pulled from or exported to other business applications
- Quantify these as much as possible from a risk and/or cost perspective.
- Establish a plan for migrating the high-risk and high-cost spreadsheets to a better solution. Bring in a 2nd set of eyes from the outside if necessary to help with this part. The objectivity and experience can be quite valuable if the pain felt by the organization is great enough.
Companies that have grown on the dependency on spreadsheets need tighter controls on how they should be used, who has access to them and where data should be stored. Without these controls, it is very easy for management to lose sight of how the business is actually operating. You will inevitably end up with multiple views of the truth as well. And it likely means your current systems that manage data aren’t cutting it, which may be reflecting people & process problems rather than technical shortcomings.