Research question and scope

This protocol measures how long a correction memo takes to close a quality failure. The clock starts when the failed check is recorded and stops when the recheck passes. The unit is one memo linked to one failed task. The measure describes the correction process, not the severity of the original error.

Set the scope to one team, one review standard, and a fixed period of failed checks. Record the memo template, the repair owners, and the recheck method used at the time. If the template changes, split the sample at the change date. A new template can change both the speed and the quality of the closure.

Keep the measure separate from a disciplinary process. A memo that is delayed because the repair needs a system change is different from one that is delayed because no owner was named. The purpose is to find where correction stalls. State that purpose in the report.

Population and unit of analysis

Define the population as every failed check that required a correction memo during the period. The unit is the memo, not the message or the meeting. One failed task may generate one memo with several actions. Count the memo once and note the number of actions separately.

Include memos that were reopened after a failed recheck. A reopened memo is still one case, but the reopening is part of the cycle time. Record the number of rechecks and the reason for each reopening. A memo that closes and reopens is not the same as a memo that closes once.

Exclude a failed check that was corrected immediately before any customer impact and did not require a memo. Report those cases separately as same-shift corrections. The two paths have different evidence requirements. Mixing them would make the memo process look slower than it is.

Definitions that make the measure reproducible

Define the start as the timestamp of the failed check in the review record. Define the end as the timestamp of the passed recheck, not the date the memo was filed. Define the repair as the change to the guide, form, system, or training material. Publish one example of each timestamp.

Separate elapsed days from working days. A memo that spans a weekend may look slow in calendar days and normal in working days. Choose one calendar and use it for the whole sample. Report the choice with the result.

Define waiting time as the period when the memo was blocked on an approval, a system change, or missing evidence. Define active time as the period when the repair owner was working on it. The two measures need different fixes. Use a short reason list to classify the waiting.

Sampling and data collection

Draw the sample from the review ledger rather than from a list of memos that were easy to close. Include memos that are still open at the observation date. Record the export date and the filters. A sample of closed memos only will understate the cycle time.

Capture the memo identifier, the failed check, the start and end timestamps, the repair owner, the recheck owner, the waiting reason, and the final status. Add the number of rechecks and the customer impact. Store the evidence with the record so the classification can be checked. The missing field is often the first clue about a stalled repair.

Interview the repair owner for a small sample of long memos. Ask what the owner was waiting for and what would have shortened the wait. The answer may reveal an approval step that is not visible in the ledger. Record the answer beside the timestamp.

Check whether the recheck used a case that would have failed under the old process. A recheck on an easy case does not prove the repair. Record the case type and the result. A closure without a valid recheck should be treated as open.

Measuring correction cycle time

Report the median and the longest cycle time, the share of memos closed within the target, and the share still open at the observation date. Show the counts behind each share. A median alone can hide a long tail of blocked memos. The open share is part of the result.

Break the result down by cause, repair owner group, and work type. A cause concentrated in system changes may need a different planning step. A work type with many rechecks may need a clearer standard. Compare groups with similar complexity and avoid ranking individuals.

Report active time and waiting time beside the total. A team that repairs quickly but waits weeks for approval has a different problem from a team that works slowly. The split shows where to intervene. Use the same reason list for every memo in the sample.

Limitations and next action

The measure cannot show the quality of the repair or whether the same error returned months later. It also cannot show the time spent on memos that were never recorded. Treat the result as a measure of the recorded correction path. Note the work that was corrected informally.

Give the result to the review owner and the repair owners with one change to test. The change might be a named approval path, a standard recheck case, or a shorter memo template. Record the effective date and repeat the measure with the same definitions. A small change to the blocking step is often more useful than a broad reminder.

Review the reopened memos separately and look for a repair that did not address the cause. A fast closure with a failed recheck is not a good result. The goal is a correction that holds, even when the memo takes a little longer to close. A repair that only shortens the memo is not a repair.