You can find the terms "AI," "artificial intelligence" and "machine learning" all over collections. You might also assume that most everyone in the industry has integrated AI and machine learning into their strategy. That assumption would be wrong.
“There are not a lot of real, true AI use-cases at the moment. It is a term that is a little bit overused right now,” says Chris Repholz, Chief Business Officer at The CCS Companies.
Substantial AI use-cases are still pretty rare in financial services, agrees Michael Orefice, Business and IT Practice Leader at Bridgeforce.
“Everyone is using [the term] AI; everyone is using the term ML (machine learning); and in many cases, they’re doing it interchangeably," Orefice says. "But in reality, there’s very few use cases where real, true, raw AI, the ability to just influence a transactional outcome...in the financial services sector are proven...to have added benefit. But the number of use cases is growing.”
The promise of AI and machine learning is definitely still there, however, and that promise is in AI-powered computing systems that find the connections for you, says Alison Sorel, VP Collections Solution Strategy at Katabat.
AI is “not needing to have a person making all of the decisions and connecting all of the patterns in the data,” she says, but rather, “using the power of computing today to be able to make those connections and help you make the best decisions.”
So, what are some good, accessible strategies to get real value from AI / machine learning now? Here are two good use-cases you can employ in the short-run.
1. Natural Language Processing
One great application for machine learning in collections is Natural Language Processing, says Shantanu Gangal, CEO at Prodigal. “It’s one place where we have seen a lot of impact,” he adds.
And it doesn’t need to be an AI voice agent, either. Gangal compares using AI in collections to a Formula 1 pitstop. “The machines are the ones that help the agents (the drivers, in this analogy), go through the pitstop quickly.”
He continues: using AI to support agents can increase their value by allowing them to talk to more consumers. “Everything that doesn’t involve a revenue generating activity is best delegated to machines.”
2. Removing Variance and Increasing Compliance
“AI doesn’t need to be better than humans,” Gangal adds. But it can do a lot of the work, and it can be “faster, cheaper, and more consistent.”
Even the best agent can have a lot of variance and inconsistency, and machine learning and AI can help remove it. Using AI and machine learning to support agents has the benefit of adding consistency and compliance, and it’s becoming more affordable to use, as well.
Gangal notes that the cost of using tools like AI and machine learning has gone down significantly since 2001.
Examples of increased consistency include:
- Talk off guides based on data the agent collects
- Assistance with writing account notes
- Guidance through compliance scripting, like the mini-Miranda and other disclosures
The Limits of AI and Machine Learning
Chris Repholz says it’s “incumbent on us [as an industry] to educate our clients on what they (AI and Machine Learning) can do and what they can’t do.”
It’s important to remember, technology can only do what we tell it to do. If it’s programmed to be compliant, it will be compliant. The abilities of AI, especially conversational AI using Natural Language Processing, is much more limited than what a lot of people envision.
“It’s not going to solve everything,” adds Peter Lyons, Senior VP at Firstsource, like avoidance tactics, “but it has proven out quite well...in driving self service and driving cost savings and a reduction in cost to collect.”
Insight from this article comes from a recent iA Strategy & Tech’s recent three part webinar series, “How to Build a Digital RFP and Measure Success of a Digital Strategy.”
Erin Kerr is the Director of Content for iA Strategy & Tech - a digital resource for collections strategy executives - and the Executive Director of the iA Innovation Council. She is a seasoned receivables management professional, with recent experience in digital strategy and a passion for crafting digital solutions for a better customer experience.