This article discusses issues associated with incentivizing effective collections. Structuring and implementing an effective incentive scheme is an integral part of the account management process. It is important that an organization implements a collector incentive scheme that will compliment the methodology of working delinquent accounts. The objective of an incentive scheme should be to motivate collectors in such a way that will improve overall collections effectiveness. This article discusses three different approaches to structuring an incentive scheme for individual collectors.
The incentive schemes discussed within this article are appropriate for ‘normal’ delinquent collections as opposed to policy group collections. The term ‘policy groups’ refers to groups of accounts that are in a state other than normal delinquent accounts. Examples of policy groups include deceased estate accounts, fraudulent accounts, insolvent estate accounts, etc. The actions taken within policy groups are different to those taken in early stage collections, and in recoveries. Specialized training is required for each collector within these areas. Ideally the policies will be documented (or driven by the collection system), to ensure a consistent approach across all collectors.
As a rule, incentive payments should be higher for the better performing collectors. Therefore effective evaluation of collector performance is required to drive the incentive scheme. The collector evaluation should include both qualitative and quantitative performance standards, for example:
Number of accounts worked per employee
The results of collections calls made
Quality of telephone calls and account record updates
Volumes and ratios of promise to pays
Volumes and ratios of honoured promise to pays
Payments collected as a percentage of total amount due
Backlog of accounts to be worked.
It should be noted that any incentive scheme is prone to manipulation by collectors seeking to maximize their rewards. One of the objectives of any scheme should therefore be to minimize the potential for manipulation. To do this, the incentive scheme should be driven by results and not simply collector actions.
It is also important to remember that the collections results will differ according to level of delinquency, risk, and account type. This should be considered when setting benchmark or minimum performance standards for each collector, team or the collections department as a whole.
Within account management strategies, accounts are often segmented into groups based on delinquency and risk. Behavior score or an alternative risk segmentor is used to allocate different levels of risk and therefore levels of difficulty (within delinquency level), to different collector queues. It follows that different thresholds would therefore be required by queue, within the same delinquency level. On the other hand, where volumes do not justify segmentation by risk into different collector queues, or where a more simplistic approach is required, all accounts within a delinquency level can be queued together.
Any incentive scheme should encompass the following objectives:
Minimum thresholds should be reached before incentives are paid, to ensure that a certain level of performance is achieved
Try to ensure that an individual is incentivized on an on-going basis, and never reaches a stage where additional performance yields no additional incentive
The individual contribution to the overall result should be rewarded
Incentives given should reflect the degree of difficulty in work assigned
The incentive should be on a sliding scale and should be of real value to the collectors (greater than 10% of salary)
An organization can also set various qualitative and quantitative measurements that must be reached prior to any incentives being paid, such as a minimum number of accounts worked, a minimum number of promise to pays taken or a minimum number of promises kept. More sophisticated schemes could use a number of these measures to set an individual’s performance grading, whereby only individuals above a minimum performance grade qualify for an additional yearly incentive (typically linked to an annual bonus or appraisal).
Three approaches are detailed below for incentivizing individual collector performance. It is important to note that in all three cases, only payments received within the same billing period/month should be eligible to be counted against the total due at the beginning of the month. This will incentivize collectors to ensure that promises are set during the same month, and not over the billing date. It will also urge collectors to try to ensure that payments occur within the same month. This will incentivize collections activity before month-end (when customers salaries are paid), and should therefore increase repayment.
Within collections strategies it is typical for ‘normal’ 1 cycle accounts, (excluding ‘special’ circumstance accounts such as first payment defaulters, broken promise to pay accounts, etc.), to be assigned to a single work queue. The level of risk represented by accounts determines when an account moves from the ‘hold’ or ‘no action’ queue to a ‘work’ queue.
Ideally a pool of collectors is assigned to work normal 1 cycle accounts and all collectors work throughout the billing cycle. Theoretically all collectors have an equal chance of working both the easiest and the hardest accounts. Therefore, within the incentive scheme, it is not necessary to assess degree of difficulty of 1 cycle accounts worked by one collector, compared to those worked by another. We assume that payments made can be attributed to collectors on the basis of an action, for example a promise to make a payment is obtained on the account.
Firstly a minimum total payment value acceptable for 1 cycle accounts should be established (minimum threshold). For example, 50% of total due is the minimum repayment threshold that must be reached in order for incentives to be paid.
At the end of the billing cycle the total payments received on 1 cycle accounts is established (total payments). The minimum threshold is subtracted from the total payments, leaving the amount by which repayments exceeded the minimum threshold. A portion of this amount (for example 30%) can become the total amount available for the incentives to be distributed amongst the collectors (incentive value).
The incentive value available is distributed amongst the collectors on the basis of the proportion of payments received attributable to each collectors’ actions (obtaining promises to pay).
The organization will determine the incentive value. Research has shown that in order to be effective, an incentive should make up a significant proportion of a collector’s salary – up to 50% or more of a collector’s salary could consist of incentive payments.
Ideally, the scheme should differentiate between full and partial instalment payments. A weighting could be applied to reward full payments to a greater extent than partial payments, thereby encouraging collectors to negotiate payments upwards, and to collect full instalments. Alternatively, the scheme could be structured in such a way that only full instalment payments qualify as payments in terms of the incentive scheme.
