Banks and other companies go to great lengths to capture market share when they offer credit. They debate for months about whether to appeal to their customers’ vanity or frugality, whether to market by Internet or telephone, or whether a pink credit card might prove more popular than a purple one. But when companies try to get their money back from those who neglect to repay, many suddenly forget their marketing savvy.

All too often collections are an afterthought. The organizational structure of one Asian bank places collections on a par with the office responsible for mailing out monthly statements. At another bank an executive confided that collections appeared to be an “elephant graveyard” where unwanted employees were sent to await retirement.

Yet collecting overdue payments is an integral part of banking and of any business offering credit or payments over time, including retailers, hospitals, utilities, and telephone companies. It poses a particular challenge in emerging markets, where customer information is often scarce and legal systems may well be inadequate. The best companies see collections as a profit opportunity, and talented people can use it to make substantial contributions to the bottom line.

In our experience there is considerable room for improvement. We estimate that companies in emerging markets can reduce overall retail credit losses by 30 percent or more by improving customer segmentation and frontline execution. In Russia one financial institution cut its losses in half by tailoring its collection methods to specific debtor groups, changing the incentives it gave to collectors, and offering new tools, such as flexible payment systems for delinquent borrowers.

While emerging markets are the focus of our work, we have learned that even in developed markets credit collection is often ineffective and ripe for improvement. To stem the significant losses tied to poor practices, top management must insist that collections adhere to the same high performance standards and analytics that prevail elsewhere. Many executives fear that the reputation of their companies will suffer if collections become too aggressive, but the careful segmentation of debtors and the flexible pursuit of delinquents—applying pressure, in other words, only where it’s really justified—can largely eliminate these risks.

The root of the problem

Around 8 to 10 percent of total credit card balances become delinquent every month at a typical bank in an emerging market, and up to 20 percent of these delinquent balances will eventually be written off. At one Asian bank such write-offs equaled nearly half its net profit in 2005. The problem is even more severe at nonbanking companies that offer delayed payments for products or services.

Part of the problem is that managers typically dismiss collections and write-offs as an operational cost that offers little room for improvement. Instead, top executives tend to focus squarely on growth and sales. They don’t understand the potential of better performance in collections and think that seemingly heavy-handed collections policies will damage public opinion. What’s more, effectiveness in collections is difficult to gauge, since external factors, such as an improving economy, can influence performance. As a result, collections departments often feel little pressure to improve.

Beyond these universal elements, the challenge in emerging markets is even greater. First, in developing countries creditors know much less about their customers than do their counterparts in developed markets. Information on credit applications can be difficult to verify and may later prove useless if the debtor defaults. We’ve seen banks that don’t have valid telephone numbers for almost half of their customers. Forged documents are common, a lack of public databases makes it hard to confirm addresses, and reliable income records do not exist where employees get paid in cash. Furthermore, in developing countries with no credit bureaus, customers have little incentive to manage their credit history, and companies have no way to research it.

Meanwhile, the legal systems in many emerging markets often frustrate efficient debt collection. At the extreme, seizing assets or salaries to recover delinquent loans may be impossible. Banking regulations can intentionally or unintentionally restrict the options available to banks. In Russia, for instance, institutions cannot write off the principal from a loan until the full term expires, since current tax law views these write-offs as taxable profits rather than losses. Such provisions make debt restructuring and settlement offers more difficult.

Companies in developing markets generally have to rely on in-house collections expertise because legitimate third-party agencies are scarce and secondary markets for nonperforming loans are underdeveloped. Moreover, the methods that loan sharks and other illegal lenders use make those available to banks look tame and ineffective by comparison.

To overcome these hurdles, companies should focus their collections strategies on a clear customer segmentation and top-class execution. Banks and other credit-related businesses need to reject the idea that a more assertive collections policy poses unacceptable risks to their reputations. On the contrary, improved segmentation and better execution allow companies to identify delinquent debtors and to offer them flexibility in troubled times. Most late payers can be left alone, with only the most recalcitrant facing a more robust approach.

Companies that meet these challenges will find themselves not only more profitable at home but also more skilled when they compete in other markets.¹ Many collections departments around the world seem out of touch with the “mobile-phone generation,” for instance, and have yet to exploit the communications tools typically used by the young. While some banks in emerging markets send text messages to defaulters, their counterparts in Europe and North America have been slow to use this channel. E-mail and other Web-based contact strategies remain almost untouched.

