.jpg){kind=link}
- Published
- Updated
A vendor may report high on-time delivery while the study team handles many missing outputs. The apparent contradiction can arise when the denominator includes only completed work. Reconcile the meaning of an eligible service opportunity and expected output before interpreting the percentage.
Scope of the decision
Cancelled, reworked, undelivered and out-of-scope tasks should not automatically be treated as one category. Assess exclusions in the study context. Comparisons between vendors also need comparable definitions; identical percentages can arise from very different volumes and distributions of risk.
Worked review
For example, one hundred files are expected during a month. Ninety arrive and eighty-nine are on time. The vendor calculates its metric using the ninety delivered files. The sponsor keeps the ten missing files visible and separates timeliness from completeness. One high percentage therefore does not describe the entire service.
Practical checks
- Show expected, received, cancelled and out-of-scope outputs for the same period.
- Explain numerator, denominator and the reasons for exclusions.
- Check whether important delays disappear within a low-volume subgroup.
Acceptance and limits
Accept a metric when both parties use consistent counting rules and unresolved service gaps remain visible. Limit historical comparisons when definitions change. An aggregate rate does not remove the need to assess an important individual effect on participants or critical data.
Frequently asked questions
- Should every output carry the same weight?
- Not always. Critical services may be tracked separately, with a clear rationale for weighting.
- Are exclusions inherently wrong?
- They can be appropriate if their definition, reason and effect on interpretation are transparent.