An in-depth look into the current debt collection landscape and how Juris Collect addresses industry pain points.

Image credit: rawpixel.com on Freepik.com

Debt Collection: the most important part of trade

Debt collection has been an important part of trade ever since the concept of loan was introduced thousands of years ago. In fact, both loan and debt are concepts older than money itself, as they both occurred before the invention of money, during the time when bartering was the norm.

While origination may be the backbone of the loan life cycle, debt collection remains the most important facet of the whole process because it is the collection of interests through which profits are made. If a loan becomes non-performing and eventually turns into write-offs – forget the interest – there may not be a way to fully recover that loss again.

Current challenges

Debt collection wouldn’t face any issues if everyone cleared what they owed, on time and in full. But we wouldn’t be here if that was the case. Challenges in debt collection have always been around since the invention of debt. For starters, the human memory is fallible, and many would just simply forget to pay their debt, until they’re being reminded to.

Increasing NPLs during recession

It has been noted that the proportion of non-performing loans (NPL) for every loan disbursed has increased by more than 50% post-financial crisis of 2007 – 2008 and throughout the Great Recession. As the economy is still slowly recovering, it remains relatively flat, causing high unemployment rate, pressure on income, and rising level of consumer debt. All these lead to high and volatile delinquency rates worldwide. The World Bank’s report of ‘Bank nonperforming loans to total gross loans’ was at 4.288 percent in 2009. While there was volatility, dropping to 3.738 percent in 2012 only to rise to 4.162 percent in 2014, we are currently seeing a downward trend, at 3.448 percent as of 2017. Malaysia is doing particularly well in this case; we had a rate of 9.39 percent back in 2005 but it has been on a downward trend since, at 1.545 percent as of 2017. But globally as an aggregate, things could still improve.

Customer indebtedness and subsequent arrears have significant impact on financial institutions’ (FI) profitability. Rising NPL causes an increase in collection costs, bad debt write-offs, and provisions against loan losses – making the management of credit loss an utmost priority.

Cutting of costs

The spike in NPLs increases collection costs as FIs and agencies have to mobilise more personnel to deal with a higher number of delinquents. At the same time, this is tugging with the fact that the increase in bad debt loss and reduction in profit, reduces credit availability of FIs causing them to reduce operating costs as well.

Companies may now be limiting their hiring of field agents, or worse, initiating layoffs. Cost-cutting also tempts them to adopt a broad, blanket approach to arrears management. This is problematic as one size does not fit all. Should the blanket collection approach be too aggressive, it may risk estranging potentially profitable debtors who may just be going through a temporary period of financial troubles.

Old technology

Perhaps the most lasting impact of cost-cutting is on the technology, solution, and software used for collection. It creates dependency on outdated technology, not to mention the increased sluggishness in capitalising on significant jumps in new technologies.

A lot of collection companies are still sticking to legacy systems; most of them developed internally many years ago for specific credit or loan products. Because they’re highly customised, they have a limited feature set and lack the flexibility to extend beyond their original purposes for which they were designed. New features cannot be easily added by business users without the intervention of developers. Legacy infrastructure may also lack the scalability to handle spikes in network traffic, due to growing customers. Not to mention that there had to be different applications custom-built for different products, resulted in a bunch of different software to maintain. To top it all off, legacy solutions are probably still being maintained by the same software engineers who may not be able to fully hand over everything to the next generation of developers due to the incomprehensibility of legacy code for new coders.

Inefficient and ineffective processes

Highly manual processes in legacy code may also cause bottlenecks. These days, debtors have multiple debts with various creditors, and the earlier a creditor reaches a promise-to-pay (PTP) deal, the more likely said creditor will get paid before the rest. The further back a creditor is in the queue, the more likely their loan will end up as a default. Therefore, speed in processing customer debt information is imperative.

As FIs tend to delegate late-stage collections to third-party agencies, it was difficult for different organisations to coordinate their collection efforts due to their using of different systems. Coordination is important for various reasons including compliance with ever-changing government, law, policies, and industry guidelines.

The system has to be able to handle and process competing priorities. This is not just limited to prioritising the methods of contacting customers in a multichannel environment to maintain or improve customer experience during collection efforts. (Is it still effective to send letters to customers instead of the more immediate nature of SMS and phone calls?) It is also about prioritising the type of customers to contact first, in order to maximise collection profits or to reduce losses. All these while keeping in mind the reduced resources and increased debt climate we’re experiencing today, where new, efficient, and more effective strategies to collect debt is a must.

