Harness Better Tools Powered by AI to Quickly Maximize Recurring Revenue

The Impact of Failed Customer Payments and Passive Churn

Failed customer payments for recurring subscriptions are often seen as a cost of doing business when using subscription revenue models. Payment failure rates approaching 25% of recurring transactions are not uncommon. And these failures can happen for a variety of reasons, including an expired credit card, an incorrect billing address or insufficient funds, for example. Regardless of the cause, the consequences for the business go beyond a missed payment event. Often referred to as passive churn, these failed payments can severely affect the entire economics of the business, and therefore treating these as unavoidable can be a very expensive mistake. The impact of failed customer payments should be of concern to many in the organization including the CEO, CFO, CMO and COO. Today, complementary third-party technologies exist that address the issue in a comprehensive way and that can have a meaningful impact on the bottom line.

These failed payments can severely affect the entire economics of the business, and therefore treating these as unavoidable can be a very expensive mistake.

Traditionally, organizations have approached failed payments using a standard escalation process. The first step is to retry the payment through repeated attempts by the processor and the merchant billing platform. If that does not work, most organizations then attempt to contact the customer via email or text. A further escalation is then to move to a phone call and a high-touch recovery process via the customer service team. This approach may achieve the recovery of that payment, but it comes at a high cost in personnel time. Furthermore, this approach may not necessarily produce a lasting recovery and can instead be an indicator that the customer will require future manual, costly efforts.

Failed payments are not just about revenue leakage; they have a fundamental impact on the organization’s revenue base. There are direct costs associated with failed payments—labor and lost revenue—but these only make up part of the complete picture. The economics of subscription business models are different from repeated one-time sales, as revenue is now spread over the lifetime of an engagement. A defined point exists at which the customer becomes profitable given the acquisition cost, cost to serve and cost of goods. Failed payments disrupt this timeline, since they interrupt revenue and since recovery costs delay the breakeven point, potentially to the point that the customer is never profitable. Manual recovery costs impact the profitability of the specific subject customer in this way, but they also have broader impacts on broader profitability initiatives. For instance, there is a high opportunity cost when tying up customer service resources with payment recovery tasks rather than allowing them to work on upselling or other incremental revenue activities.

While many organizations have some automated payment retry processes built into their billing platform, new tools further address failed payments by using AI-driven techniques to complement these existing recovery processes. Vendors of these AI-based systems train models on their extensive histories of payment successes and failures to predict the best recovery methods to use. These algorithm-driven applications reduce passive churn and enable organizations to reduce the cost of repeated additional manual outreach and resolution. Labor can then be freed up to focus on other revenue-generating activities. These recovery strategies minimize the cost of remedying failed payments and accelerate the ROI of each individual customer.

Augmenting Existing Processes with New AI-Based Tools

Vendors are developing AI-based systems that complement existing standard automatic retry capabilities offered by payment processors, further reducing the need for high-cost customer service personnel to be involved in high-touch payment recovery workflows. These applications can improve the incidence of automatic recovery beyond the rates resulting from retry efforts based on payment system capabilities. Using machine learning on large sets of historical responses and reason codes from many different merchants across payment providers, specialist recovery vendors can better categorize failures as “soft” or “hard” failures. Wider comprehension of network and merchant advice codes coupled with real-time account update information results in more payment failures that are able to be automatically resolved. Smarter retry strategies can be automatically applied at scale in ways that are beyond single processor and merchant capabilities. In addition to making the cost of recovery lower, this also frees up customer service agents to spend more time on positive customer engagement activities that retain customers and grow product and service revenue.

If the recovery can happen automatically, customer service does not need to pursue a recovery workflow that the customer will likely perceive as negative.

For example, in the case of “box subscription” companies that ship physical goods at a periodic rate, failed payments are problematic because the payment must usually be recovered before the box can be shipped. But it is also important that the box be shipped on time to keep the customer happy. If the recovery can happen automatically, customer service does not need to pursue a recovery workflow that the customer will likely perceive as negative. Instead, they can engage in a positive way with a promotional campaign opportunity or something similar.

These systems perform automated follow-on steps for failed customer payments and can also help provide pre-emptive future resolution. Retry logic within an organization’s existing systems will yield some amount of recovery, but third-party AI-driven algorithms, trained on much larger data sets than those found within any individual organization, can drive insight to resolve more error types and pursue a wider range of recovery methods to solve not only for the one failed payment, but also to proactively heal future payment failures that are related to the current issue. In most cases, the specific error will not typically occur again, thus limiting passive churn and solving the missed payment problem for that specific customer unless a different issue arises. Improving the resolution rate of payment failure and involuntary churn in this manner is cheaper than trying to acquire new subscribers to continue revenue growth.

