The High Costs of Errors and How to Avoid Them in Your Business
A staggering 250,000+ people are killed in the US each year as a result of medical errors, according to a BMJ report. This makes medical error the third-leading cause of death in the US, behind only heart disease and cancer, at 9.7% of all deaths. Surprisingly, here in Canada, we don’t seem to be collecting or accurately reporting the same information. That said, a University of Toronto report published in 2015 suggested 7.5% of all hospitalizations in Canada “had an adverse event that harmed patients” and a Canadian Institute for Health Information (CIHI) report found that medical errors affect nearly 25% of Canadians.
I start with this because it’s about as “costly” as an error can be when we look at the loss of human life. Furthermore, it stems beyond the root of the error and illustrates the need for proper data collection and evaluation of tangible cost (beyond the loss of life).
For example, in that same CIHI report, further exploration is done into the types of errors reported and it showed that these errors cost 1.1 million extra days in hospital. One can imagine the hard costs associated with that amount of extra care, care that was not originally assessed or anticipated — not to mention the potential liability from malpractice suits and the costs incurred by the patient (lost work, ongoing treatment, etc).
The reality is, humans make mistakes. While we have made incredible progress in many fields, perhaps most significantly in medicine, things are still forgotten, overlooked and in some cases — ignored. Environmental stresses, priorities and unforeseen circumstances can exacerbate this — all of which is outside of personal control areas like lack of sleep, illness, mental distraction and of course ignorance.
When we shift the focus onto business, the loss of life likely (and hopefully) exits the scenario, but grave consequences of errors and omissions still come into play.
These “consequences” or costs can take many shapes and forms depending on the specific case and are certainly not always financial in nature. Some of the more prevalent, sometimes easier to calculate, examples might be:
This could be hard costs or losses, missed revenue, liability, additional time, cost to resolve, and so on.
Time spent fixing an error, finding the error, resolving the error, delays in production — the impact of time can be just as detrimental as hard costs, and of course still have a financial measure.
Errors impact clients, vendors, staff, even partners — they can all be lost as a result of error. This is amplified when errors are repeated.
While there are certainly other ways errors can impact a business, those are a few that most would agree are critical in nature. After all, if you’re losing time, money and people — you likely won’t be in business long, right?
All that said, errors and mistakes are definitely a part of life. They cannot be completely avoided. In fact, they sometimes drive great innovation, but, they need to be controlled and certainly not all categories of error are going to lead to the discovery of something great… like popsicles (yup, totally a mistake!).
So, how can we reduce and better manage errors?
DISASTERS: How to Avoid, Reduce & Handle Errors
In no particular order, I’ve assembled nine key areas you can work on to avoid, reduce and handle errors — and to avoid DISASTERS. Admittedly, I was pretty proud of myself for being able to develop such a fitting acronym for this. Many are easily implemented and while there are certainly more, specific to your individual cases, many of these are great steps towards error reduction.
D is for Data:
Data is king. Many businesses have a number of data sources, many also don’t. Some are fragmented, some are raw and some are nicely organized. Ultimately, the more data we have, the more we can draw from it to drive insights, predictions, identify areas to address and items to resolve. Not using data means more guesswork, longer times to resolution and implementation — all of which can ultimately lead to further errors. Using existing data is key, but finding new datasets to harvest can assist further. Ensuring that available systems are used and reporting is enforced can go a long way in preventing and reacting to errors and issues.
For example, capturing data on the rate of returns in a product business, and the reasons for those returns, can mean honing in on specific challenges. If “wrong product received” is a common error, you can hone in on your order entry process, your product descriptions, or your e-commerce setup. Without this data, you will have a harder and longer time finding these causes, if it all possible.
I is for Importance:
Sometimes the steps are right, but the order is wrong. Sometimes steps are missed, or even skipped. And, sometimes, steps get missed or confused because of external factors and stressors. Taking a step back and creating policies and frameworks for prioritization and assessing the importance of steps is critical. Lastly, conveyance of that importance — and the importance of processes and systems in the first place, is a critical step. People will pay closer attention if they know the consequence of their misstep.
For example, if you inform your team that not calculating shipping rates ahead of a quote could mean rejected orders or loss of sales commission if absorbed, they may be more inclined to double-check their steps when preparing orders for their clients.
