AFG pilots new AI platform

The new Red Alert warning system has been integrated into AFG’s SMART marketing platform to help identify residential home loan clients who might be thinking about changing their current situation.

It builds on traditional email automation (for example, notifications for end of fixed term periods) by utilising AI to predict whether a client may be looking to change their loan or circumstances. The update intends to help brokers more effectively develop their communications in a bid to improve retention.

According to AFG’s head of marketing, Jon Amery, the system was built after undertaking workshops with brokers and staff to ascertain some of the key triggers that may lead clients to reassess their residential home loan arrangement, which was then overlaid against AFG data and built into an algorithm.

The initial algorithm correctly identified more than a third of all borrowers who would eventually leave their broker, he said.

Mr Amery noted that while there was no single demographic or data point that could reliably indicate if a customer is more likely to churn, the algorithm could identify ‘micro-groups’ of customers that met several key variables/patterns of behaviour that could be thinking of changing their loan or circumstances.

Those patterns are then compared with existing customer data points to identify which customers are the most likely to be making a change in the immediate future, and why.

The results are then automatically displayed for brokers so they can reach out to the customer or use the SMART Marketing Automation tool to trigger a scheduled communication.

Mr Amery told The Adviser: “Brokers are alerted/provided with a list of Red Alert customers directly into FLEX and then have the option to automate an email or to contact them directly.

“Because this is an AI algorithm, it’s not necessarily one or two big data points that will flag an alert, rather a combination of many micro data points. 

“This is the big difference between triggered communications that are fairly obvious, such as a customer coming to the end of a fixed rate, and using AI to predict a customer’s behaviour.

“An example might be a combination of factors that indicate a new home buyer is looking to move to a new home after ‘getting in the market’ years prior,” he explained.

“There may be hundreds of data points but some of these could include: number of years since purchasing a property, increase in suburb property value by X per cent, age or change in age of buyers and family unit, suburb profile/change in suburb profile, and the equity in property.”

However, Mr Amery added the new system is designed to “augment the service that brokers provide their customers, not replace them”.

“It will make the brokers more efficient, allowing them to have more meaningful conversations with their customers, and help brokers anticipate their evolving needs – well beyond when their home loan was originally settled,” he said.

“The Red Alert system also enhances competition and choice for customers by suggesting credit options at an appropriate time. This helps brokers to continue to act in their clients’ best interests.”

According to AFG’s head of analytics and insights, Alex Maund, the system could reach 60 per cent accuracy as it goes through live testing.

AFG has said that a subset of “randomly allocated” AFG brokers will be using the new AI SMART platform over the next few months to “continue to test and optimise while running it alongside the existing algorithm”, before it rolls out to all SMART brokers next year.

[Related: Why the mortgage broking industry needs to embrace, rather than fear, AI]