For businesses that operate on a subscription model – magazine publishers, fitness gyms, wine clubs, your TV viewing – the issue of involuntary churn is a deep problem with a long history.
That’s the situation in which a customer hasn’t cancelled the subscription but their payment method isn’t working for some reason. This type of churn costs subscription companies up to $440 billion each year, according to Butter Payments, a San Francisco startup that sells payment-recovery tools to companies that rely on recurrent revenues.
Butter says its machine-learning algorithms leverage 128 unique points of data from a customer’s card and payment patterns to better understand what may be occurring and how to resolve the failure. These tools can lead to higher card authorizations and annual recurring revenue growth of as much as 10% higher, according to the company.
Butter’s customers are primarily in the media and health-fitness industries, which rely heavily on subscription models. They include Fabletics, the athletic wear maker; online instruction firm MasterClass; and The Athletic, a sports publication owned by The New York Times. Butter either prices its software with a service fee or can keep a percentage of the revenue it collects on behalf of a customer.
Butter has 32 employees and has raised $37.3 million to date. Charles Rosenblatt, Butter’s former chief commercial officer, was named CEO in December.
Editor’s note: This interview has been edited for clarity and brevity.
PAYMENTS DIVE: What’s the source of most card failures?
CHARLES ROSENBLATT: Non-sufficient funds is usually the leading source, but another one of the big ones is suspected fraud. I’ll give you a great example. A lot of subscription companies here in the U.S. will try [to make] the transaction at 2:30 a.m. in the morning. And so what happens? The processors flag that as potential fraud. And some of them not only do the first transaction at 2:30 in the morning, they set it up that all the retries are at 2:30 in the morning, so they’re just perpetuating the problem.
Collecting owed money sounds like a core finance department function, what every company must do, not something for a vendor.
We’ve spent six years and have billions of dollars of transactions to make decisions off of. So, we can predict these things and we know the best way to get that, versus individual companies who, one, solely have their own data, and two, don’t want to build the technical resources on which to do this. We charge for incremental (recovery). We say ‘If you think your internal people can save 15 of the 300 people who fail, great, go ahead and do it. We’re going to only charge you for the 16th and beyond that we’re saving.’
How do companies address the costs of providing a service or goods or ending a subscription after the payment failure?
The period post-rejection, before you send it to a collections agency who threatens to steal your kids and all that kind of stuff, before you get to that is called a dunning window. It’s a period of time you’ve chosen, which could be seven days, ten days, 20 days. There is the subscription and actually sending the product or turning off the subscription. Most companies technically have not figured out how to decouple those two things.
Their technology doesn’t allow them to flip a switch and turn off the product at the same time. Unless they want to make a unilateral, across the board decision: The moment the payment fails, we shut [the subscription] down, then we [seek payment]. It doesn’t let them make variable decisions. And the reality is, in some cases … it is a really bad customer experience for the 85% of the people who are going to stick around, but just had a payment fail. It’s not worth making a terrible experience for them, [just] for the 15% I’m giving a free ride.
What accounts for companies being unable to split that decision process for their different customers based on what they know about them?
For subscription companies, you have a payment processor – Stripe, Adyen, Global Payments, whoever. And then you have this other entity in the middle called a billing platform. The reality is, the decision [on] what you do on the product sometimes lies in a different area. You are likely in a series of vendor decisions. That’s why we partner with those billing systems. Our APIs meld into those systems, which allows us to manage those systems in coordination. Because most of those companies don’t do what we do. We are another added-on service.
Has a company like Stripe or anybody else made an offer to buy Butter?
We get approached quite often by folks. As with any company that has a Series A [funding], I have investors who always would like a favorable outcome.