Inflexible collection strategies

If there is one thing that makes or breaks a debt collection software, it is in its collection strategies. Priorities can be thought through but as mentioned earlier, the cutting of costs may incentivise creditors to adopt a single collection approach and that is bound to be ineffective and inefficient. This is because no two debts are alike, and due to time and cost constraints, a lot of strategy ideas may go untested. The existing collection software may also not have the flexibility to easily integrate several strategies for testing purposes.

There are a number of perspectives on which the strategies could operate. While it is logical to first approach the collection strategy by the size of debt, it is instead better to use predictive models to estimate the potential collectable amount. Increasingly linear automation could also affect accurate assessment of the debt pool value – if the current strategy focuses on contacting debtors with low contactability and low willingness to pay, the many attempts to reach said debtors will be wasted, and that energy is better spent on contacting those with higher propensity to pay, for example.

Trends and what’s important

Before we dive into the features of Juris Collect and how it could address the aforementioned challenges, it pays to first identify what are the trends and what’s important to FIs too. According to The Corporate Executive Board Company (CEB) 2015 Technology Adoption & Investment Survey, the key value drivers for collections technology are:

• Reduce risks: 50%.
• Improve processes: 46%.
• Improve service and experience drivers: 23%.
• Comply with regulations: 23%.
• Keep up with competitors: 8%.

While all of the value drivers above are important, two of the most high-impact ones are to reduce risks and to improve processes. These two requirements should be kept top-of-mind when in the market to adopt a new collections solution or to upgrade an existing one.

Enter, Juris Collect

Juris Collect was one of the founding solutions of JurisTech when the company was formed in 1997, at the beginning of the Asian financial crisis. Our chief executive officer, See Wai Hun said in an earlier interview that we ‘started at the tail end of the debt value chain’ by way of offering software and services to help banks and lawyers manage litigation cases and take defaulters to court. The debt collection and legal recovery solutions proved to be ‘recession-proof’, as The Malaysian Reserve article noted. Our venture into debt collection and legal recovery with the creation of Juris Collect and Juris Legal became a huge success, the former of which turned into one of our tent-pole solutions.

Right from the start, we made Juris Collect with the goal to be an easily-but-fully configurable platform. Through this, various collection and recovery strategies can be implemented and tested for different segments of the customer base, to address the challenges detailed earlier. The system isn’t just designed for financial institutions; it is capable of handling collection operations from other industries such as utilities and telecommunications, both in public and commercial sectors. Juris Collect also comes complete with legal features such as civil, foreclosure, repossession, and write-off handling.

Stages in collection

Over here in JurisTech, we break down the debt collection process into seven chunks, here with their corresponding actions:

• First-time defaulter: special calling campaign.
• Habitual late payer: SMS reminders.
• Early delinquent: early-delinquent calling campaign.
• Late delinquent: calling campaign, rescheduling, restructuring, NPL managing, external agency.
• Write-off: write-off and write-back.
• Asset disposal: asset repossession and auction.
• Legal action: track letter of demand to execution.

Juris Collect modules

Juris Collect comes with modules for managing each stage of delinquency:

• Collection: strategise contact channels using predictive behavioural scoring.
• Recovery: manage late-stage accounts using proactive recovery strategies.
• Agency management: place hard-to-collect accounts in a fully collaborative framework.
• Legal recovery: recover debt by pursuing legal action including bankruptcy, writ of seizure and sale (WSS), judgment debtor summons (JDS), and winding up.
• Write-off: write-off accounts that have exhausted all avenues through interface-to-host and general ledger.

Artificial intelligence

Early on, we found that artificial intelligence (AI) and predictive analytics are going to be the biggest differentiator in the debt management space. The longer an account goes delinquent, the less likely it will be repaid, therefore it is imperative to identify potential defaulters as early as possible.

Juris Collect is powered by an AI-based, self-learning technology that also incorporates behavioural scoring. It helps you implement and test various collection strategies for different types of customers to achieve the most efficient use of resources. The incorporated behavioural scoring tracks customers’ payment and contact patterns to maximise collection efforts by recommending the right pre-emptive measures for different delinquency tracks.

With the AI-based algorithm embedded, Juris Collect makes better predictions over time as it absorbs more data, and overall helps you identify potential NPLs with up to 90% accuracy, and up to six months ahead.