Sorting out the debtors

Not all debtors are created equal, including the delinquent ones. Most people who take out loans or buy on credit expect to fulfill their obligations. In our experience, up to 70 percent of the borrowers whose payments become overdue are well-meaning men and women who missed a payment and have every intention of getting up to date quickly. A further 10 to 20 percent have fallen on hard times, such as unemployment or illness, and temporarily can’t make their payments.

The remaining 10 to 20 percent or so have little intention of paying, because they either consciously mean to commit fraud or want to take advantage of a company’s lenient policies. This group presents a particular problem in emerging markets: the lack of credit bureaus and other kinds of infrastructure makes it difficult to screen out these customers, and ineffective legal systems limit the options when loans become delinquent.

Yet collections systems in developing countries often make the mistake of treating everyone equally. In a typical process, a bank begins by calling the borrower once a payment is, say, five days overdue. Standardized reminders—usually a polite phone call from either a call center or a branch—continue until the payment is 60 or 90 days overdue. At that point the account is transferred to a more aggressive collector, who might eventually instigate legal action.

Two problems stand out with this approach. First, it wastes resources, since up to 70 percent of the early calls go out to borrowers who would pay without a reminder. Second, collectors, knowing that a majority of the recipients are well intentioned, tend to be relatively unassertive in these early calls, inadvertently giving the willfully delinquent more confidence to withhold their payments. Experiments have shown that polite reminders from banks actually encourage such debtors not to pay. One South Korean bank found, for example, that it collected 3 percent more money by not calling anyone during the first two weeks of delinquency. Call center managers are deluded into believing that the courteous approach is successful by the large portion of late payments collected. Many of these, of course, would have been paid anyway.

The South Korean credit card crisis of 2003 and 2004, which gave us a unique opportunity to analyze a group of customers who had defaulted at both of two local banks during those years, demonstrated that different segments require different approaches for optimum collections (exhibit). The first bank gave all of its customers an easy way out of their debt problems by offering low monthly payments; the second bank insisted on the full payment of outstanding debts, with no flexibility. For about half of the customers we analyzed, flexible payments were crucial: the first bank collected roughly 50 percent of its outstanding balances by offering the flexible-payment option. The second bank collected only 5 percent of its outstanding balances from the same customers. However, for a second group of customers—about a quarter of those examined—the first bank’s flexible-payment option simply gave them a way to delay their obligations, exposing the bank to unnecessary risks and losses. While the bank with the stricter approach recovered all of the outstanding debt from this group, at the lenient bank many of these customers defaulted on the flexible-payment program a few months later, and in the end that bank recovered only 70 percent of what they owed. The two banks were equally successful with the remaining customers.

Instead of a routine based on the length of delinquency, banks should use marketing analytics, such as cluster analysis, to develop a behavioral segmentation of their delinquent customers and to develop tailor-made risk scorecards to segment late payers further. These scorecards, which must be designed using insights into the local market, can be useful with surprisingly little data. When the credit department of one Brazilian retailer built up the needed capabilities, for example, it segmented its credit customers in an effective way by using just 20 pieces of information (mostly drawn from internal credit applications) relevant to loans and the people who received them.

A first step toward an effective segmentation would be to create self-cure scorecards drawing from demographics and past behavior to estimate whether each delinquent borrower was likely to pay up without any action from a bank. Customers who are habitually a day or two late with payments or who have a long relationship with the institution, for instance, would rank high on the self-cure scale. Indeed, so would most delinquent debtors, and the bank would therefore take no action against them until an obligation was more than two weeks overdue.

For the remaining 20 to 30 percent, a write-off scorecard would gauge the likelihood that a debt might end in default. The higher the rank on this scorecard—earned, perhaps, by previous broken promises when payments were late—the more assertive the bank should be early in the collections process. These assertive measures could include more frequent contacts by phone, special delivery or courier-borne notices, or even visits by bank staff. (Low labor costs in developing countries make such visits economically feasible.) Also, collectors can push for promises to pay, monitor closely whether debtors meet those promises, use a more aggressive script that focuses on the consequences of defaulting, and resort to a clear chain of escalation for borrowers who do not meet their obligations.

