Dunning and Involuntary Churn Recovery (Retention)

Dunning and Involuntary Churn Recovery (Retention)

1. 01

   Pull the trailing 90 days of failed payment events and segment them by failure code

   Separate hard declines (expired card, closed account) from soft declines (insufficient funds, issuer timeout). Soft declines recover at a materially higher rate and deserve a different cadence than hard declines.

2. 02

   Map every failure code to the gateway response the client's stack actually returns

   Chargebee, Recurly, and ChargeOver each surface decline reasons differently, and a code that reads as soft in one gateway can be hard in another. Normalize the taxonomy before writing any retry logic.

3. 03

   Set the retry ladder by decline type, not by a single global schedule

   A common pattern is three retries across 10 days for soft declines and a single retry plus card-update prompt for hard declines. Document the ladder in the client's billing runbook so support staff do not improvise.

4. 04

   Sequence the outreach: in-app prompt, then email, then a human touch before suspension

   Card-update links embedded in the first email recover more accounts than a bare past-due notice. Reserve the account manager call for accounts above the client's stated revenue threshold.

5. 05

   Define the suspension and reactivation rules in writing with the client before go-live

   Grace period length, data retention after suspension, and whether reactivation restores prior entitlements are the three questions that generate the most client escalations. Get sign-off in the retainer scope.

6. 06

   Instrument recovery reporting so the client sees recovered dollars, not just failure counts

   Report recovered MRR, average days to recovery, and the share of failures that never recovered. This is the number that justifies the retainer line item at renewal.

7. 07

   Review the ladder quarterly against actual recovery data and adjust thresholds

   Retry timing that worked at 200 subscribers can suppress recovery at 5,000 because issuer behavior and card expiry distributions shift with the customer base.