Consider another example in which a media subscription service costs $14.99 per month. Using standard customer acquisition costs of $70 per account and $40 for cost of goods sold, the break-even point for a customer in this model is month seven. With an average consumer subscription lasting only seven months, the profitability of a customer will be called into question if payment failure recovery is either unsuccessful with standard retries, or if human outreach is utilized. Even assuming an average cost of only $15 for manual recovery, this activity pushes the customer into loss territory. When more failed payments can be resolved via an automated service (and proactively healed so as not to fail again), the organization reduces the duration of the subscription needed for these customers to cross the break-even point to profitability.

The Role of Automated Recovery

It is important to reiterate that these new systems are not a replacement for an organization’s internal billing and retry processes, nor are they meant to remove active churn or to compensate for poor products and services. Instead, these systems should be considered a way to top up those preexisting processes and to maximize the revenue recovery that would otherwise need to be handled by customer service. These systems also play no role in the billing cycle, so the algorithms have no visibility into potential customer issues until the recurring cycle results in an actual payment failure. They are also not meant to lead to a head count reduction. Rather, they should lead to increased headcount effectiveness that can affect top-line growth.

According to our Analytics and Data Benchmark Research, nearly nine in 10 organizations (87%) either already use or plan to adopt AI/ML, and yet, 65% of organizations feel they don’t have enough skilled AI/ML resources available. Applications that harness AI in an automated fashion can relieve some of the burden in that discrepancy since the tools do not require special “AI skills” from the teams employing them. But it is not enough to simply have skilled resources since effective AI tools equally depend also on data and domain expertise. Specialized AI-assisted payment recovery applications from established vendors benefit from an extensive historical data set that crosses industries and payment processors, offering a much richer training set than that available to an individual organization. These pre-packaged payment recovery solutions can immediately add value since the service developer has tested and validated which AI algorithms are best used. This approach removes the need for the organization itself to build, test and maintain proprietary tools without technical AI and skilled data science resources.

Organizational Impact of Revenue Performance Improvement

For subscription businesses, solving failed payments is not just a matter of cost avoidance. Rather, it is important to the viability of an organization. In digital publishing, for instance, magazines and newspapers can see significant impacts on subscription numbers by failed payments because customers are more prone to dropping a service that encounters problems. Profitability is already a fine line for these organizations given the shift in how consumers now acquire information, so any degraded part of the customer experience comes with heightened risk to subscription counts.

Given the importance of these retention activities to the organization, it makes economic sense to outsource otherwise simple activities such as recovering failed payments to an automated system.

As mentioned, retaining a customer is far less expensive than acquiring a new one. Depending on the industry, the expense of acquiring a new customer can be quite significant. Consumers tend to buy from brands they trust, and converting a new customer from a competitor takes more effort than maintaining the loyalty of an existing one. Resources across the sales, marketing, customer migration and customer success teams must be engaged, not to mention the cost of marketing across multiple platforms. Retaining a customer, however, can be achieved by focusing that level of effort on providing a superior customer experience. Given the importance of these retention activities to the organization, it makes economic sense to outsource otherwise simple activities such as recovering failed payments to an automated system. In most scenarios, the improved recovery comes at fraction of the cost of advanced manual efforts.

In addition to eliminating the recovery costs that accrue with the use of live agents, using automated payment recovery applications frees up customer service resources to pursue revenue generating activities such as promotional outreach and bolstering customer loyalty.

The impact of an efficient and effective payment recovery reaches far more than just finance. It is fundamental to the sustained economic proposition of the subscription business model and will therefore impact the health of the entire organization.

Key Takeaways

• Treating failed customer payments as simply “the cost of doing business” in a subscription model business can be a very expensive mistake. Manual recovery efforts push an individual customer’s point of profitability out in time, potentially even to the extent that they are never actually profitable.
• Organizations must consider whether their automated efforts and manual customer service interventions are yielding recovery results that are timely and profitable. In particular, examine whether manual interventions are pursued at too high of an opportunity cost when considering how else the customer service team can be utilized to improve revenue performance.
• Third-party AI-backed tools are trained on massive data sets and can more comprehensively address failed payments and reduce future payment failures. These tools are designed to supplement current systems, process and personnel.