Research ·
SLA pause accounting and excluded time in outsourced support
How pause rules and excluded time change what a service target can show, and how to report the customer's real wait.

Key stats
time figures reported
pause rule per review
days before comparison
Methodology and findings
Question and scope
This review asks how pause time and excluded time change what a service-level measure can support in an outsourced help desk. The population is ticket queues with a written service target and a status model that stops the clock. The unit is one ticket with at least one pause interval during a declared observation window. The question matters because a target that excludes time can look healthy while the customer is still waiting.
The scope excludes targets based only on first response and excludes queues with no recorded status history. It also excludes time spent in a client-owned system that the desk cannot observe. Those boundaries keep the review on pauses the desk can document. A finding should help a reader compare reported performance with the customer experience.
What a pause actually measures
A pause records that the desk was not expected to act, not that the customer stopped waiting. The clock may stop when the ticket waits on the customer, a vendor, or an approval. The customer continues to experience elapsed time even when the service target does not. A report that shows only active time hides that difference.
Pause reasons are not equivalent. Waiting for a customer reply is different from waiting for a client engineer who has not responded. Waiting for a scheduled maintenance window is different from waiting for an approval that nobody owns. Grouping all pauses under one label removes the information a manager needs.
Counting rules for excluded time
Write the start event, stop event, and rounding rule for each pause reason before extracting data. A pause that starts when an agent changes a status is not the same as one that starts when a message is sent. Record which clock the tool uses, because the queue clock and the shift clock often disagree. Ambiguity in the rule becomes a difference in the reported result.
Cap the pause at the point where the desk should have followed up. A ticket waiting on a customer for thirty days is not paused in any meaningful sense. Set a follow-up interval and count the time after it as active or as a separate aged state. The cap should be written into the target, not applied after the report looks bad.
Where pause evidence breaks down
Status changes are often entered at the end of a shift rather than at the moment they happen. That delay moves time between active and paused without changing the work. An agent who clears several tickets at once can create a cluster of pauses that never existed. The record shows the entry time, not the real transition.
Some tools overwrite the current status and keep no history. Without history, the desk cannot reconstruct how long a ticket waited or who moved it. Export the status log before a tool migration and test that the export includes the pause reason. A measure built on a field that is rewritten is not comparable across periods.
Reading pause ratios with case mix
A high pause ratio can mean good discipline or a queue that depends on other teams. Compare the ratio within the same request type and the same client before drawing a conclusion. A billing queue that waits on approvals will look different from a password queue that rarely pauses. The mix explains more than the average.
Look at the distribution, not only the mean. A few tickets with very long pauses can lift the ratio while most tickets move quickly. Report the median and the count of tickets above a threshold. The reader should see whether the pause is routine or concentrated in a small set.
Reporting pause time to the client
Report active time, paused time, and total elapsed time as three separate figures. Name the pause reasons and the rule used to cap them. Show the customer-facing elapsed time even when the service target excludes it. A client who sees all three can judge the service rather than the formula.
Explain any change in the pause rule before presenting a trend. A rule change can move the result without any change in the work. If the old rule was wrong, say so and restate the period where the data allows. A clear note protects the credibility of the current number.
Limits of the finding
The review cannot measure the customer's actual wait when the ticket moves between channels. It also cannot recover transitions that the tool did not record. The result describes the pauses that were documented, not every pause that occurred. The desk should state that limit in the report. Report the channel change beside the wait figure so the reader can see the gap.
The finding is affected by reviewer judgment when the pause reason is chosen from a list. Calibrate the list with examples and check a sample of entries. A reason that agents interpret differently cannot support a comparison. Record the calibration date beside the definition. Record the calibration date with the definition so a later reviewer can follow it.
Applying the review
Take one month of tickets with pauses and extract the status history. Group the pauses by reason and compare the active, paused, and elapsed time. Read ten tickets with the longest pauses and look for missing follow-ups. The exercise should produce one rule to change, such as a cap or a follow-up interval.
Apply the change for a month and report the same three figures. Compare the ratio within the same request type and note any change in case mix. The aim is a target that describes the customer experience without hiding the work. A modest, documented rule is better than a precise number with no history.
Sources
- ISO 18295-1:2017 customer contact centres - Contact-centre performance measurement guidance.
- Atlassian service-level agreement guide - Service target and measurement concepts.
- NIST SP 800-53 Rev. 5 security and privacy controls - Audit and record-keeping control concepts.
- U.S. GAO Assessing Data Reliability - Data reliability assessment guidance.