S is for Sources:
What we’re looking at with sources is two-fold. It’s the number of sources, or “points of failure” and also the ability to identify the origination of an error. Are you entering data into a spreadsheet and then uploading to a software, or to multiple? Are you then manipulating that data at another point? Aside from the system being fed bad data as a potential first error, the steps that follow all invite further mistakes. Having a single entry point can mean fewer steps, less wasted time, and higher accuracy. Secondly, interacting with elements of multiple systems can mean great difficulty in tracking the source or origination of a given error or unpleasant circumstance. Consider mapping out your processes, your tasks, your systems and considering where steps can be reduced, systems can be consolidated or even where things can be more logically placed to lend to better execution.
For example, we recently worked with a client who had staff members transcribing data into a spreadsheet at the end of each data that was taken from written forms. The team members would then send those sheets to a manager who would compile them into a master sheet. From there, the sheet was sent to a fulfillment manager and also used to create invoices. A common occurrence was fulfillment errors — but how easy do you think it was to find the source of those errors? Was it the original paper form that resulted from a verbal discussion, or was it one of the 2–3 transcriptions and manipulations of that data that followed? Now you have fulfillment and billing being impacted and need to go back to isolate the error and consider how to avoid it. Even by simply directly entering data into the spreadsheet from a tablet in the original conversation, a step could be avoided. Replacing all of those steps with a web-based system ended up being the best solution to consolidate steps and automate calculations, but simple steps can often be taken before a new system is implemented.
A is for Automation:
Stemming from the previous section on sources and building onto systems, it’s very often the case that once a workflow is created and optimized, that it can be digitized and automated. This can deliver massive impact to an organization in areas like error reduction, time reduction, accelerated delivery times, accuracy and more. While many business owners and managers assume things like digital process automation, AI and other forms of automation are reserved for enterprises — the reality is sometimes simple requirements have simple solutions — and the ROI can be astronomical.
In my previous example, we were able to turn 5 steps into 2 and introduce checks and balances. We were able to automate some of the calculations and conditions that were being applied in one of the steps. We were able to pre-populate select data from pre-verified information. All of this resulted in the vast reduction of error and also in time, turnaround time and billing accuracy.
S is for Safeguards
Here we’re looking at those “checks and balances”. If a particular step, or process, is critical, consider how you are ensuring its completion and accuracy. That could mean a step in the workflow where another individual is reviewing something. It could be a check against another data source. It could be an automated logic in a system that is built to ensure valid data entry.
Leaving everything to one individual and not accounting for missteps, busy days, oversights and things of the sort is a surefire way to happily invite error into your organization.
For example, if we look at web-based forms as a point of data entry, consider required fields. This prevents the user from proceeding without an important piece of information. Taking that a step further, the format, the syntax, can be dictated to ensure the right information is going into the right spaces. Another example would be a “pending” status on that form submission until a superior reviews the information.
T is for Training
Along the same lines as information, training is an important step to consider. Training on software and particular aspects of a job are common, but training on processes and prioritization are sometimes not well documented, covered or even present. As complexity increases, reliance on mental storage becomes more challenging — and when combined with distractions and other external factors, is welcoming errors into the equation. Creating checklists and tools and then providing training them helps keep things on track and accessible.
Secondly, training should exist for what to do in the case that an error occurs. How should it be escalated, who needs to be involved, what reporting is required?
E is for Encouragement
Doing everything you can to avoid errors is something to strive for, but expecting them to never occur is unrealistic.
Conveying the importance of your systems and processes, technologies and software iscritical. While non-compliance or repeated errors must be addressed, creating a stressful situation around the occurrence of an error may prevent employees from reporting them — or at least quickly. You want to foster an environment of acceptance and communication and to get as much insight and data into each error as possible. This cannot be done if people are terrified to report them or if there are no systems in place to even be aware that they occurred.
Provide a positive work environment that promotes focus, quality and accountability — and prioritize quality of work (for example, accuracy over speed).
R is for Roles & Responsibility
This one can be tricky, especially in small businesses, but it refers to who is doing what. There’s a couple of aspects to look at when it comes to roles. Firstly, if everyone is doing everything, it’s difficult to know who is ultimately accountable for a particular task and it’s difficult to have a “specialist” in an area that another can turn to for help or escalation. Having clearly defined roles means interfacing with specific processes or systems and it should create accountability.
