It was after a collection conference workshop on employee compensation that we were motivated to ask ourselves some interesting questions about the way collection companies are managed and operated today. As we pull ourselves out from a long, difficult period, the subject is probably more relevant. We believe that it would fair to describe the average owner or

collection manager as conservative and conventional. For most, although everything around them is changing rapidly, the basic business process has not changed in many years. Although some radical thinking does exist, it is often the exception. That is not a weakness – change is often difficult and costly.This discussion is intended to provoke dialogue and questions about different business models in the collection industry. The thoughts presented are based on many years of collection experience as well as a unique opportunity we have had to introduce radical thinking to the industry.


Let us start with the concept of compensation. The conventional wisdom is that we pay people for collecting money. Almost every compensation model in the industry is based on that model. It may have worked when we were recoveing 35% at rates of 50%. But is it the best approach in this new era of collections? Should we start paying people for finding the debtors who do not pay? Why should we consider such an outrageous idea? Would the following justify it?

  • Over 85% of accounts are not collected
  • It is easier to find a person who will not pay than a person who will pay
  • Everynon-collectable account that is identified quickly allows us to devote more resources to the collectable accounts
  • If less than 15% is collected, why (and at what cost) do we put the same effort into the other 85%?

Some compensation models do attempt to limit commissions paid to collectors if the payments are generated prior to an account being worked by a collector. That makes sense, but what about the accounts that are eventually worked by collectors? Should we compensate a collector in a different way if an account is more collectable? We usually do not. What if we could identify accounts that are more likely to pay? We may not even need experienced collectors to work them initially. Of course this is a different business model, but one that has to be considered as we try to collect more at a lower cost.

Finding the people who will pay

As we consider different aspects of a collection operation, the experienced manager will realize that a collection operation is not a matter of simply bringing together areas such as people, a few large clients, technology, scoring, automated dialing, collector compensation and

client services. All the different facets of an operation are related and tied together by a single important (and often neglected) item – your business model. The area of scoring is a great example of the value of the business model.

With “scoring”, we are trying to determine who is most likely to pay. Eliminating even a small percentage of the large numbers who are not likely to pay will also help to isolate possible payers faster. We could then use our results to classify our accounts from highly collectable to least likely to pay. How do we score accounts?

• We obtain “external” scores • We use the (internal) information and knowledge we already have within our systems and companies

• Our system creates dynamic scores based on hundreds of changing parameters

• Collectors and managers make individual judgments

Why do we continue to give up on accounts because they are small? That has to be a poor scoring model – Talk to companies that collect parking tickets! We must begin to focus on a collection model that is driven by collectability. We can classify every account and focus our efforts on the 60% who are most likely to pay. What about the other 40%? Many clients will not want to see their accounts closed within 24 hours even if they knew that they would never collect on most of them. Because people cost much more than letters, we must consider working those accounts using mail and messaging as opposed to telephone calls and contacts, making sure that they are never overworked. How do you do that? Simple – you automate, automate and automate.

If we put less effort into less collectable accounts we will need fewer collectors! Those collectors would collect more and would consequently demand higher commissions. We surely need to revisit our operating model. If costly technology is applied to

accurately predict paying debtors, the compensation models must change.


Today, the automated dialer is established, accepted and mature technology within the collection industry. In spite of its productivity, many of us still do not have a business model with a high percentage of predictive calling. Why? It is because we know that people will collect more when they are “responsible for an account” for a longer period. But the ownership (cradle-to-grave) model comes with a price. You cannot dial those accounts predictively.

So why not take your large balance collectors, change the ownership model and let them work in a pooled environment? It is probably because most of us cannot measure individual productivity in a pooled environment. If we could do that and find a way to allow collectors to hold accounts for just enough time to convert positive contacts to money, we could improve collector

motivation and increase contacts by 100% compared to an ownership model with no predictive dialing. Yes, we need to go back to rethinking our business model. And when collectors make many more contacts, they will probably collect more. Do we pay them more because of your investment in technology? Isn’t it interesting how we keep circling back to the issue of compensation?

Dialers need to be a part of our business model. Often, they are not.

• Dialers must work without being managed in terms of the accounts and phone numbers to be worked, time of day to call or the pacing of calls.

• We must find ways to improve the quality of the numbers we send to the dialer. Bad numbers must be removed quickly and automatically.

• We must dial responsibly – predictive dialing must not be allowed to become a liability.

What more can we change?

In addition to the above, there are several other areas that merit attention within a business model. Some of these are:

  • If automation plays a greater part, a collector’s role could be reduced to “working their accounts”. Decision-making and thinking must be consistent and should be controlled by management as opposed to individual collectors.
  • How do we credit individuals when multiple people (e.g. collectors, skip tracers, insurance billing staff) were involved in generating the payment?

A new business model?

If we changed our business model, how would the traditional agency change? Let us consider a company with 50 collectors and 5 supervisors. We have analyzed and classified the accounts based on collectability. We may not need experienced people to work the most collectable accounts, and the least collectable accounts will be worked with the least effort being expended. In the new business model, our collection operation could possibly look like the following.

since the accounts are not likely to generate many payments.

• The last 35% will be worked by the system with letters and automated messages. Inbound traffic and any outbound calls could be handled by the same 20 representatives.

The following represents the collector allocation by collectability “segment”.

  • • We must not be forced to manage workflows or ensure due diligence with human resources. These are jobs where computers excel. 
  • The new model calls for only 40 collectors compared to 50 in our traditional model. We may be able to manage with fewer supervisors too. In addition to reduced numbers, costs are lower because we do not require experienced people in every area.

Where do we start?

In addition to changes in management thinking, you will need new tools to implement progressive workflows and higher levels of automation. In the 1990’s several major companies were challenged to

redesign their business processes. IBM reengineered their operations, retailers changed to “just-in- time” inventory models and automakers invested in robotics. For collections, it is unlikely that traditional software models will provide the power and flexibility to undertake major reengineering. The most cost-effective solutions will probably utilize knowledge-based technology. This technology uses stored knowledge to make management decisions, allowing you to frequently change your logic for scoring accounts or adjusting your work plans without the need for programming changes.

The ideas presented may seem radical to some of the readers but they have been tested and implemented with exceptional results. Where do we go to find the next great increase in productivity? We have always looked to technology but that may not be the only answer. After nearly 25 years of automation, perhaps it is time to take a new look at how we would design our solutions if we had to do it today. And it all starts with a business model. While technology does help, it does not solve problems – people do.