Story image

Xero launches machine learning automation to improve coding accuracy for SMBs

11 Oct 17

Building on the success of its first machine learning project, Xero announced the next step in its journey towards code-free accounting – the automation of account codes and bills for all small businesses and their partners.

With more than 500,000 bills entered into Xero every day, and each line of a bill edited individually, the automation of billing account codes is set to transform accounting practices, ensuring greater accuracy and reducing the time small businesses spend creating bills.

While billing is the second-most commonly used feature of Xero, it has the second-highest rate of defaults, with every small business using the system differently. 

Fifty percent of businesses use 10 or more expense codes, while others create their own codes, which means that information is often entered incorrectly. 

Xero’s new artificial intelligence system will consider each individual business’s characteristics, then recommend account codes based on what it has learned.

“We set out to further develop our machine learning program to reduce the number of mistakes being made when creating bills. In doing so, we’ve created a system that not only learns from the individual needs of our customers, but can also make objective decisions about which account their transactions should be coded to,” said Andy Neale, Xero data science and automation head.

With accuracy at the forefront of Xero’s latest development, the account code automation is rolling out to all Xero users, with customers continuing to code their accounts for bills as normal.

A minimum of 150 bills is required to be entered and as more bills are entered and accepted or corrected, the better the suggestions become.

“Account coding has the potential to impact both small business finances and reporting. So, when applying machine learning to Xero’s billing system, we needed to ensure accuracy was at the forefront of any changes we were making. While we’ll only know the true impact of the technology when people start using it, on average, we know we can achieve more than 75% accuracy after 150 bills have been entered.”

“Account codes can be time intensive task for small businesses. We are using machine learning to simplify the system overall, so that more time is spent growing the business, rather than on low-value, administrative tasks,” Neale says.

Commerce Commission report shows fibre is hot on the heels of copper
The report shows that as of 30 September 2018 there were 668,850 households and businesses connected to fibre, an increase of 45% from 2017.
Dr Ryan Ko steps down as head of Cybersecurity Researchers of Waikato
Dr Ko is off to Australia to become the University of Queensland’s UQ Cyber Security chair and director.
Businesses in APAC are ahead of the global digital transformation game
“And it’s more about people and culture - about change management - along with investing in the technology.”
HubSpot announces fund for 'customer first' startups
HubSpot is pouring US$30 million (NZ$40 million) into a new fund to support startups that demonstrate ‘customer first’ approach of not only growing bigger, but growing better.
Mac malware on WatchGuard’s top ten list for first time
The report is based on data from active WatchGuard Firebox unified threat management appliances and covers the major malware campaigns.
LearnCoach closes $1.5m seed round
The tutorials are designed for students who want to learn NCEA subjects but can’t make it to a physical classroom.
Bin 'em: Those bomb threat emails are complete hoaxes
A worldwide spate of spam emails claiming there is a bomb in the recipient’s building is almost certainly a hoax.
Marriott sets up call centres to answer questions on data breach
Marriott has released an update on the breach of the Starwood guest reservation data breach which affected 500 million guests.