While most creditors, in both developing and developed markets, can probably stem losses by using a basic segmentation of delinquent borrowers, creditors with the highest aspirations deploy more advanced behavioral-segmentation techniques. Using a range of marketing tools—market research (such as interviews with delinquent debtors), product design (flexible-payment options for customers in temporary financial troubles), and sophisticated analytics—creditors can move beyond a debtor’s willingness to pay as the key distinguishing criterion and create a more complete view of the various segments. Then they can craft finely tailored approaches for each.

By combining scorecards and behavioral-segmentation rules, one Asia-Pacific bank identified a group of customers with very high balances in their savings accounts when credit card payments became delinquent. Historically, the bank wrote off 18 percent of the debt such customers owed, since many would empty their savings accounts before the bank became serious about collecting the overdue debt. With the new segmentation approach, the bank called these customers within days of their missing a payment and used a tailored script to persuade them to pay off the entire credit balance, executing the transfer immediately via phone banking. Losses all but disappeared in that segment.

Young customers often require a special approach to behavioral segmentation because neither risk nor willingness to pay is terribly helpful in categorizing them. Instead, a collections strategy for this segment should focus on teaching basic financial planning and the consequences of neglecting financial commitments. To help young customers avoid missing payments in the first place, creditors can offer debt-management tools such as automatic reminders (for instance, by text message) if outstanding balances or total weekly expenditures on credit cards exceed predetermined thresholds. One Australian bank sent government booklets on financial planning to all of its delinquent young borrowers. Besides minimizing losses, this approach enhances a bank’s image as a responsible lender.

Collecting better from one place

In addition to improving the segmentation of overdue borrowers, banks in emerging markets must execute collections more successfully. For many, a critical move will be to transfer responsibility for debt collection from branch offices to a centralized department. (An exception would be microlending operations, which rely heavily on personal relationships.) Using branches to collect debt breeds inconsistency, since each outpost evolves a different set of tactics and rules. Branch-based sales representatives, who often double as collectors when necessary, aren’t trained to collect debt effectively and can feel uncomfortable confronting the same customers to whom they originally sold products. While tailoring strategy to local circumstances is often an advantage, in many developing markets decentralized collections efforts have little analytical rigor, waste resources, and rely on unskilled staff.

Centralized collections centers, on the other hand, can employ staff specifically trained to handle even the most unyielding borrowers. Improved training allows collectors to speak with one voice, basing their approach on a clear understanding of the bank’s strategy and a complete view of the customer’s status (for instance, a customer defaulting on credit card payments might also hold a mortgage in good standing). Along with formal training, coaching is a vital part of improving skills, especially with new hires. But in many emerging markets, branch leaders of banks that take a decentralized approach lack even the most rudimentary tools (such as software allowing them to listen in on calls) that would help them monitor trainees.

Centralization also improves performance management. Key performance indicators can be developed for workers whose sole responsibility is debt collection; at branches this approach isn’t possible because collections will often form only one of many roles an employee has. Technology also plays a part. A centralized collections department allows banks to invest in specialized equipment such as telephone dialers, which can double or triple the number of calls employees make. Specialized software linked to the phone system can track the productivity of individual collectors, with metrics such as calls per hour, promises to pay extracted from borrowers, fulfilled promises, and dollars collected per promise kept. If collectors are widely disbursed, such efforts will be nearly impossible.

Clear metrics open avenues to performance pay by giving managers a yardstick for measuring success. The best performers in a collections department often bring in several times as much money as the poor performers do, and these stars should receive sizable bonuses to mark their success. While performance pay often conflicts with a bank’s standard salary scale, collections managers should strive to implement some version of it or risk losing their best talent to competitors. One South Korean bank took the extreme step of basing the entire compensation of its collectors on performance, specifically to make them more aggressive. To safeguard against excessive eagerness and ensure that collectors follow policies closely, companies should institute compliance audits—for example, by having supervisors listen in unannounced on collectors’ phone calls.

An added benefit from centralization is that it helps stimulate a friendly and productive rivalry among peers. Centralized operations make it easier for banks to stir excitement around performance improvements, for instance, by pitting individuals or teams against each other in playful competitions, with prizes awarded to the most productive participants.

Banks and other credit-giving businesses often relegate collections departments to the dark corners of the organization and neglect them. But by sharply reducing write-offs ascribed to bad loans, efficient collectors can greatly increase the profitability of lending operations. This won’t happen, though, unless top executives insist that the collections department achieve the same high level of performance they demand from the rest of the institution and unless the department stops treating all debtors alike.