Financial institutions are experiencing an increase in the number of defaulting customers and are beginning to feel the consequences through falling profits brought on by the rising costs of bad debt. Additionally, the customer experience is increasingly strained as banks progress more and more customers through the collections process. Not only have financial institutions increased their provisions for bad and doubtful debts, most have adopted a much more stringent lending policy: it is a downward spiral.
Lenders are also coming under increased pressure to assist their customers. Based on the 2007 Independent Review of the Banking Codes and following the recommendations of the independent reviewer and response by their Code of Sponsors, there are a number of new revisions that place the onus on the lender to demonstrate ‘treat customers fairly’ principles.
Under these new codes of practice, lenders are required to use their information technology systems for the pro-active identification of customers who may be falling into financial difficulties. Institutions should then contact such customers in order to emphasise that, if they are in difficulties, a ‘sympathetic and positive’ approach will be taken and to provide details of independent, free, money advice agencies.
In this instance, the lender should also suspend any automatic offers that may further the customer’s difficulties, such as credit limit increases and, furthermore, weed out such customers from credit marketing lists.
How can the retail banking industry predict which customers are likely to get into difficulties before they start to default and identify and keep those customers who, in the long term, are likely to spend more money with them? A strategy is needed to deal with this and pre-delinquency management is the upcoming methodology. Anyone can spot a customer who stops making payments, the trick is to spot the ones who are about to stop making their payments.
It would be great to have a crystal ball to identify those customers that were simply going through a tough time with their finances but in the long term would remain high-value customers, and those that would fall delinquent and become a bad-debt statistic. Pre-delinquency management (PDM) is discussed here as a strategy, methodology and technology that emulates that crystal ball.
Predicting delinquency
Organisations are now able to put a strategy in place to identify customers likely to fall into bad debt before they do so. There are many indicators which, when analysed together through sophisticated modelling, can help predict imminent payment defaults.
Modelling algorithms are set to work on customer data to identify changes in customer behaviour. The resulting score card identifies the propensity to default.
Through analysis of existing behavioural and transactional customer data from all sources and channels, strong predictive analytical models can be created to indicate the most likely future outcome. The key here is the early identification of potential delinquents.
Having developed the predictive models, speed of analysis is of the essence. As soon as possible after the customer data has been refreshed, the predictive models should be applied to the customer base as a whole. This will result in a list of customers most likely to default. The sooner the list can be created and utilised to facilitate initial customer contact, the less chance there will be for error.
As a result, many accounts will avoid moving into collections, returning safely to the sales and service phase and increasing customer retention periods. Alternatively, early identification and activity provides the option of managing out people who are always going to be bad debtors. Both approaches reduce costs and increase profits.
The focus of traditional pre-collections or early arrears activity has not allowed lenders sufficient time to have a significant impact on customer behaviour. A successful PDM programme focuses on early identification of stressed behaviour; that is, the period six to nine months prior to delinquency. This provides adequate time to get accounts back on track.
Over time, the models used to predict delinquency will need to evolve. As customer behaviour changes, in line with both personal and external financial pressures, the customer attributes and values that influence the predictability of the models will also change. The flexibility and speed of such change will have an impact on the overall long- term success of any PDM programme.
Once customers most likely to default are identified, the next step is how best to handle them. The secret is recognising how to profitably manage different types of customers through this phase, while ensuring fair treatment. One size does not fit all. Strategies need to be specific, appropriate and dynamic. Managed properly, treatments have the ability to not only retain profitable customers but also to drive up lifetime values. This is the result of a positive customer experience generated through supportive PDM interactions.
A single view
With a single view of each customer across multiple channels, up-to-date, real-time information optimises the value of every customer interaction. This will also enable re-evaluation of circumstances to ensure that every interaction and treatment is validated and relevant – be it on the internet, on the telephone or in the branch. It is beneficial to contact the customer first to determine the full extent of their circumstances. This, in turn, may affect the appropriate remedial activities.
Treatment strategies need, therefore, to be handled with skill and experience in order to generate the best customer reaction possible. These could include the automatic generation of a letter or a telephone call to offer help, such as changing credit limits or suggesting a new payment schedule.
Whatever best next action is chosen, the treatment should be managed centrally with intelligent prompts to call-centre advisers to guide them through these actions. Outbound interactions across multiple channels may also be appropriate, depending on the level of pre-delinquent risk. Intelligent software also needs to be fully integrated with outbound marketing to ensure the suppression of inappropriate new offers.
New framework
A technology solution must be designed to support the delivery of a flexible, multi-channel PDM solution that can either operate as a stand-alone pilot application or enhance existing sales and service platforms. This allows the following ideals to become reality:
The solution needs to be based around four distinct PDM actions:
This provides a solution that allows lenders to predict potential delinquency, help customers to manage their debt and reduce write-offs as a result.
The significant reduction in the provision for bad debt and less stringent lending policies is the lender’s dream. The dream can become a reality if pre-delinquency management strategies and tools are in place. If delinquency is prevented or at least reduced using a pre-delinquency solution, there are tangible benefits for organisations and their customers. These include adherence to the new guidelines in the Banking Code and the enablement of ‘treat customers fairly’ principles.
The provision for impairment can also be reduced, as can the number of accounts moving into collections. This allows revenues to be maintained by keeping profitable accounts alive. Furthermore, the lifetime value of customers can be increased, simultaneously improving the loyalty of delighted customers – who will be likely to tell their friends. Meanwhile, customers that are likely to remain bad debtors can be identified and handled before the debt has amassed.
