Recently, we had a good call with a large credit card issuer. We talked about different aspects of the collections process. More specifically, I inquired about promise-to-pay, the option offered to customers to get past a short term cash crunch. I view promises-to-pay as securing a commitment for a payment by some date. It’s a form of temporary relief for a customer who is struggling with cash issues currently.
The card issuer’s strategist for collections remarked on something that caught my attention, saying this:
Promises-to-pay are not very good predictors of payment performance by a delinquent borrower.
I admit this threw me off. I hadn’t thought of promises-to-pay in a risk context. Rather, I was thinking of it as a better match for a customer’s cash flow to a payment obligation they have:
I came into the conversation with the green highlighted branch in mind. The collections strategist was more focused on the red branch above. I understand the perspective. While a delay in making a payment is certainly an accommodation to a customer in a cash crunch, it’s also a form of credit. The borrower is being extended time to make good on their obligation. It doesn’t mean the promise-to-pay will be honored. As consulting firm McKinsey notes:
The customer needs to keep the promise to pay. This is a complex decision with plenty of opportunities for derailment.
Understood in that light, does it make sense to grant promises-to-pay to every delinquent customer? No. But how would you decide who qualifies?
BackOnTrack’s promise-to-pay proxy
A key part of the BackOnTrack® offering is that the delinquent customer must get their account current in order to receive a statement credit. The customer has the discretion to make the required payment right at the conclusion of BackOnTrack, or to pay later.
Just over half of BackOnTrack completers click to Pay Now. But what about those who click Pay Later? Remember, the offer is a statement credit for: (i) completing BackOnTrack; and (ii) getting one’s account current. That statement credit is just sitting out there, waiting to be earned when the customer makes a payment. The customer has an incentive to follow up on this Pay Later commitment.
While there is not a specific date, those who click Pay Later are essentially making a promise-to-pay. What can we learn from them?
Measuring promise-to-pay follow-through
We have a set of data that enables analysis of this proxy for a promise-to-pay.
timeline for delinquent accounts analyzed
We have an initial month of early stage (1-30 DPD) customers. We can check to see if they rolled or not the very next month. Then, 6 months later, we can check their account statuses.
We run an initial analysis of those early stage delinquent customers who clicked Pay Later, our proxy for a promise-to-pay. What do we learn initially?
| Promise-to-pay outcome | Roll rate @ 6 mo. |
| Kept promise-to-pay | 15.3% |
| Broke promise-to-pay | 79.0% |
People who make a promise-to-pay and follow through are significantly better performers. People who break a promise-to-pay exhibit very poor payment performance in the months that follow. For the customers who broke their payment promises, the bank would have been better off applying different treatments and more experienced collectors early. This would drive better results than offering a promise-to-pay.
The question is: how do I know which customers merit a promise-to-pay?
Quickly assessing a customer’s risk
In a previous post, we introduced the concept of Collections Personas. These are nine different personas built on the 3 C’s of Delinquency: cause, capacity, control. In that post, you can see a clear separation between the first five personas and the last four in terms of payment performance.
What would those personas look like for offering the option of a promise-to-pay? We’re able to analyze that for those BackOnTrack completers who clicked the Pay Later option. The table below provides the payment performance for these promise-to-pay customers.
| # | Persona | Roll rate @ 1 mo. | Roll rate @ 6 mo. | Count | Pct to total |
| 1 | financially secure forgetter | 4.8% | 4.8% | 62 | 8% |
| 2 | will figure it out forgetter | 0.8% | 11.9% | 118 | 16% |
| 3 | get through it forgetter | 5.6% | 7.4% | 54 | 7% |
| 4 | scrappy forgetter | 1.4% | 17.6% | 74 | 10% |
| 5 | resolute payer | 5.6% | 13.0% | 54 | 7% |
| 6 | scrambling after a setback payer | 11.5% | 25.0% | 104 | 14% |
| 7 | overwhelmed payer | 10.3% | 22.4% | 58 | 8% |
| 8 | on the edge payer | 12.5% | 28.1% | 96 | 13% |
| 9 | under an avalanche payer | 24.3% | 41.7% | 120 | 16% |
| 740 | 100% | ||||
| personas 1 – 5 | 3.0% | 11.3% | 49% | ||
| personas 6 – 9 | 15.6% | 30.7% | 51% | ||
Look at those results. By asking the customer to answer three BackOnTrack questions, we can convert their responses into analytics that tell us A LOT about their future payment performance.
In this case, a simple screening process identifies customers who are good candidates to offer the promise-to-pay option. Let’s say the first five personas are appropriate for promises-to-pay. In terms of customer relationships, a bank can go a long way satisfying its customers with just a small accommodation.
The last four personas are the ones where early intervention will be more beneficial. Rather than offer a routine option for a promise-to-pay, these customers could use more substantial treatments. If you’re assigning accounts to collectors, these are the accounts for your more experienced professionals.
Bottom line: there are very good, simple ways to assess who should be offered the promise-to-pay option. A simple triage will pay meaningful results in terms of customer satisfaction and reduced charge-offs.
Click here to learn how BackOnTrack analytics can fuel a smarter approach to offering promises-to-pay for your delinquent customers.
