How AI is empowering banks in the race for new customers

From a business perspective, Canada and Australia often look to each other as a benchmark. Both have relatively small populations which have continually grown through immigration. They’re enormous geographically, with many people concentrated in a few major cities. And they share the challenges of these characteristics when building and maintaining key infrastructure, such as for transport or communication. 

The similarities extend to banking, where both countries are led by a small group of heavyweights while a second tier of challengers, online and regional banks compete for customers – particularly those new to the country or switching banks.

As Australia’s banks face an increasingly competitive market, could the answer be to borrow with pride the solutions that have worked across the North Pacific?

Through 2020 and 2021, Deloitte Canada developed and deployed Acquisition.AI – a cutting-edge customer acquisition toolset. Combining a wide range of rich data assets, primary research on banking preferences and the latest machine learning techniques, it helps banking clients target opportunities to increase their new-to-bank acquisition rates.

And it’s now being used right here in Australia.

Acquisition.AI draws from over 4000 data points relating to the country’s population. These include:

• Demographic data such as age, gender and family mix
• Deeper insights such as spend behaviours, financial product holdings and credit bureau data (aggregated to small geographical areas)
• Unique attributes such as attitudes to risk, reasons for changing banking relationships and priorities when selecting a bank brand and its products.

Importantly, these data assets were collected and modelled using machine learning to cover the country’s entire population, not just a single bank’s existing customers. And as the data covers the whole population, it doesn’t rely on people being active online or revealing possible intent based on their use of social media.

This whole-of-population data asset is leveraged in a series of machine learning models that predict the number of people actively changing banking relationships in any micro-market area: the ‘accounts in motion’. 

Further, through the richness of the data and sophisticated modelling, Acquisition.AI offers detailed profile descriptions of the customers who are in motion and what they are looking for in their banking provider, products and features. 

Finally, integrating our banking client’s data into Acquisition.AI creates a powerful ‘extractability’ machine learning model. This model identifies the micro-market areas which contain the bank’s best opportunity – the ‘next customer’ or ‘next home loan’ view of the market.

As part of this process, the model identifies where to find the three key factors needed for successful customer acquisition. It’s the intersection of people who are actively looking to change their banking relationship, are of high banking value and are highly likely to specifically select our banking client’s value proposition – the ‘extractable’ customers. 

Each banking client has their own unique extractability model that is aligned to their specific strategy, identifying their best opportunities, benefiting from a whole of market view of movements (not just their own historical acquisition) and a deep understanding of what extractable customers care most about.

Working with several leading Canadian banks, Acquisition.AI has delivered a detailed view of the most ‘extractable’ micro-market areas. Our Canadian banking clients have used this insight to create digital and physical audiences for hyper-targeted acquisition campaigns, which outperformed their existing acquisition strategies by up to 100%. These new-to-bank customers brought value not only in their numbers, but also in their higher profitability and tendency to bring more than one product as they were acquired.

Against this growing backdrop of success for Deloitte’s banking clients in North America, bringing Acquisition.AI to Australia is an incredibly exciting prospect for our clients.

The rich data collected in our Australian research, as highlighted in the second blog of this series, has been combined with rich demographic data, financial product holdings, credit data, and consumer behaviour and movement data. This has created the data asset needed to bring Acquisition.AI and its AI-driven model to the local market.

This powerful capability will empower our banking clients to target where their next home loan or everyday banking clients will come from. They will be able to execute on these insights through cross-channel campaigns and have a rich, detailed understanding of what parts of their product and value proposition these new customers care most about.