Failed payments are 20 to 40 percent of all churn and the cheapest to fix. Retry by decline code and refresh cards before they fail.
Frequently asked
Related resources
Solutions
See it in the product
The same loops this note describes run 24/7 against your customer base. Watch the workspace decide, experiment and execute 1:1.
Vocabulary
Definitions an answer engine can quote, each one a page of its own.
Compare
Neutral reads on the categories buyers evaluate against a decision layer.
All comparisonsDunning is the process of recovering failed recurring payments through retries, card updates and customer contact. The best practices that move recovery rates are unglamorous: retry by decline code rather than by calendar, keep card credentials fresh before they fail, and measure dunning separately from voluntary churn so neither pollutes the other.
Failed payments are the largest recoverable revenue leak in most subscription businesses, typically 20 to 40 percent of all churn. Yet dunning is often run as a fixed email sequence designed years ago. The split between this involuntary churn and deliberate cancellation is covered in voluntary vs involuntary churn; this article is the operational playbook.
A fixed retry schedule (day 1, 3, 5, 7) treats an insufficient-funds decline the same as a stolen-card decline. They are opposites. Soft declines, such as insufficient funds or issuer timeouts, recover well with retries timed to when money arrives: month-end, known paydays, local payroll cycles. Hard declines, such as stolen or closed cards, never recover by retrying and only generate fees and issuer distrust. Route on the code:
The cheapest failed payment is the one that never happens. Account updater services from the card networks refresh expired or reissued credentials automatically, and network tokens replace the card number entirely so reissues do not break the mandate. Together they prevent a third to a half of failures in mature implementations. Pre-dunning email, asking customers to update an expiring card two weeks ahead, adds a few more points at near-zero cost.
How long service continues after a failed payment is a policy decision with real economics. Cutting access immediately maximises urgency but burns goodwill with customers who had no idea their card failed. A grace period of three to seven days with clear, non-threatening messaging recovers more at lower reputational cost. The message matters: 'your payment did not go through, update your card' outperforms anything framed as a warning, because most of these customers intend to pay.
And keep dunning out of the retention team's numbers. When involuntary churn is pooled with voluntary churn, retention programmes get credited for payment recoveries and blamed for card expirations, and both teams optimise the wrong thing. The same separation discipline applies to save flows at the point of cancellation, covered in churn deflection.
Markin treats payment recovery and retention as separate decisions with separate holdouts, so each is measured on the outcomes it actually caused. See Retention decisioning or the fintech view.
