The dunning management process
A mature process has five steps. Detect the failure. Classify the decline reason. Retry on a schedule that matches the reason. Communicate with the customer in a tone that fits the account. Escalate to a human when automation has run out of moves. Every case ends with a documented outcome: recovered, card updated, refunded, or churned.
Automation versus human touch
Automation handles volume. Most declines are ordinary and resolve themselves with a good retry schedule and a clean update-card page. But some cases need a person: large accounts, ambiguous declines, and customers where an automated dunning email would do more harm than good. A good dunning management setup makes the handoff explicit.
Metrics that actually matter
- Recovery rate on failed invoices (recovered / attempted)
- Time to recover, measured from first decline to resolution
- Involuntary churn as a percentage of total churn
- Revenue recovered per active subscription per month
- Outcome distribution: recovered, updated, refunded, churned
Common failure modes
The most common mistake is treating every decline the same way. A hard decline needs a different response than a soft one. The second most common mistake is sending automated dunning emails to enterprise contacts, which torches the relationship in ways the finance team never sees. Good dunning management routes by account value and decline type from day one.
How Chaser runs dunning management
Chaser is dunning management built for B2B SaaS on Stripe. It classifies every decline, runs the right playbook automatically, and escalates unresolved or high-value cases to a command center with full account context. You get branded save flows, operator-set thresholds, and closed-outcome reporting for every case that ever entered the funnel.