Self service portal vs. machine learning-based collections

By on December 6th, 2016 in Industry Insights, Machine Learning
TrueAccord Blog

Consumer behavior is changing. As more of us are glued to our mobile phones, emails, and social media accounts, it’s clear that the old ways of collecting debt are quickly becoming irrelevant. Still, the market doesn’t offer a multitude of collection solutions aimed at responding to the digital consumer. When we present our machine learning-based solution to prospective customers, we’re often asked about the difference between our solution and a self service portal. Although both solutions are digital, they cannot be less alike.

Passive or active? Proactive and reactive strategies in collections

A self service portal is a passive asset, available for consumers to use when they choose to. Often, consumers choose not to use the portal since they would rather not deal with their debt conundrum. It’s only human to do so. Since the portal is passive, the consumers who find it are those who look for a way to pay online, without talking to an agent. They are a minority, and their liquidation is far from worth the price of implementation. In fact, many or the collectors and creditors who use online portals have accepted that less than 10% of their paid dollars may come from their self service portal. Others have given up completely. Compare that with TrueAccord’s whopping >90% paid through our online portal.

A machine learning-based system is an active one. Our collection engine creates collection strategies that are both proactive and reactive. The system proactively reaches out to the customer, automatically and using pre-written content, across multiple channels, to get them to engage online. Every communication has a link to the online and mobile-optimized portal, and phone calls are aimed at supporting this strategy rather than the only tool for contacting the consumer. Once the consumer interacts with our communication and browses the website, the system tracks their behavior so it can react to it – personalizing follow up flows based on emails they opened, pages they visited or words they said to agents on the phone.

Cross channel integration and halo effect

Using a machine learning based collection system has another advantage over passive a self service portal: it is integrated across all channels. Self service portals are often an add-on, their data is separate from the main collection system and dialer. They exist to supplement the main dialing strategy, often giving consumers the feeling that they are “fleeing” the collectors to use the online portal. It’s a wasteful strategy that also yields consumer complaints.

A machine learning based system harvests the halo effect, that causes consumers to react better when they are contacted across multiple channels. For example, for some groups, following up on an email with a single call attempt at the right time and number is twice as effective in getting a response as sending another email. For others, calling within five minutes of a consumer browsing through payment options without paying increases liquidation significantly. Shifting between channels increases effectiveness but also increases efficiency, reducing outbound call volume by up to 95%.

Bottom line

Machine learning based solutions are significantly more effective than a passive, stand alone self service portal. More than 90% of paying accounts in the TrueAccord system are resolved using the online service – and liquidation beats traditional agency performance by 30% and more. Using data and the maturing world of machine learning algorithms is the future of debt collection – and we’re happy to help.