Brought to you by
Aurelia logo
Aurelia are the automation and software experts for accountants and finance leads. We want to to...
Save content
Have you found this content useful? Use the button above to save it to your profile.

Tech won’t save you from accounting mistakes

13th Feb 2023
Brought to you by
Aurelia logo
Aurelia are the automation and software experts for accountants and finance leads. We want to to...
Save content
Have you found this content useful? Use the button above to save it to your profile.

Been there, done that. Closing the books at the end of the month and realising something doesn’t add up. Yes, there’s a mistake in the data or calculation. 

What you feel is perhaps just frustration- was it your quick fingers? Was it a document that the client hasn’t sent? Or maybe it was your tech set up that has failed you. Cause yes, that happens more you’d like to. 

Whatever it was, now you’re facing a couple extra hours in front of the computer screen trying to figure out what happened and how to fix it while your friends have gone out to have dinner already. This article is for all the tech-savvy accountants who rely on their tech stack to minimise mistakes yet still face them on a weekly basis. So what can you do to make your tech stack more fool-proof?

Tech won't save you from accounting mistakes | Aurelia on AccountingWeb
Andrea Picquardo // Pexels

Even with software, mistakes happen 

So where are the most common places where an ‘ooops’ can creep in? Here’s what we observed - even though it’s not a comprehensive list.

Data extraction & data entry

This usually applies with data from PDF invoices, scanned receipts being made digital. Whether you retrieve this data manually or ask a piece of software to do it for you, things can go wrong. It could be a small typo in the name of the vendor, or software extracting literally every information from the invoice - forcing you to waste around 10 to 15 min matching it to the correct fields, it’s still time spent on correcting rather than doing actual work. A small typo probably won’t cause an avalanche (although it may raise eyebrows with some auditors), the wrong amount extracted from the invoice usually leads to incorrect tax being paid. 

Closely related to the above is data entry. Manually typing in data into either accounting software or bank to make payment happens more often than we’d like to see with this level of technology adoption. As in the example above, the software may fail in extracting data from the invoice so you need to do it yourself and with quick fingers a £2,000 can turn into £3,000. Same when manually entering payee details in the bank. While most banks have introduced extra safety nets to make sure you always pay the right person, one wrong digit in the account number and the £10k invoice was just paid to some unknown person in Worcestershire. Good luck getting it back. 

System failures

This is often overlooked. You may feel like you have the best tech stack in the world and then suddenly your client’s bank account connection to Xero breaks and your client cannot get it to work. Now you are a couple of days behind with work. 

Another one we love to discuss is Zapier. We see more and more accountants choosing it to integrate their tech stack. The problem is though, Zapier is not designed to handle financial data. Unfortunately, while not an official data the company would publish,  >20% failure rates on workflows are not uncommon from our conversations with financial controllers and accountants. It’s an even bigger problem when these failures lead to a sale or a purchase not getting recorded in sales, or giving you even more work to clean up after a failure because there’s no way to retry just a step in a Zap. What’s even more insulting is that if you want to retry a zap, it costs more than a regular Zap run!

Another issue with Zapier and a lot of other similar systems is that when things go wrong, you won’t know what went wrong and how to fix it. So on top of data clean-up you now need to also search through help articles, forums and set up the Zap again. 

The real cost of accounting mistakes

We don’t have to tell you that mistakes in accounting cost you time. They unfortunately also cost you reputation and money.

The more time you spend fixing mistakes, the slower you are. And the slower you are the less time you have to provide the personalised service to your clients - something they really expect. And if you work internally, you won’t have time nor mental space for forecasting and providing expertise that the business really needs. You just won’t be seen as future-proof as you may want to be. 

And then there’s money. The best way to illustrate it is using  The 1-100 rule from George Labovitz and Yu Sang Chang. The idea behind this rule is that validating  data at the point of entry costs about a dollar. Then, fixing it later on costs 10 dollars. Doing nothing and failing costs 100. And you still need to spend time in all scenarios.  

And there are some real horror stories about how overlooking something can get your clients into trouble. Like this one about the overcharged VAT  because the company didn’t have a tax invoice among other things. Fixing mistakes can delay tax returns and HMRC will not hesitate to apply a penalty in that case too. There was a very loud case recently of tax returns being overdue for years but even a couple of weeks lateness can incur a couple 1% penalty fee!

And now think of your reputation if you were the accountant for those clients!

Not to mention the toll on your mental health and productivity, but also self-esteem from stress and frustration of dealing with stuff like this. 

So what can you do to make yourself a bit more fool-proof? 

Your tech stack can be more fool-proof

The less you rely on manual work the better and you know this. As mentioned before, we don’t doubt you use digital tools to improve efficiency and accuracy of your work but again, this is fallible. 

Tech shouldn’t be the first answer to becoming less prone to accounting errors. Start with the process. Identify the loopholes in your day-to-day work. Where do you still rely on manual work and manual data transfer? Could any of this be automated? And by automated we mean using a proper tool that’s designed for finance and accounting and not a workaround you need to spend time building or zapier-like, generalist connector.

Once you’ve identified the parts of your process where you still rely on manual work or rickety tools you should be ready to pick a tool that’d make you more resilient to accounting mistakes. When picking a piece of software, the features below are an absolute must for greater accuracy.

Well-integrated, end-to-end tool

You want something that will be able to complete at least one process in full, without relying on workarounds or additional plugins. Let’s take accounts payable. The pipe dream would be to have seamless flow from the moment of collection of an invoice all the way to payment or transaction happening and then being marked as paid in the accounting software. And we’re relentlessly trying to make this pipe dream come true - with Aurelia’s Invoice Inbox you can collect, process and pay. 


What we mean when we say a tool is robust is firstly that it won’t freak out if asked to handle more complicated use cases like applying VAT rates to different countries of sale or managing currency exchange. Robustness also means stability. All connections and workflows built in the tool should be reliable enough so they don’t randomly break and even if they do, you should be able to firstly know exactly what, when and how it got messed up, secondly how to fix it and lastly, re-run it retrospectively so that nothing is lost. 

A smart tool that learns

Firstly, mapping a human process one to one in a piece of software is a disaster. Let’s take an example of data extraction. The tool shouldn’t just take all the data from the invoice and make you decide what’s relevant and what’s fluff. It should be able to pick the relevant information as per your accounting software and discard what’s not needed. Obviously, technology, just as humans won’t get things right 100% of the time. The advantage software has over the human brain is it’s not getting brainfogged on Friday afternoon and it never makes the same mistake twice. You want a tool that learns, for example, applying the correct category so that over time, you can rely on it more. 

Final word

Even as a tech-aware accountant you undoubtedly still encounter mistakes in accounting on a weekly, if not daily basis. And they come from all places where you thought you already had it nailed. And then dealing with these costs you more than time, it’s also money and reputation. Can things be better? Can you protect yourself from accounting mistakes for good? Perhaps not completely but once you clean up your processes and identify the spots with manual work, you can rely on smarter, robust and well-integrated tools that are indeed available now to make your work more fool-proof and efficient. 

And if you still struggle with any particular part of your process or software that doesn’t make the data more accurate and just adds up work, give us a shout at [email protected] and we’d love to help!