Secondly, we can look at origination. If we know the nature of the original task, we can likely narrow the search of who may have made the mistake and where far quicker if we individuals are tasked with different components of processes or areas of the business.
In an example like users having a single set of credentials they share for access to a particular system, this creates chaos. Knowing who did what and when is nearly impossible if you are bypassing the way the system has been set up for individual users.
S is for Systems
Systems here don’t necessarily refer to technologies. It can be simple checklists, posters with reminders, meeting structures and a variety or combination of other elements. The key is to develop a system that thoroughly addresses the steps involved in a given task or process that has been reviewed, refined, prioritized and made widely available (and trained upon). Sometimes simplicity is key. You want these tools to be used and adhered to. When it comes to the efficacy, yet beautiful simplicity, of a checklist — I would highly recommend checking out The Checklist Manifesto by Atul Gawande.
Once processes are created, workflows, and those “systems” are adopted, very often a technological solution can be introduced to further optimize that system and to automate logic and highly repetitive aspects to avoid error, save time and expand possibilities.
Errors affect all organizations, big and small, and range in size from small to big. Some errors can be minor, some cumulative and some disastrous. By taking action to get ahead of these with plans of reduction, plans of management and plans of action, you can begin to reduce the occurrence and/or impact of errors on your business. Like many things in life, if you don’t take action things will remain the same. If errors are a reality in your organization, there’s no time like the present. If you’re one of the lucky ones who hasn’t yet encountered critical errors, I’d urge you to consider digging deeper to ensure they are not going undetected and then to consider what the cost of one significant error could inflict upon your business.
To the latter point, I’ve assembled a bit of a “blunderlist” below. You may or may not share the same sense of interest in seeing the massive impact of error, but at the very least these are sure to raise some eyebrows.
Errors: By the Numbers
I’m quite sure the well runs deep when it comes to major errors that have occurred in history, and keeping that list up-to-date would likely be impossible. To illustrate the potential impact of errors and to push past the assumption that your business is somehow immune to error, I’ve put together some key examples and figures.
I recently wrote an article specifically honing in on the negative impact that spreadsheets have on businesses, as with the use of such a manual tool inevitably comes errors. In that article, I shared that 88% of spreadsheets were found to contain errors in a series of studies, and I went on to highlight some of the examples that’s had on business — which of course relate to this article. The most notable example was JPMorgan Chase suffering a north of 6 billion dollar loss in part stemming back to a copy/paste error in their spreadsheets — a staggering example of the true cost of errors when dealing with manual input.
There’s an age-old, humourous acronym in IT circles known as a PICNIC error — breaking out to mean “problem-in-chair-not-in-computer” — or variants thereof. Essentially, these “errors” are referring to cases where a “tech” problem is actually the result of the human operator — not the technology itself malfunctioning.
For example, in 2017, Amazon’s AWS services crashed after a team member executed the wrong command — originally attempting to speed up processes. After a nearly five hour outage, impacting giants like Netflix, Airbnb and countless others, Cyence (a cyber-risk firm) estimated the cost to these companies to be in the neighbourhood of 150 million dollars in lost revenue — amongst other calculations.
While ultimately avoided, Toy Story 2 was nearly completely deleted after a member of the Pixar team ran an incorrect command.
A single typo in a line of code can result in an array of outages, unexpected behaviours and more when it comes to looking at errors in the use of technology.
Human Error Woes
Human error is the leading cause of data loss in business — sitting far ahead of its competitors like corruption, hardware failure and malicious attacks — at a hefty estimated 75% of all occurrences. An older report from Marketwire back in 2008 found that human error costs businesses in the US and UK north of 37 billion dollars in lost productivity. As technology has advanced dramatically since then, you would think that companies have taken significant steps to reduce these figures, however, cases come out far too regularly for that to be the case.
Even in critical areas, like the health care example, there’s still far too much room for human error.
Perhaps a large financial example to illustrate this would be Wall Street in 2010, in which investors lost 1.1 trillion dollars in a “flash crash” that occurred with the market plunging 1000 points in minutes. A costly example of what can happen when a simple typo is made by a well-intentioned individual — in this case a letter B being placed on a sell order instead of an M, representing billion and million respectively and sending the automated markets into chaos. Luckily, much was recovered — probably faster than confidence in the market.
