Key takeaways
- Completion without evidence is not a reliable measure.
- Task timing matters as much as task occurrence.
- Small samples cannot support performance claims.
Define the task set
Manager onboarding tasks are often described in a policy document and tracked informally. The study defines a small task set with a checkpoint and an evidence type for each: before start, day one, week one, day thirty, and day ninety.
The task set must be realistic. A list of twenty tasks will produce low completion and a dataset that says more about the list than about the managers.
| Task | Checkpoint | Evidence | Limit |
|---|---|---|---|
| First one-to-one | Week one | Calendar entry and note | Self-reported |
| Buddy assignment | Before start | Recorded assignment | Assignment only |
| Thirty-day check-in | Day thirty | Completed checklist | Timing varies |
| Probation pack | Before review | Pack shared in advance | Manager judgment |
Code timing and evidence
For each task, record whether it occurred, the date, and the evidence. A task that occurred late is not the same as a task that did not occur. A task with no evidence is recorded as unverified.
Two reviewers should code a sample independently and compare. Disagreement usually means the evidence standard is unclear.
Analyze by cohort and context
Segment by manager, team, location, and start cohort. A task that is consistently late in one team may reflect workload or an unclear expectation rather than individual neglect.
Report counts with denominators. A completion rate based on five hires is a pattern to investigate, not a performance conclusion.
State the limits
The design is observational. It cannot show that a task causes retention or performance. It cannot capture the quality of a one-to-one or the appropriateness of a manager's judgment.
Use the results to improve task definitions, timing, and support. Do not use them as a standalone performance rating.
Sources and methodology
Observational review of manager task records across cohorts, with timing and evidence coded separately.
- Society for Human Resource Management current.
- U.S. Department of Labor current.
- EEOC Employer Guidance current.
- U.S. Bureau of Labor Statistics current.
Source count: 4. Last verification date: October 22, 2026.
Questions about the findings
What is measured?
Whether the task occurred, when, and what evidence supports it.
Does completion improve retention?
The design cannot establish that; it is descriptive.
How should late tasks be handled?
Record the date and the reason rather than treating late as missing.
Put the finding to work
Review the full research library, compare cluster coverage inside Manager Enablement, and pair these findings with our VA candidate screening support.
