Research question and scope

This protocol measures how often a cancellation request exceeds the written save-off limit. The unit is one cancellation request tied to one account and one retention attempt. An exceedance means the offer value or duration was above the approved limit without a recorded approval. The measure describes the retention control, not the customer's decision.

Set the scope to one product, one billing system, and a fixed period. Record the approved offer table and the approval path in force. If the limit changes during the period, split the sample at the effective date. A different limit creates a different denominator.

Keep the measure separate from a churn rate. A customer may accept a compliant offer and still leave later. The purpose is to test whether the desk stayed inside its authority. Report retention outcomes separately.

Population and unit of analysis

Define the population as every cancellation request that reached the retention desk during the period. Include requests where no offer was made and requests where the customer declined. The unit is the request, not the customer. A customer who requests cancellation twice is two units.

Include requests that arrived through a reseller or payment provider, and mark the channel. Exclude cancellations completed entirely through self-service with no desk contact. Report the excluded volume. A rate built only from saved customers will overstate compliance.

Exclude a request that was later withdrawn by the customer before any offer. Record the withdrawal separately. The desk did not make a retention decision in that case. State the exclusion count beside the denominator. Record the reason for each exclusion.

Definitions that make the measure reproducible

Define the written limit as the maximum value and duration published for that product and customer segment. Define an exceedance as an offer above either dimension without an approval reference. A verbal approval is not enough unless it is recorded with the approver and time. Publish one example of a compliant and an exceeding offer.

Define the offer value as the total concession cost, not only the monthly discount. A free month, a waived fee, and a plan downgrade all have a cost. Use the client's cost model where one exists. Record the model version with the result.

Decide how to treat a package of offers that is individually compliant but jointly above the limit. Sum the value and compare it with the limit for the account. Record the combination as one offer. A set of small concessions can still exceed the authority.

Sampling and data collection

Draw the sample from the retention queue and the billing system for the full period. Record the export date, the filters, and the row count. Do not build the sample from saved accounts alone. A saved-only sample cannot show the offers that failed to retain.

Capture the request identifier, the account, the product, the offer type, the value, the duration, the approver, and the outcome. Add the cancellation confirmation time. Store the export with the report. The offer fields and the approval reference are the core evidence.

Check a small overlap sample with a second reviewer. If the reviewers disagree about the offer value, the cost model is too vague. Rewrite the rule and retest before the full sample. Agreement matters more than sample size.

Record the time spent matching offers to approvals. A report that requires reading every call note will not be repeated. Note the fields the billing system does not capture. The collection cost belongs beside the result.

Measuring offers above the limit

Report the share of offers within the limit, the share above the limit with approval, and the share above the limit without approval. Show the counts behind each share. The three groups describe different controls and should not be collapsed. A single exceedance rate hides the approved exception.

Break the result down by product, tenure, and desk shift. A new product may have an unclear limit, while a tenured account may attract larger offers. A night shift may lack a fast approver. Compare only groups with similar volume and limit.

Report the median and maximum exceedance value. A small overage is different from a concession that doubles the approved cost. Include the approval time for the exceptions. The distribution shows whether the problem is the rule or the path.

Report the later churn rate for customers who received an offer, by offer size. A large offer with high early churn is a poor use of authority. Compare it with the no-offer group where the data allows. The retention outcome belongs beside the compliance result.

Limitations and next action

The measure cannot show whether a customer understood the offer or whether the cancellation was honored. It also cannot show a concession granted outside the billing system. Treat the result as a measure of recorded offers. Note the products and channels that were not included.

Give the result to the retention owner and the finance owner with one change to test. The change might be a required offer field, a faster approval channel, or a lower limit for a specific product. Record the effective date and repeat the measure with the same definitions. A controlled change is easier to judge.

Review the boundary cases before lowering the limit. A few long-tenured accounts may explain a large share of approved exceptions. Fix the segment rule first if that is the pattern. The goal is a clear authority, not a lower number by definition.