# Help-desk assessment measurement worksheet

Define the observed decision before an authorized assessment. This worksheet includes fictional arithmetic to demonstrate the method; no counts are GhostEye customer results or industry benchmarks.

## Assessment definition

- Primary decision and expected response:
- Approved procedure and scenario version:
- Scenario boundary and stopping point:
- Assessable eligible-encounter definition:
- Separate count: decision reached, outcome unknown:
- Treatment of repeated contacts and transferred calls:
- Observation window and clock start:
- Exclusion and uncertainty rules:
- Coverage groups:
- Legitimate-service comparison:
- Evidence custodian and authorized reviewers:
- Stop conditions and receiving operational owner:

Use stable encounter references. Keep names, real credentials, recovery secrets and unnecessary personal information out of the analysis export. Detailed evidence belongs in the approved restricted system.

## Freeze the denominator dictionary

| Metric | Numerator | Denominator or population | Missing-data rule to define |
| --- | --- | --- | --- |
| Connection rate | Connected interactions | Scheduled attempts | Record nonconnection separately |
| Assessable eligibility rate | Encounters that reached the decision and have assessable outcomes | Connected interactions | Separate pre-decision endings from unknown outcomes |
| Decision criterion | Assessable eligible encounters meeting the criterion | All assessable eligible encounters | Report decision-reached unknowns separately, outside this denominator |
| Handoff routing | Required handoffs routed | All required handoffs | Do not silently drop unrouted or unknown cases |
| Handoff acknowledgment | Required handoffs acknowledged within the window | All required handoffs | Preserve the count without acknowledgment |
| Conditional acknowledgment | Routed handoffs acknowledged within the window | Routed handoffs | Label this conditional denominator explicitly |
| Completed acknowledgment time | Distribution of observed elapsed times | Acknowledged handoffs only | Report incomplete outcomes alongside; never assign them zero time |

Define eligibility before seeing outcomes. “Eligible” in this worksheet means the encounter reached the predefined decision and has enough evidence to classify its outcome.

If it reached the decision but the outcome is unknown, count it separately as “decision reached, outcome unknown.” Publish that count beside the assessable eligible count; exclude it from the assessable-case rate. Do not relabel it as a pre-decision exclusion or a confirmed failure. A high unknown count limits the usefulness of the reported rate.

## Worked fictional baseline

| Condition | Scheduled | Connected | Eligible | Criterion met |
| --- | ---: | ---: | ---: | ---: |
| Daytime | 60 | 46 | 40 | 32 |
| Evening | 40 | 34 | 32 | 22 |
| Total | 100 | 80 | 72 | 54 |

- Connection: 80/100 = 80%.
- Eligibility among connected interactions: 72/80 = 90%.
- Decision criterion: 54/72 = 75%; 18 eligible encounters did not meet it.
- Daytime: 32/40 = 80%. Evening: 22/32 = 68.75%.
- Difference: 11.25 percentage points. This is a descriptive gap, not its cause.
- Pre-decision endings: six daytime and two evening. These eight connected interactions did not reach the predefined decision point. A real report must retain their specific reasons.
- Decision reached, outcome unknown: zero in this fictional dataset. All 72 eligible outcomes are assessable; the arithmetic is unchanged.

## Worked fictional handoffs

Of the same 72 eligible encounters, an overlapping subset of 24 required handoff. Eighteen were routed and 15 were acknowledged within 30 minutes of the escalation decision. Handoffs are not an additional stage every encounter must reach.

- Routing: 18/24 = 75%.
- Acknowledgment among required handoffs: 15/24 = 62.5%.
- Acknowledgment among routed handoffs: 15/18 = 83.3%, rounded.
- Six required handoffs were not routed; three routed handoffs lacked acknowledgment within the window.

Fictional completed acknowledgment times, in minutes: 2, 3, 3, 4, 4, 5, 5, 6, 7, 8, 9, 10, 12, 15, 22.

The median is 6 minutes among the 15 acknowledged handoffs. Report that beside the nine required handoffs without acknowledgment within the window. Acknowledgment does not establish incident resolution or ownership acceptance unless separately defined and observed.

## Matched retest record

- Same decision and scenario family:
- Same procedure/scoring rules, or changes documented:
- Coverage and routing differences:
- Participant relationship retained where authorized:
- Prior exposure or other comparison limitations:

Fictional paired example: the same 32 evening participants each produce one eligible encounter in each round.

| Paired outcome | Participants |
| --- | ---: |
| Met criterion in both rounds | 20 |
| Met initially, did not meet in retest | 2 |
| Did not meet initially, met in retest | 7 |
| Did not meet in either round | 3 |
| Total | 32 |

Initial: 22/32 = 68.75%. Retest: 27/32 = 84.375%. Descriptive increase: 15.625 percentage points, with seven improved and two worsened classifications. Do not call this an intervention's causal effect or a prediction of real-world incident rates.

## Decision-ready report

- Question assessed:
- Population and conditions actually observed:
- Result with numerator, denominator and window:
- Exclusions, unknowns and incomplete outcomes:
- Evidence limitations:
- Improvement owner and action:
- Retest condition and expected observation:

Have a second reviewer apply the definitions to a sample of records and all consequential disputed cases. Resolve definition changes before comparing results. Keep the dictionary with the report so another reviewer can reproduce the calculation.