Dashboard with analytics

The comprehensive customer data captured in Juris Collect with the help of AI are then populated on a powerful dashboard with customer analytics. Said analytics can be based on collection and contact activities, account movement and flow-through for the various delinquency buckets, plus agency and legal productivity analysis.

This graphical dashboard reporting ensures that data are presented effectively to maximise comprehension.

Web-based user interface

Part of the ease of development and deployment is due to Juris Collect’s web-based nature. In fact, most of JurisTech’s solutions are web-based. The web-based software allows multiple developers to work on it at the same time while having testers examine the software without disrupting development. Compare that with an install-based software, where an updated code has to be uploaded back to the repositories before others can work on it. This significantly lowers not just development costs but operating and maintenance costs as well.

Being on the web, Juris Collect gives FIs the ability to automate, handle, and streamline the collection process in a single system, with information synced and accessible by anyone with a web browser and a connection.

Enhancing usability for collector effectiveness is also not overlooked here, because collection agents need to act fast according to the information they receive. The intuitive user interface with minimal learning curve allows anyone to quickly grasp it. The positioning of information in a single screen that makes sense, improves the user experience by maximising screen real estate and minimising clicks. All this is useful while gathering information, making decisions, and managing cases.

Customer-driven software

Unlike other collection solutions that are campaign-driven, Juris Collect is and will always be, customer-driven. It allows a 360-degree view of delinquent accounts, including information about whether or not an account treatment is effective after contacting the delinquent customer. Besides facilitating and tracking PTP installments, the system also manages both calls and collections. Putting it all together, the system remembers customers’ behaviours for the purposes of improving calling strategies.

When a customer information file (CIF) is pulled up, all key information is immediately displayed on the top row, like the overdue amount itself, months in arrears, last payment date, etc. The middle sections are for call history as well as for customer behaviour (aging buckets, payment behaviour, and activities since due). Actionable items are available on the same screen, like the contact status of the customer (promise to pay, uncontactable, etc), and the status of the case (rescheduling, repossession, etc).

Juris Workflow Engine

Juris Collect and almost every solution of ours come with the Juris Workflow Engine which is a key element of our systems. This core engine has a graphical user interface (GUI) with drag-and-drop features, built to be fast and scalable using PHP. This advanced workflow engine automates the collection process and the treatment of accounts receivable, in various stages. A typical workflow would look something like this:

Call customer > update outcome > decide on next action > skip tracing / PTP / refer to legal > monitor payment > repeat if necessary

Specifically for Juris Collect, Juris Workflow Engine allows for the creation and deployment of collection strategies. For each customer segment, strategies created consist of treatment steps and activities. These activities can be performed as an immediate step (run once immediately) or as a batch workflow (run daily, weekly, or monthly). Steps include contacting customers via phone calls, WhatsApp, SMS, email, and even the printing of reminder letters from PDF or Word. Of course, the frequency of these notices is fully configurable.

Self-cure prediction

Self-cure is a very important part of Juris Collect as it is the single, most-effective way to save on collection costs to avoid making less-than-optimal contact with customers. Usually in a pool of delinquents, there will be a significant portion of self-curing cases. Because FIs tend to not have enough personnel to contact all accounts, calling cases that are self-curing are considered unnecessary. Earlier, we mentioned that debtors have multiple debts with various creditors, and the earlier a creditor reaches a promise-to-pay (PTP) deal, the more likely said creditor will get paid before the rest. Therefore, it is imperative for FIs to quickly identify non-self-curing cases that are most likely to promise to pay, and contact them the soonest. Juris Collect could help with that, with its AI and predictive analytics.

Filtering delinquent accounts

Another important aspect of Juris Collect is its ability to filter delinquent accounts from core banking. Accounts from the core banking are loaded and inspected. Out of those accounts, those that have not exceeded one day past due (DPD) are given the all-clear. But for accounts that have exceeded one DPD, they will be sifted into the collection and recovery stages.

Segmentation and aging

Once an account is in the collection system, it undergoes segmentation based on criteria such as the type of loan product, previous records, tagging, geographical location, etc. These segments are then given different methods of treatment, based on a given customer’s DPD. For instance, the customer gets an email and SMS on one DPD, and a phone call on three DPD. Juris Collect supports automated generation, assistance for human agents, and mediation for outsourcing, on various channels for treatment:

• Automates in generation of: email, snail mail, WhatsApp, SMS, phone call.
• Assists in reaching via: SMS, phone call.
• Outsources to: debt collection agencies, lawyers.

