Research question and scope
This protocol measures how often a new specialist clears the thirty-day quality gate without rework. The unit is one specialist assigned to one bounded work lane. Clearing without rework means every sampled item met the gate definition and none required a correction after review. The measure describes first-month readiness for the current scope, not long-term performance.
Set the scope to one client, one lane, and a fixed start cohort. Record the onboarding guide, the gate definitions, and the sample size in force. If the gate changed during the cohort, split the group at the effective date. A different gate creates a different denominator.
Keep the measure separate from retention or attendance. A specialist may clear the gate and later leave, or stay while missing the quality standard. The purpose is to test the first-month evidence. Report retention and attendance separately.
State the decision the result will support before collection begins. The client may use it to change onboarding, adjust the sample, or revise the gate dimensions. Write the intended action beside the scope. A measure with no decision becomes a monthly scorecard.
Population and unit of analysis
Define the population as every new specialist who reached day thirty in the lane during the period. Include specialists hired through any source and any shift. The unit is the specialist-lane pair. A specialist who changes lanes starts a new unit with a new gate.
Exclude specialists who left before day thirty or whose scope changed before the gate. Record the excluded count and the reason. A person who never reached the gate cannot clear it. Report the exclusion so the reader can judge the cohort.
Include specialists whose work was reviewed by any named reviewer. Record the reviewer role and the sample draw. Leaving out a reviewer hides part of the evidence. State the inclusion rule before scoring.
Decide how to treat a specialist who was coached during the month. Coaching is part of the process, not a disqualifier. Record the coaching date and the behavior. The gate should measure the result after reasonable support.
Definitions that make the measure reproducible
Define the gate dimensions from the written guide, such as accuracy, completeness, timeliness, and escalation judgment. Define pass, needs coaching, and fail for each dimension with an example. Publish the definitions before the cohort starts. The reviewer should not invent a standard at day thirty.
Define rework as any item returned for a correction after the first review, regardless of who made the correction. Record the reason and the time. A specialist who catches their own error before submission has not produced rework. The distinction keeps the measure about output.
Define a clear as a sample where every item meets the pass definition and no item required rework. Define a conditional clear as a sample with coaching notes but no failed dimension. Define a stop as a sample with at least one failed dimension that the coaching plan did not repair. Record all three definitions in the gate note before scoring begins.
Define the sample draw by queue, date range, and item type. Use a random or complete draw rather than items the specialist chose. Record the draw method and the count. A selected sample cannot show first-month readiness.
Sampling and data collection
Draw the sample from the specialist's completed work during the thirty days. Record the export date, the filter, and the row count. Include a mix of routine and exception cases if the lane contains both. A sample of only easy items overstates readiness.
Capture the specialist identifier, the lane, the start date, the review date, the sample count, and the result for each dimension. Add the rework count and the coaching notes. Store the evidence with the gate record. The sample and the written definitions are the core evidence.
Check a small overlap sample with a second reviewer. If the reviewers disagree about whether an item needed rework, the definition is too vague. Rewrite the definition and retest before scoring the cohort. Agreement matters more than sample size.
Record the time spent reviewing each item and the number of systems opened. A gate that takes a full day per specialist will not be repeated. Note whether the work tool exposes a stable review history. Collection cost belongs beside the result.
Measuring gate outcomes
Report the share of specialists who clear without rework, clear with coaching, and stop. Show the counts behind each share. The three groups describe different readiness levels and should not be collapsed. A single pass rate hides the person who needed repeated correction.
Break the result down by lane, shift, and reviewer. A lane with a weak guide may produce more coaching outcomes. Compare only groups with the same gate version and a similar sample size. A rate without the group size is hard to interpret.
Report the most common failed dimension and the most common rework reason. A missing source may point to onboarding, while a late handoff may point to scheduling. Record one example of each. The pattern should lead to one fix.
Report the number of coaching actions before the gate and whether the same issue repeated. A specialist who improved after one correction tells a different story from one who repeated the same error. Record the sequence. The trend matters more than a single score.
Limitations and next action
The measure cannot show how the specialist will perform after the scope expands. It also cannot show whether the gate dimensions match the client's real priorities. Treat the result as first-month readiness for a bounded lane. Note the lanes and work types that were not included.
Give the result to the client owner and the provider lead with one change to test. The change might be an earlier source check, a smaller first scope, or a clearer rework definition. Choose one change and name the owner who will apply it. Record the effective date and repeat the measure with the same definitions.
Review the boundary cases before tightening the gate. A specialist may need more time because the client changed the tools during the month. Fix the scope or the timing first if that is the pattern. The goal is a fair decision, not a lower pass rate.
Schedule the next measurement for the same point in the following cohort. Keep the lane and sample filters unchanged so the comparison is fair. Record any gate change in the note. A trend built from different standards is not a trend.