Once the delinquency level is greater than 1 cycle, typical strategies will allocate accounts to different queues based on risk and balance during the month. It is likely that these queues will be assigned to different collectors.
There are two work queues for these 2 cycle delinquent accounts, ‘Normal’ and ‘Harsh’. At billing, high risk low balance accounts are allocated to the Normal queue while high risk high balance accounts are allocated to the Harsh queue. High scoring accounts are all placed in the Normal queues. As the month progresses, accounts are moved from Normal to Harsh queues according to risk and balance.
Different queues have different levels of risk, and therefore difficulty of collections, associated with them. In order for the incentive scheme to be fair, a methodology is required which takes account of the varying degrees of difficulty in obtaining payments from different types of accounts.
The behavior score represents the probability of an account reaching the predefined ‘bad’ status and therefore indicates the degree of difficulty associated with obtaining payments from an account. Where a behavior risk score is available, the odds can be used to weight the value of the payments received.
The following example shows 2 accounts with the same opening balances and the same payments received, but with different behaviour scores.
The calculation used to obtain the weighted payment value uses the Balance at Risk Factor. B@R Factor = 1/(1+Odds). Weighted payments take into account both the risk of the account as well as the value of the payment. For example, the B@R Factor for Account 1 above is calculated as follows: 1/(1+15) = 0.0625. The weighted payment is calculated by multiplying the actual payment by this factor: $100 x 0.0625 = $6.25.
The weighted payment column shows that the weighted payment received on Account 2 is greater than that on Account 1, due to the relative difficulty of obtaining the payment indicated by the higher risk/lower behavior score.
Again, a minimum threshold, total payment and incentive value should be established. The incentive value is then distributed amongst the collectors on the basis of the total weighted payments made due to their actions, as a % contribution of the grand total weighted payments amount.
The methodology of weighted payments applies whether accounts are split into different queues according to risk or not. By using the weighting of the Balance at Risk factor, it does not matter if collectors actually worked different queues.
This example scheme again requires that individual collectors work allocated groups of accounts, and that the allocated groups can be monitored for payments taken. The incentives in this scheme have been stated as percentages of salary.
Here, incentives are based upon payments as a % of the balance or amount due within each allocated group of accounts. The allocated accounts should be by debt category types. Examples of debt category types would include:
1 Cycle ‘Normal’ Accounts
1 Cycle Broken promise to pay
2 Cycle ‘Normal’ Accounts
2 Cycle Broken promise to pay, etc.
||Incentive > Avg, < Max
|1 Cycle ‘Normal’||66||70||5%||10%|
|1 Cycle Broken promise||50||55||5%||10%|
|2 Cycle ‘Normal’||45||60||5%||10%|
|2 Cycle Broken promise||20||35||5%||10%|
Average repayment as % of total due within each debt category for the last x months (for example a rolling 6 month period) is calculated. Then maximum repayment as % of total due within each debt category over the last x months is calculated. Lastly, the incentive table for each debt category based upon the averages and the maximum repayments as % of total dues within each work queue for the last x months is derived.
Incentives are granted as a % of salary, where 5% is granted to collectors who collect more than the average but less than the maximum, and 10% is granted to collectors who collect more than the maximum.
A shortfall of this particular model is that the %’s of salary are low. In addition, the incentive scheme is capped, in that a maximum of 10% of salary is awarded, irrespective of what the collector recovers once the maximum has been reached.
The benefits of using this methodology to set the benchmarks are that incentives are only granted if the collector achieves better than the average results for that queue. In addition, the top incentive is only given if the previous maximum is reached or bettered. The benchmarks should be reassessed on a monthly or quarterly basis.
It is important to test the incentive scheme before implementation (simulated application and monitoring) to allow for changes to be made before the scheme is communicated to the collectors. This is to ensure that the incentive levels are correct, and that the probable payouts are within the allocated budget.
Ideally, incentive schemes should only be based upon reporting that is automatically (as opposed to manually) produced by the collections system and host system, and has been validated as being accurate.
The method of presenting the incentives should be considered. Incentives can be presented openly, therefore promoting competition between collectors. The promotion of the competitive element is only feasible, however, if large teams of workers are working the same types of accounts.
On a monthly basis, the top x collectors should be identified and some form of recognition program implemented. This may be in the form of a reward/incentive, or simply a published list to recognize exceptional performance amongst a peer group. Over a number of months, the consistent top performers can be identified and recognized.
In a similar way, the bottom x collectors can be identified and some form of coaching/development program implemented. This may be in the form of a training exercise or a regular audit review with collection supervisors to establish the areas that need development. The consistently high performing collectors may be assigned to this group as mentors/trainers to develop these individuals. This can ensure that departmental and team based incentives can be achieved more easily. In addition, reward programs for these mentors, based on improved performance of the collector, should be considered.
PIC Solutions provides consumer credit solutions to a wide range of blue-chip organizations. We are experts in the fields of credit, risk and software and have an established track record of success powered by solutions. Our ongoing commitment is to enable clients to optimize the credit life cycle.