After a year of development, Xero is nearly ready to pilot a machine learning system that will speed up invoice preparation by suggesting account codes for invoice items.
For example, "Sales - Labour" would be suggested for an invoice for time spent on site, reducing the risk of it being inadvertently applied to "Sales - Materials."
The system will adapt to the processes used by the business and its advisor.
In early testing, after just one invoice, machine learning techniques understood invoicing behaviour in general better than Xero experts, according to Xero officials. By the fourth invoice, early implementations were accurate over 80 percent of the time, and by the 50th they consistently reached over 90 percent.
"Simply talking about machine learning isn't enough. The truth is, it's really difficult to do beautifully. While many claim to use machine learning principles, actually implementing them across complex platforms to address customer's real needs takes intricate, specialised knowledge of the technology, the platform and the user," said Xero business product lead Luke Gumbley.
This feature is seen as the first step towards providing a personalised assistant for small businesses and their accountants that will reduce administrative chores, prevent mistakes, and enable them to spend more time growing their business.
"Our machine learning automation will give each Xero business and their advisor the unique insights and processes they need to cut the burden on manual tasks and let them spend more time growing their business. We're not asking them to learn something new - we're saving them time and money by ensuring the system learns from and for them," added Gumbley.
"We expect the wider adoption of machine learning to lead to further insights and improvements that will help small businesses, their advisors and other partners fundamentally change the way they work. In the future, Xero will be able to use aggregated, non-identifying data to identify trends, patterns, behaviours and industry insights. This is just the beginning."
It doesn't necessarily require AI, but it could be useful to learn how your business compares in various ways with similar enterprises. For example, discovering that you were spending significantly more on electricity than your peers do would likely prompt checks on your energy efficiency and the deal you have with your electricity supplier.
The machine learning automation for invoicing will be available to all Xero users later this year after it has been tested by a group of small business customers and their accountants.