Strategies

Juris Collect’s framework allows for a user-configurable profile, strategy, and treatment, to adapt to constant-changing needs. A strategy is a particular configuration of treatments (phone calls, emails), set to be deployed each in predetermined times, and applied to a specific segment of the customer base. A debt collection software usually houses multiple strategies, to cater to different customer segments. Strategies applied through the use of Juris Collect allow for an automated and streamlined collection and recovery processes.

The most fundamental version of the approach to design strategies is using the value vs collectability matrix:

• High value, low collectability: more aggressive strategy.
• High value, high collectability: field visits.
• Low value, low collectability: send reminders.
• Low value, high collectability: call campaigns.

Customer categorisation can be done through two methods, either with scorecards using the rule builder, or with predictive analytics using AI.

Each strategy can be broken down into several stages along the aging timeline. This allows the construction of different campaigns within a strategy, reflecting on real-world methods such as the automated reminder channels, assistance in facilitating the contact of customers, and outsourcing to agencies and lawyers.

There are typically two resolutions a customer could use to settle their arrears:

• Promise to pay (PTP): the customer receives a reminder and initiates an arrangement to pay (ATP) with the FI. After ATP, auto-reminders from Juris Collect will be turned off until repayment, or until the PTP date has passed.
• Rescheduling and restructuring (R&R): the customer submits a request to alter the terms of the loan before the rescheduling calculation happens. A credit report and debt ratio calculation are later made, before the approval and acceptance phase; all of which can be facilitated by Juris Collect as well.

Whiz and champion challenger

The strategy manager in Juris Collect, named Whiz, is a visual setup assistant that lets you make and maintain treatment strategies without requiring programming or configuration expertise. The campaign can be divided into stages by first defining the number of ‘buckets’ and the number of days in arrears in each bucket. Method of communication (call, SMS, WhatsApp, letter, email, etc) can be easily dragged-and-dropped into a bucket to create a treatment plan. The parameters of the communication method can be further adjusted, by adding/modifying the content, assigning it to someone, etc. On the same screen, the aging timeline can be viewed graphically in tabular format, complete with all the communication methods and corresponding DPD numbers. Once everything is configured and the ‘Start Strategy’ button is clicked, Juris Collect’s Whiz will do the rest. Of course, multiple profiles can be create including treatment for self-curing customers, bad paymasters, high-risk customers, etc.

Whiz is coupled with the champion/challenger module which allows for the creation of a champion strategy, which serves as a benchmark. You can then spawn a challenger strategy by tweaking some of the treatment settings, deploy both strategies, and compare their effectiveness. Based on the comparison, you can decide to either retain the existing champion strategy or promote the challenger strategy to be the new champion. This is an extremely effective tool to experiment with different strategies to maximise effectiveness.

Cases assignment and distribution

Whenever cases are distributed automatically, specific rules can be put in place to govern the assignment process so that it can be optimised. Here are some of the methods of assignment which can be implemented:

• Pool-based: customers’ cases are distributed to a pool of personnel who would be taking turns in performing their duties.
• Sticky collection: once a customer’s case is assigned to a member of staff, the staff will remain in charge should a future case of the same customer enters the collection system again.
• Worklist method: certain staff can view everything within the worklist and cherry-pick which case to work on, which also means that the staff could override the system distribution.
• Weighted: the number of cases received by a staff compared to others is adjusted according to certain considerations, like experience and seniority.
• Location-based: the distribution of cases takes into consideration the physical proximity between customers and staff members. Agents nearby are prioritised, while agents too distant can be prevented from being assigned faraway cases.
• Product-based: this method governs the relationship between the staff and loan product categories (eg: credit cards, mortgages, etc). Agents can be either be prioritised or prevented from being assigned to cases of particular products.
• Capping limit: sets an upper limit of the number of active cases which can be handled by a staff at any given time. Besides basing it on number of cases, you can also base it on the monetary value of cases.

Juris Collect’s distribution engine also allows you to collect by CIF or account, and perform centralised or decentralised collection by branch.

Collaboration, coordination, and gamification

Because FIs, lawyers, and collection agencies connect to the same system, Juris Collect facilitates collaboration among all parties that are involved. The collaboration and coordination are well-handled by the system all the way through to the late-stage collection phase, by capitalising on its multichannel contact capabilities via predictive dialler, email, SMS, WhatsApp, and physical letter.

