The Data Analytics Blog

Our news and views relating to Data Analytics, Big Data, Machine Learning, and the world of Credit.

All Posts

Right-Party-Connect Rates In 2018

September 22, 2017 at 1:19 PM

Collection Cascade

When looking at the collections cascade and it’s diminishing returns, we always emphasis the responsibility factor relating to each metrics. Whom is responsible for these areas, and how do we monitor the accuracy of these metric results?

If we take a closer look at the right-party-connect rates, you'll notice several dispositions that occur regarding this metric:

  • RPC KPI’s for collection agents
  • Manual logging of RPC’s and subsequent impact
  • Quality Assurance monitoring of RPC’s
  1. Right-party-connect is a direct result from available data, analytic insights/models and dialler strategies, but we still see companies including RPC as a KPI for collections agents. This is concerning due to an agent having no influence if the right-party answers a telephone call or not. RPC outcomes are only as good as the data and strategy used. If your company uses RPC as an agent KPI, how sure are you that the logging of RPC’s is accurate? Remember, agents always find a way of manipulating outcomes to suit their needs.
  2. If we look at how we log a RPC in our operating systems, we can all agree that for the most it is still a manual action by an agent. But is this really optimal, and what is the risks associated with these manual actions? Even though we may not assess agent KPI’s on RPC outcomes, we still assess their conversion of a RPC to a promise-to-pay (PTP). We know that agents can very quickly find a way of manipulating outcomes to suit their needs. So how comfortable are we in the accuracy of our agents’ negotiation outcomes, knowing that there is still room for manipulation?
  3. I recently had a conversation with a colleague regarding their comfort around their QA process and the validity of the outcome. What QA penetration % on calls would you feel is sufficient to draw conclusions from? Would you bet your money on the validity of that penetration? % NO... And I agree fully.

We all know that most companies penetrate less than 0.5% of all calls, and with those odds we are very far from being comfortable to draw any concrete conclusions to influence strategy.

What if we could penetrate 100% of all our calls, would that change the face of our RPC outcomes? Would that result in a focus shift?

Looking at where the collections solutions are going in terms of data analytics, machine learning, AI agents and voice analytics – isn’t it important to completely take the RPC metric away from the call centre agent and their ability to manipulate it?

Let’s start talking about voice biometrics and the importance of it within the operational call centre...

increase your collection and recovery yields

Perry de Jager
Perry de Jager
Perry has been involved in Collections and Recoveries for the past 12 years, spending time in different market segments ranging from law firms to investment companies. At Principa, Perry has worked on extended projects within both South Africa and the Middle East with some of the largest financial organisation, providing on-site consulting within the collections and recoveries space covering strategy, process, people and technology.

Latest Posts

10 Machine Learning Books To Read For Budding Data Scientists

Machine learning and artificial intelligence are exciting fields, and we've been writing about these topics for a couple of years now. While a lot of what we talk about on our blog is advanced implementations of machine learning and can be overwhelming to beginners, the core concepts of machine learning are actually pretty easy to grasp. There are many resources and cheat sheets available online, but we believe the old fashioned way of learning is sometimes the best: with a good book. Few resources can match the in-depth, comprehensive detail of a good book.

Our Top Artificial Intelligence, Machine Learning And Data Science Podcasts

Our data scientists are keen readers and avid podcast listeners. In this blog, we list a few of the podcasts that cover topics such as data science, machine learning and artificial intelligence, and that we’d recommend if you’re looking to start exploring the world of podcasts.

Where To Learn Data Science Skills In South Africa

In this blog, we’ve created a (non-exhaustive) list of courses you should consider if you want to learn essential data science skills in South Africa. These courses are mostly classroom training from South African institutions, but if you’re more interested in online learning, check out our blog Where To Learn Essential Data Science Skills Online.