Engineers at the Mars Climate Orbiter weren’t so lucky when their nearly 328 million dollar spacecraft was destroyed after units of measurement weren’t properly converted. In this case, it really wasn’t “rocket science”…
Manufacturers have it the worst though, with 23% of their unplanned downtime resulting from human error — a pretty big bump over the 9% found to be the average across other sectors in a Vanson Bourne study.
Life and Death
As we’ve already explored, human error can cost more than lost profits and workdays and is a leading force behind injuries and deaths across the globe.
Many of the last five decades’ most serious accidents at work could be traced back to human error. An article from IRMI outlined a number of examples of this, including:
- 20,000 people dying in a 1984 plant explosion at Union Carbide
- 167 workers killed in a Piper Alpha oil explosion, also resulting in a major oil spill
- 53 workers killed in Arkansas at a Titan missile silo
Unfortunately, those are only a few pulled off a much-too-long list of incidents resulting from human error and causing loss of life at work.
When we turn to focus on the commute, we still have more to do, as an estimated 90% of all vehicle crashes are caused at least in part by… you guessed it — human error.
When we look at the legal system, bit of an “oldie but a goodie”, but our friends south of the border published a report back in 2000 that analyzed death penalty cases since 1973. They found that more than two-thirds of those cases had “serious error-mistakes” caused by either police or prosecutors. The result? Of the analyzed 4,500 death penalty cases, 68% were thrown out or reconsidered in the appeals process. On the other side of the coin, consider how many people are executed in error — meaning they were innocent. This data is far more difficult to collect and is referred to as a “dark figure” in the Proceedings of the National Academy of Sciences.
As more and more notable cases surface of ransomware attacks, data breaches, phishing attacks and other cybersecurity attacks, businesses are paying more and more attention to their cyber-hygiene — which is great to see, but dramatically undermined when we see that the driving forces are often unintentional in nature.
The 2017 Equifax breach that exposed 146 million customer records was traced back to a mistake of employees failing to follow security warnings and code reviews in the course of implementing the software fixes that would have prevented the breach.
Data from 2016 and 2017 pertaining to overall unauthorized exposure of regulated data backs this up, finding 92% of all incidents and 84% of all data breaches were unintentional or inadvertent in nature.
While these are all “cyber” in nature, companies have to further invest in training, compliance, accountability and tracking measures — as their biggest threat resides inside their own walls (be they physical or metaphorical).
The types of errors that can occur — whether considering the application (financial, operational, product, etc) or cause (data entry, intrinsic, environmental, transpositional, etc) — are variable, plentiful and easily combined. What they all have in common, however, is that they are nearly always avoidable — and always costly in some way shape or form.
Companies need to be actively formulating error reduction strategies, monitoring them, evaluating them and refining them. While we must acknowledge that errors and mistakes are inherent in life and in business — we also cannot allow that to invite complacency or inaction.
I reference a great article by Health Leaders Media in illustrating what that complacency would look like, and why we should always be looking for new and improved methods of reducing and managing errors in our business. In that article, they explored what some key aspects of American life would look like if we embraced 99.99% accuracy.
- O’Hare Airport would experience 2 unsafe plane landings daily;
- Each week, 500 surgical operations in the US would be done incorrectly;
- 50 newborns would be dropped, daily, by doctors at birth;
- Each hour, 22,000 cheques would be deducted from the wrong bank account; and
- More than 114,500 shoes would ship each year mismatched (gasp!).
While 99.99% seems pretty darn good, it’s a strong reminder that none of us should (or in some cases can) be satisfied with “good enough” — we can always do better.
Humans are, of course, the centre of your business — and mistakes happen. But you can take steps to reduce how much manual entry is happening in your business, to increase your visibility into data and you can implement checks and balances to ensure small mistakes don’t cascade into colossal disasters. A leading strategy for doing this is in evaluating the implementation of digital tools in your business — automating key tasks and processes, allowing your team to spend time on what matters. For help identifying what processes to evaluate and how to take those initial steps in your business — I’d invite you to check out some of the work we’ve done for other great brands like yours, or to slot off some time to pick my brain. I’d also welcome any examples of error impact or error reduction strategies you’ve seen successful in your business.