Naturally, the ease of collaboration and coordination will improve the collections performance of agents, but we didn’t just stop there. Juris Collect also gamifies agency performance by measuring the value collected and number of days spent, as metrics. The leaderboard displayed is visible to all parties involved.

Localisation

The collection endeavour, at times, must be subject to multiple local procedures and policies. Juris Collect allows for these variations through localisation features, where activities and rules are aligned in accordance to appropriate, local regulations. Some countries exhibit standards and familiar regulations, while others – though still familiar – have the stages ordered differently. Some are very lax and have straightforward policies, while others have their procedures way more complex than usual.

A real-world example would be the differences between Malaysia and Vietnam. After judgment:

• In Malaysia, it will either proceed to: bankruptcy petition, prohibitory order, writ of seizure and sale, or garnishee order.
• In Vietnam, it will either proceed to: orders for specific performance, award for liquidated damages, or injunctive relief.

Integration with other JurisTech solutions

Juris Collect is an integrated solution that covers the whole collection life cycle end-to-end, from early-stage delinquency, to the final approval of write-off. This closed-loop collection ecosystem also ensures that contactability of a customer will be followed by an action with monitoring and tracking.

At every phase of the debt collection life cycle, we have other solutions to complement the workflow and its processes:

• Juris Mindcraft (early detection): predicting self-curing customers and behavioural scoring.
• Juris Astra (early-stage and late-stage collections): provides a complete outbound call-centre telephony solution.
• Juris Agency (collection agencies): connects debt collectors and FIs.
• Juris Legal (litigation): connects lawyers and FIs.

Another big differentiator is its ability to integrate with third-party services as well, such as CCRIS and CTOS.

Other features

Juris Collect packs a whole lot more we couldn’t cover here, but here are some of the notable ones:

• Scheduler: a built-in scheduler allows for the scheduling of data extraction and data processing jobs. They can be run on specific days of the week or month. This is useful for managing imports of multiple files which are then used for integration with other systems. The scheduler supports both text file transfers and imports.
• Customer repository: once data has been fed into Juris Collect, the customer repository engine will begin calculating DPD and month in arrears (MIA) from day one the delinquency occurs. Data includes information on the customer, contact, facility, security, billing and payment, number of occasions classified/declassified as non-performing, plus instances of invoices defaulted and their corresponding dates.
• Document template generator: users of Juris Collect can define standard credit proposal templates which can then be generated at any point during the collection life cycle. These documents support variables we called unified field objects (UFO) such as the account holder name, IC/ID number, address, etc. Management of these documents is easy as it allows storage for later retrieval, tracking of document status, and logging of document transaction.
• Remarks: you can find sections throughout Juris Collect for the purposes of inputting additional remarks. Remarks entered will appear in the customer case history, but the main remark will be displayed prominently at the top of the screen, each time a case is accessed. An example remark could be something like, ‘Don’t call the customer at this number.’
• Reports: Juris Collect contains a rich set of reporting functions that can be used to produce reports upon request, through menus or scheduled batch jobs. Of course, these generated reports can either be viewed on the screen, exported as Excel or PDF files, or printed. Reports include but not limited to: MIA movement report, team performance report, PTP report, aging report, skip tracing report, etc.

The JurisTech advantage

Juris Collect’s powerful features is but a reflection of JurisTech’s unique values and this is apparent in all our solutions as well. We have been incorporating AI in our software where applicable and the resulting predictive analytics is especially valuable in detecting self-curing customers, to be able to focus contact efforts efficiently. This efficiency in collection lends a positive impact on business financials. The single, web portal connects FIs, collection agencies, and solicitors together for a close collaboration. The collaboration is further enhanced through gamification to incentivise collection efforts. And by using the strategy manager, Whiz, it enables them to easily plan collection strategies and test them in the real world to see which one works best. The flexibility in configuring and using the system is also reflected in the same flexibility possible in capital expenditures (CAPEX) and operating expenses (OPEX) of organisations that use Juris Collect.

Juris Collect was presented with MSC Malaysia APICTA 2012 Best of Financial Applications Award. In the same year, it also took on the Asia-Pacific stage to compete with other solutions from 13 member countries, and won the Merit Award in the prestigious ‘Best in Financial Applications’ category.