Debt collection in finance is starting to be disrupted by artificial intelligence due to the availability of massive amounts of historical records of customers for banks Debt collection in finance is starting to be disrupted by artificial intelligence due to the availability of massive amounts of historical records of customers for banks

Artificial Intelligence can improve debt collection systems in SA

Jun 23, 2019

According to a report released in 2018 by the National Credit Regulator, of the 25 million ‘credit-active’ consumers in South Africa, a frightening 10 million were behind in their loan repayments.  Even the World Bank has deemed South Africa to be the most indebted country in the world.

While regulators and accounting rules are looking to strengthen how risk is managed, we can’t seem to stop the need to consume more credit at a household level. South Africans have been hit hard by fluctuating fuel prices in the past year. The National Energy Regulator of South Africa (Nersa) announced an increase in levies for power, which could see consumers in the red. With the increased cost of living, Debt Rescue’s Livhuwani Naledzani has noted an increase in consumers who have approached the company, asking for assistance in clearing their debt.

While President Cyril Ramaphosa and his cabinet try to salvage the economy, businesses have to start pondering on the use of collections analytics as a more efficient way of dealing with debt. The best way to balance regulations, costs, demand and capacity is analytics, in particular mathematical optimisation. Not every organisation will realise this off the bat, but those interested in achieving true competitive collections competence will look to analytic models, decision engines and automated, omni-channel communications.

FICO South Africa Country Manager Derick Cluley acknowledges the need for AI in local institutions.

“AI-powered analytics can improve automation in collections in many areas, from optimizing contact strategy settings to ensuring human agents make sophisticated decisions when restructuring debts or even calculating the right remedial action on whether to sell, keep or place debt with an external agency and if the latter, which one, at what expected account level yield?,” Cluley said.

 In a 2018 survey conducted at FICO’s FutureCollect event in Tokyo, seven out of ten senior collections managers revealed they plan to implement and integrate AI into their collections systems within the next two years.

This finding was consistent with FICO’s insights on AI last year, predicting companies will focus on operationalising AI. Almost half (48 percent) of banks believe that the use of AI will help them optimise their collection decisions, while 41 percent believe it will enable them to accurately predict consumer behaviour.

Artificial Intelligence may seem a distant wish for many, but FICO has repeatedly driven business value for clients who:

  • Can provide access to data
  • Know their objectives) – or the KPIs they wish to move
  • Understand the constraints they need to work within – including regulatory, financial, and operational constraints

And we have evidence that AI and automation works in collections. Banks that use automated collections to contact customers indicated it leads to faster bill payment. The time taken to collect payments was reduced by anywhere from two days to two weeks.

Will smart robots be contacting you in the future when you miss a payment? It’s looking increasingly likely.

 

By Doug Clare. Vice President at FICO.

  1. Opinions and Analysis

Calender

« July 2019 »
Mon Tue Wed Thu Fri Sat Sun
1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30 31