Since it launched in 1985, Microsoft Excel has been a powerhouse in the world of data analysis and business processes across the globe. A whopping 1.1 billion people use Microsoft's productivity suites across the world. While this doesn’t give us an exact figure on the number who use Excel, it’s reasonable to surmise that about 1 in 8 people on the planet uses Excel.
So, you’ve probably got the gist that Excel is pretty popular and is depended upon by many businesses. So, you would expect the software to be a polished, reliable and foolproof tool, leaving no room for error. However that is not always the case. While it is an effective tool, it brings with it its own set of risks, with the companies below having learnt the hard way.
Kodak suffered an $11 million severance error
Back in 2005, photography giant, Kodak, suffered an $11 million overstatement on an employee’s accrued severance and retirement benefits record. The reason for this wasn’t even a complex one, but simply a typo in which too many zeros were added to the spreadsheet. The error came at an already bad time for the company who were already losing over $100 million every quarter. The error was spotted and rectified meaning Kodak were able to readjust their financial statements and restate a combined $11 million worth of losses to two quarters. This certainly didn’t rectify the company’s overall financial situation, and they still made an overall loss of $1.038 billion.
Barclays Capital spend millions on worthless contracts
In 2008, Barclays Capital bought 179 contracts they didn’t want, or even know they were buying. The mishap happened when Barclays Capital were buying contracts from Lehman Brothers, who had gone bankrupt. Barclays Capital prepared an excel report that contained the contracts they intended to purchase, however due to rows being hidden, instead of deleted, the company ended up acquiring 179 contracts they didn’t want. The hidden rows were only discovered when the Excel was converted to a PDF, by which stage they were legally bound to these acquisitions. The exact cost was never declared, but we imagine it wasn’t small.
Public Health England misplaced nearly 16,000 positive COVID-19 tests
During the pandemic, data was clearly a massive concern. Whether it was the number of positive cases, the number of people in isolation, the number of countries in lockdown, or sadly the number of deaths, one thing was certain - the data was overwhelming. And for Public Health England, this caused serious issues resulting in nearly 16,000 covid cases going unreported.
The data error which led to 15,841 positive tests being left off the official daily figures, meant that 50,000 potentially infectious people may have been missed by contact tracers and not told to self-isolate.
Much of the reporting was done manually, with individual testing labs sending spreadsheets containing their results. The error occurred when one lab provided their daily test report in the form of a CSV file, which was then loaded into Excel. Sounds simple, right? Well the problem occurred because Excel has a limit to the amount of rows it contains, and the version the PHE had could only process 65,000 rows of data. This meant that the bottom of the CSV file was completely missed. After the incident, PHE changed the process going forward where each lab were required to break down the test result data into smaller batches to ensure none hit their cap. Problem solved? Well, yes, however this method still sounds extremely inefficient, and surely there is a high risk of some of the spreadsheets being lost or forgotten about.
MI5 bugged the wrong phones
In 2010, a formatting error on an Excel list of phone numbers to be tapped by M15 resulted in the UK security service accidentally tapping 134 people, who were completely unrelated to the investigation. The error occurred when the last 3 digits of phone numbers were changed to 000 within the spreadsheet. Once the error was discovered, all of the material was destroyed and phone numbers had to be manually checked from then on, which would definitely have been a time consuming and laborious process.
London Olympics oversold Olympic event by 10,000 tickets
The London Olympics were publicly embarrassed by an innocent Excel typo that led to a swimming event being oversold by 10,000 tickets. Instead of typing 10,000, an employee typed 20,000 meaning twice as many tickets went on sale than intended. The blunder was only spotted when the number of tickets sold was cross-checked with the seating layouts and configurations.
After the mistake was noticed, organisers had to contact over 3,000 customers to offer them replacement tickets for other events, many of which were deemed to be better events than the swimming heats that they originally opted for. Great news for those customers, however embarrassing press for the organisers. Every other ticket sale was then manually double checked thereafter to ensure the error was not repeated.
AstraZeneca accidentally released confidential information
Probably now best known as the pharmaceutical partner for the Oxford COVID-19 vaccine, AstraZeneca hasn't always had the best press coverage. Back in 2011, they accidentally embedded confidential company information into a spreadsheet that was distributed to a large list of analysts who were only expecting to receive financial guidance reports.
Errors like this happen more often than you might realise when sharing spreadsheets, when certain cells, rows or entire sheets are hidden temporarily, then mistakenly unhidden or not deleted fully. Confidentiality is of vital importance to all businesses, but particularly a pharmaceutical one where they have mountains of data on patients and drugs.
JP Morgan hit with $6 billion trading loss
US financial services company JP Morgan suffered a massive loss of over $6 billion in 2012 from a simple copy and paste error in Excel. The company were using spreadsheets to create value-at-risk (VaR) models and was copying and pasting data from one spreadsheet to another. An employee mistakenly copied the wrong data into a spreadsheet, resulting in the model massively underestimating the risk. The user had accidentally used the total sum, rather than the average of two values, producing extremely inaccurate results, leading to the company taking much larger risks than they ordinarily would have done, had they had access to the right information.
TransAlta lost 10% of their profits
A Canadian electricity company, TransAlta Corp, lost 10% of their profits in 2003 due to Excel errors. At the end of each month, they submitted bids for various transmission hedging contracts for the following month. This process sees transmission paths with higher demand having higher prices. In April of 2003, they were shocked to discover they had won a series of contracts that they didn’t need, unfortunately at much higher prices than normal. The $24 million error was caused from a simple copy and paste error where rows of information in the spreadsheet were misaligned, meaning higher bidding prices were assigned to routes with much lower demand. Unfortunately for TransAlta, the rules state that once a bid is submitted, it cannot be edited so they were forced to continue with the transaction.
How to avoid making spreadsheet errors
So you might want to pause for a second before hitting send on your next Excel file. Human error can be avoided by setting up stringent processes, however it cannot be eliminated entirely. And no matter how many Excel training courses you go on, it cannot prevent and guarantee that there will not be typos or errors in your spreadsheet.
So if spreadsheets aren’t reliable, then what is? We’re glad you asked. Hurree's analytics dashboards removes any doubt and gives you 100% confidence in your data. Data is automatically populated on your dashboard, directly from the source, giving you peace of mind that the data in front of you is 100% accurate. Try now for free.
You May Also Like
These Related Stories