The term ‘machine learning’ has become a buzzword in the past year or so. Billions of dollars are invested in artificial intelligence (AI) technology each year – between $26 billion and $39 billion in 2016 according to the McKinsey Global Institute Study – of which nearly 60% went into machine learning.

Financial services are leading early adopters, together with high tech and telecom. McKinsey predicts that these three industries will continue to lead AI adoption in the next three years.

This technology is becoming very sophisticated, and early adopters are starting to see clear benefits. In the financial services industry, AI adopters with a proactive strategy achieve approximately 12.5% higher profit margins than non-adopters.

Xero, a cloud accounting platform for small businesses, introduced machine learning to their software earlier this year.  


What is Machine Learning?

Machine learning is a subfield of artificial intelligence (AI). According to Arthur Samuel, a pioneer in the field of AI who coined the term ‘machine learning’ in 1959, machine learning gives “computers the ability to learn without being explicitly programmed”.

Machine learning uses neural networks that are designed to function the same way as a human brain. When algorithms process and analyse enough data, they start to recognise patterns, make connections, and classify it according to the elements it contains.

Computers aren’t as smart as humans, but because they can process data much quicker than people can, they’re very fast and generally very accurate in their conclusions.


How Machine Learning Is Impacting Finance

Accounting software is getting smarter, and it is already performing tasks that previously required human intervention. Repetitive, manual, and tedious tasks are eliminated so that bookkeepers and business owners can now spend less time on keeping their accounts up-to-date, and more time on other important tasks.

Here are some examples of specific tasks that this technology already affects:

  • Accounting apps learn invoice coding behaviours and suggest where transactions should be allocated. For instance, if the sales clerk usually allocates a product to a particular sales account, next time that the sales clerk adds that item to an invoice, the accounting app will automatically allocate it to the right account. It also looks at errors that bookkeepers and accountants fix. For instance, if the business owner allocates something to the incorrect account and the accountant fixes the error, the accounting app will take the accountant’s selection as the right one.
  • Bank reconciliations are automated. Again, technology learns from previous allocations and account choices and then makes the right recommendations for new bank transactions.
  • Banks use AI chatbots to help customers resolve common queries. Chatbots from accounting programs, like Xero, let you query the latest financial data, like how much money is in the bank, when a certain bill is due, who owes you money, and it can even connect users with Xero advisors from their directory.

Additionally, this technology will also affect the work of auditors in the near future. Currently, auditors only study a select sample of transactions. They employ large teams of accountants who work overtime to finalise audits by deadline. The vast amount of transactions that flow through companies limits the number of transactions that auditors can inspect manually.

According to PWC, in the future auditors will be able to audit 100% of companies’ financial transactions. Machine learning algorithms will process and review the data, recognise anomalies and compile a list of outliers for auditors to check. Instead of spending most of their time checking data, auditors can apply their skills to investigating and deducing the reason behind a pattern or anomaly.


Why This Is Good News for the Industry

Working with non-accounting-savvy small business owners, you’ve probably seen that they find it challenging to keep their books up-to-date and to remember where to allocate transactions. This causes lost time and unnecessary errors which you, as their accountant, must correct later on. It also means that their accounts are never accurate, leaving them in the dark as to their financial performance. Machine learning technology that automatically suggests or completes accounting codes eliminates errors and saves a lot of time.

Apart from the time-saving component, if auditors can check a company’s every transaction, their financial information will be more accurate, and auditors can spend more time analysing the financial data to give better advice to their clients.

Technology can do the heavy lifting, number crunching, and report compilations, while accountants focus on judgment-intensive tasks. Machines cannot think like humans. They do not have our emotional intelligence either. Technology makes accountants more efficient and productive so that accountants can interpret data to provide better insight and business advice to their clients.


How Does the Future Look for the Accounting Industry?

According to CLSA, an investment firm headquartered in Hong Kong, AI will create more jobs than it ‘destroys’, but the transition will be painful. Over the next five to ten years we can expect to see significant changes in the finance arena and accountants will need to learn to adapt quickly.

According to Accenture, by 2020 more than 80% of traditional financial services will be delivered by cross-functional teams that include AI. While AI will crunch data, look for anomalies, and compile reports, human accountants will analyse data and provide informed recommendations to their clients based on their experience and knowledge.

If you’re not a tech-savvy accountant, don’t let this scare you. It’s just a different way of thinking, one that is easily learnable. Start with a cloud accounting program like Xero to get your toes wet. Once you get used to this, you’ll find it easy to adopt other intelligent technologies and to use them to your advantage.

Watch this interesting Ted talk by chess legend, Garry Kasparov, about how we should not fear intelligent machines, but rather work with them.


Source: McKinsey Global Institute, Artificial Intelligence, The Next Digital Frontier:

Want to learn more about marketing for number-crunchers?