Treasury News Network

Learn & Share the latest News & Analysis in Corporate Treasury

  1. Home
  2. Payments - Receipts
  3. Accounts Receivable Management

Machine learning drives payment productivity in “End of ownership”

A fundamental change in what consumers and businesses want and how they behave - The FT call it “The end of ownership” -  is happening at ever increasing speed. Instead of owning music tracks, cars, bikes, etc., they are paying a (typically) monthly subscription (recurring payment). This then makes the payment systems and services that collect the monthly payments absolutely vital. Any declines in monthly payment collection efficiency can spoil the profitability of the whole business model.

Understanding subscribers

Understanding the ebbs and flows of subscribers behaviour is complex. Not surprisingly, as with other areas of payment collections, service supplier platforms are turning to artificial intelligence machine learning technologies to improve efficiency, e.g. reduce declines.

Recurly explain that, “Every subscription business encounters credit card declines. These declines increase churn, reduce revenue, and can negatively impact subscriber relationships. But, with the right subscription management platform, you can minimize their impact.”

Recurly, like the other subscription plaforms, work with thousands of subscription businesses which come from a wide range of industries, which gives them:

  • access to hundreds of millions of data points that encompass billions of attributes from many different types of companies in both B2B and B2C categories
  • understanding and expertise in recurring payments.

This is where AI machine learning comes in because it enables the patforms to learn, and iterate from this experience and improve understanding. This is the basis for the new Recurly Revenue Optimization Engine. 

Revenue Optimization Engine

Recurly explain that, “The Revenue Optimization Engine takes a highly tailored approach to each invoice, using statistical models and machine learning to improve collections -- increasing our customers’ monthly revenue by an average of 9%.

Credit cards and recurring transactions

Credit and debit cards are the most common method of payment in the U.S., used in a wide variety of transactions. In subscription commerce, these transactions are recurring transactions where the card is charged anew each billing cycle. Recurly has found that on average, 13% of recurring transactions are declined. 

Recurring transactions may be declined for a number of reasons including insufficient funds, credit/debit card restrictions, technical issues, or other reasons beyond the subscriber’s control. These declined transactions can lead to involuntary subcriber churn. 

How it works

Every declined transaction is different, which makes a static, one-size-fits-all retry schedule less effective. Recurly’s Revenue Optimization Engine uses machine learning “to craft a more intelligent retry schedule that is specifically tailored to each individual declined transaction.”

Using this technology, subscription businesses recover an average of 61% of failed subscription renewals. This increased payment success rate increases revenue and decreases involuntary subscriber churn that results from failed transactions.

Recurly claim that by creating and utilizing a tailored retry schedule:

  • It helps resolve more credit and debit card payment issues in a shorter timeframe than a static schedule can
  • subscribers benefit from having uninterrupted access to your product or service. You don’t have to suspend their account while payment issues are resolved.
  • subscribers receive fewer communications regarding payment issues, such as asking them to update their billing information.

CTMfile take: AI machine learning is a key ingredient in improving all types of payment collection, particularly for recurring payments.

Like this item? Get our Weekly Update newsletter. Subscribe today

Also see

Add a comment

New comment submissions are moderated.