Research question and scope
This protocol measures how often an attachment reaches a delegated support inbox without a recorded safe-handling decision. The unit is one attachment-ticket pair. A missing decision means the ticket does not show whether the file was classified, opened in an approved tool, or routed to a specialist. The measure describes the intake control, not the intent of the specialist.
Set the scope to one support queue, one case system, and a fixed period. Record the attachment types the queue accepts and the handling rule in force. If the rule changes during the period, split the sample at the effective date. A different rule creates a different denominator.
Keep the measure separate from an incident count. A file can be handled correctly without becoming an incident. The purpose is to find where the record is silent. A neutral framing helps the team report the gaps without fear of blame.
Population and unit of analysis
Define the population as every ticket in the period that carried at least one attachment. The unit is one attachment-ticket pair, not each file inside an archive. A ticket with three separate documents is three units only when the system records them separately. State the rule before sampling.
Include tickets that were closed without opening the file, and mark them as held or routed. Include tickets where the attachment was deleted after review. The unit still exists if the ticket records the decision. Exclude tickets with no attachment evidence at all.
Exclude internal files created by the support team, such as saved replies or screenshots of the case system. The measure is about files arriving from outside the team. Keep the boundary visible in the report. State the exclusion count with the denominator.
Definitions that make the measure reproducible
Define a handling decision as a recorded classification plus a recorded action. The classification may be routine, sensitive, unexpected, or blocked. The action may be open in viewer, store with case, route to specialist, or quarantine. Publish one example of each.
Define a missing decision as a ticket with an attachment and no recorded classification or action. A note that says reviewed without naming the action is not a complete decision. A decision that was made in a chat outside the case system is also missing from the record. The case record is the unit of evidence for the main rate.
Define a risky attachment as a type or sender pattern listed in the handling rule. The list may include archives, executables, password-protected files, and unsolicited payment instructions. Use the rule in force on the ticket date. Do not apply a later rule retroactively.
Sampling and data collection
Draw the sample from the case system for the full period rather than from security reports. Record the export date, the filters, and the row count. Include every channel that feeds the queue. A sample built from reported incidents cannot produce a handling rate.
Capture the ticket identifier, the received time, the sender type, the file type, the classification, the action, the reviewer, and the close time. Add the storage location and any routing reference. Store the export with the report. The classification and action fields are the core evidence.
Check a small overlap sample with a second reviewer. If the reviewers disagree about whether the decision is complete, the definition of action is too vague. Rewrite the rule and retest before the full sample. Agreement matters more than sample size.
Record the time spent matching attachments to tickets. A report that requires opening every file will not be repeated. Note the fields the case system does not capture. The collection cost belongs beside the result.
Measuring missing safe-handling decisions
Report the share of attachment-ticket pairs with a complete decision, the share with a partial decision, and the share with no decision. Show the counts behind each share. The three groups describe different levels of control and should not be collapsed into one rate. A partial decision is not the same as an open file.
Break the result down by file type, sender type, and queue. Archives may be handled differently from invoices. An unknown sender may trigger a check that a known customer does not. Compare only groups with similar volume and risk.
Report the share of risky attachments that reached a ticket without a stop or route. Include the disposition of each one. A risky file that was opened in the approved viewer is different from one that was downloaded. The disposition shows where the rule actually holds.
Limitations and next action
The measure cannot show whether a file was malicious or whether a specialist opened it outside the approved tool. It also cannot show a decision made verbally and never recorded. Treat the result as a measure of the case record. Note the channels and file types that were not included.
Give the result to the support lead and the security owner with one change to test. The change might be a required classification field, a safer intake form, or a weekly sample. Record the effective date and repeat the measure with the same definitions. A controlled change is easier to judge.
Review the boundary cases before changing the whole rule. A few senders with repeated archive files may explain a large share of the gaps. Fix the sender path first if that is the pattern. The goal is a record that shows the decision, not a lower number by definition.
