Small-business outsourcing research
How often do failed subscription payments recover without manual contact?
A protocol for measuring how many failed subscription payments recover through the retry schedule before a person has to intervene.

Research question and scope
This protocol asks what share of failed subscription payments recover without a manual contact from the business. The unit is one failed payment attempt, not one customer and not one subscription. The study measures recovery after the first failure, the attempts used, and the payments that needed a person to resolve. It does not judge whether the customer should have been retained.
Set the scope to one business, one billing system, and one retry schedule. Record the processor, the retry timing, the message sequence, and the service-suspension rule. If the schedule changes during the window, keep the original rule for failures already in progress and analyze later failures separately. A new card-update page can improve recovery without any change in customer intent.
Define the population as every failed payment attempt during the window, including failures that were later reversed and failures on accounts that were already past due. State any exclusion, such as a test transaction or a fraud block, and apply it to every attempt. Test the definition on two billing cycles before collection begins. The reader should be able to rebuild the population from the processor export.
Definitions that make the measure reproducible
Define recovery as a successful payment for the same subscription and billing period after a failure. Record whether the recovery happened on the first retry, a later retry, or after a manual contact. Define manual contact as any outbound message or call initiated by a person rather than the scheduled sequence. An automated reminder is not a manual contact even when a customer replies to it.
Use the billing period and the amount to match a recovery to the failure. Do not count a payment for a different plan or a partial payment as a full recovery unless the business rule allows it. Record the failure code from the processor and the date it occurred. A generic decline code is less useful than a specific one, and the protocol should say which code was available.
Define the retry window as the period from the first failure to the last scheduled attempt. Record the time zone used for the schedule and whether weekends and holidays were skipped. A payment that succeeds after the window is a late recovery, not an in-window recovery. The distinction matters because the customer may already have lost service.
Evidence standard and data dictionary
For each failed attempt, capture the subscription identifier, the customer segment, the plan, the amount, the failure date, the failure code, and the retry dates. Add the recovery date, the attempt number, and whether a manual contact occurred before recovery. Record the message versions sent and the channel used. Keep the processor transaction reference linked to the billing record.
Capture the reason for manual contact when it is documented: a customer reply, a hardship request, a suspected fraud, or a repeated failure. Do not infer a reason from the absence of a scheduled message. If the reason is unknown, mark it unknown and report the count. A guessed reason makes the automation look more effective than the evidence allows.
Record whether service was suspended, when, and who approved it. Note any refund, credit, or goodwill adjustment tied to the failure. A recovered payment that follows a service suspension is a different outcome from one that never interrupted access. Keep those events in the same record so the analysis can separate them.
Sampling and collection
Collect every failed attempt in the window rather than a sample, because the retry rule applies to each attempt. If the business has several plans or billing platforms, collect each separately and record the group. A combined rate can hide one plan with a persistent card problem. Report the number of attempts and the value behind every percentage.
Draw dates from the processor export and the billing system audit log, not from a spreadsheet maintained after the fact. If the audit log is incomplete, say so and reduce the claim. Record the export date, the filters, and any attempt that could not be matched to a subscription. A missing audit trail is itself a finding about the control.
Ask the billing lead to identify weeks when a processor outage or a holiday changed the retry timing. Those weeks can produce failures that have nothing to do with the schedule. Record the event and analyze those attempts separately. The reader should be able to tell whether the rate reflects the rule or the calendar.
Analysis and reporting
Report the share recovered without manual contact, the share recovered after contact, and the share that never recovered. Show the median number of attempts and days to recovery. Present the results by failure code, plan, and retry sequence. Use the same denominator for every rate.
Separate a recovery that happened on the first retry from one that required the final attempt. A schedule that recovers most payments late may still create a poor experience and a higher support load. Report both because they lead to different fixes. The business may accept a late recovery but should know what it costs.
Inspect the failures that never recovered and describe which stage failed without naming the customer. A missing card-update path, a blocked message, and an exhausted schedule each point to a different repair. Compare the outcome with the service-suspension rule and the customer's reply pattern. Report attempts with incomplete timestamps rather than dropping them from the denominator.
Limits of the evidence
The processor record can show that a payment succeeded, but it cannot show why the customer updated the card. A customer may have acted because of the message, a bank alert, or a conversation the business never saw. The protocol also cannot capture a phone call that was not logged. State that limitation before presenting the rate.
A small business may have too few failures to support a stable trend. Describe the attempt count and the plan mix instead of ranking the billing team. If one owner handles every exception, the result may reflect that person's availability more than the process design. Do not use the rate as a performance score without the context of volume and seasonality.
Do not compare a business that retries for two weeks with one that suspends after one failure. The two operate different schedules even when the recovery rate looks similar. Report the retry window and the suspension rule beside the rate. A target that can be met by delaying service for every customer is not a good target.
What to change
Give the result to the owner and the billing lead and ask them to choose one observable change. The change might be a clearer card-update link, a different retry spacing, a pre-expiry reminder, or a named exception owner. Record the owner, the effective date, and the measure that will show whether the change worked. Repeat the protocol with the same definitions after the change.
If manual recovery rises while automatic recovery falls, the messages may be reaching fewer customers even though the team is working harder. If both rise but complaints also rise, the schedule may be too aggressive. Report the movements together before declaring success. Close the review when the owner accepts the finding and the next cycle date is set.
Ask the billing lead which failure was hardest to recover and why. A concrete answer often reveals an expired card, a bank block, or a customer who could not find the update page. Use that answer to improve the path rather than to blame the customer. The next measurement should test whether the same delay shrank.
Sources
- Federal Trade Commission, negative option rule, Federal Trade Commission; checked 2026-11-17
- Federal Trade Commission, advertising and marketing guidance, Federal Trade Commission; checked 2026-11-17
- U.S. Small Business Administration, manage your business, U.S. Small Business Administration; checked 2026-11-17
- U.S. Government Accountability Office, internal control standards, U.S. Government Accountability Office; checked 2026-11-17
Frequently asked questions
Does a favorable sample prove the routine caused the result?
No. The design is observational and supports a bounded operating decision, not a universal causal claim.
What should an owner review first?
Start with high-consequence defects, missing denominators, exceptions, and cases that required the owner to reconstruct the evidence.
Related research
What share of customer referrals are rewarded within the stated window?
A protocol for measuring how many qualifying customer referrals receive their reward inside the window the business promised.
What share of lost-package claims close inside the carrier claim window?
A protocol for measuring how many lost-package claims reach a final carrier decision before the claim window closes.
How long do job applicants wait for a first response from a small business?
A protocol for measuring the time between an application and the first substantive response from a small-business hiring process.