Case study: How finance brands use data segmentation to improve the customer journey

Case study: How finance brands use data segmentation to improve the customer journey
Ben is an expert in credit risk management, providing analytical consultancy to clients in financial services, telecoms and utilities. With over 18 years’ experience, his expertise spans credit scoring, loss forecasting and stress testing, provisioning, portfolio monitoring, value modelling and the analytical components of capital management. As well as deep subject matter expertise, Ben has a natural talent for communicating complex information with clarity. Ben runs the Consulting Risk Practice at Jaywing and since joining in 2001 has led analytics teams for a number of clients in the UK, Europe and Australia, including a variety of substantial engagements in the UK banking sector.


As consumer confidence returns, the rise in people looking for credit provides lenders with a real opportunity. Ensuring those consumers can afford the credit remains top of the agenda, with recent regulatory changes designed to provide stronger protection for the consumer.

The key is for lenders to make the checks needed to satisfy the regulators and protect the customer but to do it in a way that avoids damaging customer experience and driving customers away. So, how can credit lenders get the balance right?

The FCA took over responsibility for the regulation of consumer credit in April 2014 and has since introduced a number of changes to the Consumer Credit Sourcebook (CONC). These rules traditionally applied to financial institutions providing products including loans, mortgages, credit cards, and overdrafts; however any provider of credit must now fall in line.

While the guidelines imply that the level of assessment around affordability can vary depending on the product type and the amount of credit being provided, the lack of detail in some areas is leading to confusion among many credit providers on how best to implement these changes. Many are finding that the guidance is unclear on the specific information they need to capture so making sure they collect the necessary level of detail without putting off the borrower is crucial.

Customer-centric process

The FCA has indicated that the consumer should be at the heart of the decision, regardless of the size of credit. This is something both credit providers and the regulators agree on. Today, consumers have a plethora of choice and their lives are busier than ever.

Making sure the transaction is clear to understand and easy to complete is important and lenders must balance the need to obtain information with minimising the drop-out rate and maintaining a good customer experience. Having different customer journeys for individuals who need a different approach to credit is the first step.

Identifying which customers require a light touch and which require more rigorous checks to ensure affordability is vital. CONC focuses on the “proportionality” of an assessment for creditworthiness, which may include a number of elements such as the type and cost of credit, but also on the past, current and future financial position of the customer.

Appropriate segmentation of customers and the modelling of disposable income allows lenders to satisfy both regulatory requirements and internal department needs, as well as ensuring a bespoke and smooth journey for each customer.

Consideration conundrum

Whether the consumer is applying online, in store or over the phone, working out a customer’s disposable income is the key to assessing affordability. But this is not as simple as it may first appear. For example, calculating expected income can be challenging if bonuses and overtime are to be properly considered.

Should you consider the household income or just the individual’s? How can you be sure their salary is accurate without asking for copies of payslips or bank statements? And is this overkill for some products and lending amounts anyway? The same can be said for working out expenditure; assessing all an individual’s outgoings could mean asking a lot of questions that negatively affects the customer journey and inevitably causes the customer to abandon the application process.

Many lenders now choose to work with average positions and refer all customers to an underwriter where their circumstances appear to sit outside of this average. This can be time consuming and expensive to manage for the lender and again has the potential to negatively affect the customer journey.

Statistical models

So how can you ensure your lending decisions, operational practices and resulting customer journey works best for both lender and customer? You can use a combination of customer segmentation techniques and statistical models to identify which customers should be sent on which journey. Then you can use external data sources to validate the information they supply and ultimately assess their affordability.

When it comes to validating income, there are a number of external data bureaux who offer income data validation. Where a customer’s answers don’t match the expectations based on the data from the bureau, additional checks can then be made. This means that the lender only incurs the operational cost when necessary and the customer isn’t asked for copies of bank statements or payslips for a relatively modest level of exposure. With regard to expenditure, using customer demographic data, such as that published by the ONS, means it is possible to model a direct relationship between income and expenditure.

Getting the balance right between satisfying the regulators and getting the customer journey right can seem like a bit of a challenge. But, by undertaking a full audit to understand the customer journey across all touch points, online, phone and in store, you can identify areas that can be tightened or where the onus on the customer can be reduced through external data sources to validate and check responses.

By having clear journeys established not just for different product lines but for different customer segments, the appropriate questions are asked at appropriate times, making the whole process simpler and easier for the customer, without compromising them by lending irresponsibly. The lending is operationally slick, meets both regulatory requirements and its own risk appetite and is ultimately more likely to make the sale.

View Comments
Leave a comment

Leave a Reply

Your email address will not be published. Required fields are marked *