Businessmen teamwork meeting to discuss the investment.

Underwriting uncovers the big picture for lenders

It lets them make affordability decisions and spot potential vulnerability with a detail and nuance that leads to better credit decisions. But underwriting also costs money and slows down the consumer journey.

Implementing machine learning and data science into underwriting is now revolutionising the way lenders interpret large volumes of transaction data to help acquire and protect consumers.

Faster, better underwriting

Banks have always had the opportunity to use their customer’s own transaction data to inform their lending decisions. The problem is that this analysis has often been time-consuming and laborious, frequently based only on what the bank knows about its customers using its own data, rather than basing it on a holistic understanding of their customer’s financial behaviour across different accounts.

Open Banking has changed this. Lenders can now use their customer’s own transaction data, and also import transaction data from other lenders with who the customer has a relationship. This can be used to create a more complete picture of what a customer can afford.

This is why Experian has developed Categorisation as a Service (CaaS) – a proven, automated transaction categorisation tool that helps underwriters see the big picture fast.

With it, lenders can categorise transactions in a customer’s current and credit-card accounts into over 180 different categories of income and spending to provide granular detail. And the picture can easily be segmented into timeframes of 30 days up to 12 months, revealing key trends.

This data can then be summarised into a set of insights around income, account turnover, balance behaviour and expenditure.

CaaS can be deployed by a lender on-the-cloud to analyse data in batch using their own data or deployed in real-time using Open Banking to analyse data from all a customer’s accounts where they’ve consented to share their data.

So CaaS helps underwriters answer key questions fast:

  • Does essential spending account for a high percentage of income?
  • How volatile is the customer’s income?
  • What other credit products are being repaid?
  • Has anything significant changed recently, in income or expenditure?
  • … and hundreds more

Find out more about Categorisation as a Service and how it can support your business

Get in touch

Underwriting challenges in 2021 and how CaaS addresses them

In March 2021, Accenture reported on the key challenges facing underwriters today:

  • Pressure to significantly reduce costs while maintaining or improving quality
  • Underwriting platforms – in general, underwriters are unsatisfied with them
  • Evolving customer needs and expectations – especially the desire to be wholly digital
  • Underwriters are generally not well equipped to evaluate, select, and integrate new data sources from the massive amount of information now available to them.

CaaS addresses these problems – providing fast, robust insights in a single, end-to-end digital platform. It makes things simpler for consumers and quicker for underwriters.

Underwriters have responded enthusiastically. “They’ve embraced it,” says Experian’s Director of Consumer Information, Richard Sunman. “Having realised the speed and reliability that CaaS provides, lenders are now operationalising its use. This is freeing time for them to capitalise on the insight that transaction analysis reveals to help them grow their business and protect consumers.”

How does CaaS work?

CaaS is powered by a series of machine-learning algorithms, which can run over clients’ own transaction data and/or Open Banking sourced information.

The algorithms are trained to interpret credit-card and bank transactions and accurately assign them to one of over 180 different income and expenditure categories. It reveals where the money is coming from, where it is going, and any significant financial changes.

What can it show underwriters?

For underwriters, CaaS offers a precise understanding of:

  • true income, its sources and stability
  • regular savings or pension contributions
  • level of spending on basic household essentials, mortgages or rent
  • patterns of expenditure associated with specific credit risk or conduct-risk behaviours
  • any reliance on short-term, high-cost credit
  • gambling-to-income ratios
  • engagement with debt-collection agencies
  • … and much more

No more asking for extra information

These automated insights let underwriters work wholly digitally for the first time. Analysts can cut the cost and time spent in going to customers for data. They can authorise new credit products faster and ultimately make better-quality decisions.

Proven benefit to lenders

Up to 6.1 million mainstream credit applications are referred for further investigation each year. By cutting the time highly technical staff spend on each case, lenders save money – while still making the best decisions for their customers.

Proven benefits include:

  • Extending more credit without increasing risk appetite: Using CaaS, clients have been able to accept more applications, offer more products to existing customers, or extend their limits – without increasing risk appetite. One leading Irish retail bank has confirmed a 5% increase in the volume of Accepts for loans in the first 3 months of using CaaS
  • More accurate credit-risk management: Work with clients has shown that CaaS makes decision models 10-15% more accurate.
  • Fewer credit decisions going bad: In acquisitions, CaaS makes affordability calculations significantly more accurate. One client has cut acceptances that go into debt by 40%.
  • Driving innovation: CaaS underpins Experian’s Boost proposition, the UK’s most successful Open Banking service to date. It helps consumers boost their credit score, access mainstream credit, and avoid the ‘poverty premium’ of paying for high-cost credit.
  • Protecting vulnerable customers:  CaaS can be used to identify and protect vulnerable customers by revealing important indicators in their spending behaviour. This may include dipping into an overdraft or frequent short-term loans to buy essentials. For gig-economy workers’ income – that may come from multiple jobs or seasonal work – CaaS can be used to validate income and properly assess what they can afford, opening up access to credit.

As Richard Sunman summarises:

“The important thing to know is that we’re simplifying and accelerating the analysis of complex data and using this to reveal new insight from which better decisions can be made for consumers and lenders.”