Learning objectives
- Translate a business task into observable capability.
- Compare candidates consistently and plan support after selection.
Describe the output before the candidate
List the role’s important outputs, the decisions involved and the consequences of error. For a reporting specialist, an output might be a reconciled monthly dashboard with documented definitions, rather than “excellent data skills”. Specify which capabilities are essential on entry and which can be learned with supervision. Avoid requirements that have no credible connection to the actual work.
Use proportionate evidence
A structured interview asks comparable, job-related questions and uses defined evaluation criteria. A short work sample can show reasoning that a résumé cannot. Use a fictional or suitably anonymized task, explain the expected effort and provide reasonable accessibility arrangements. Do not request unpaid productive client work under the guise of assessment.
Separate evidence from confidence
Record what the person did, the criterion applied and the remaining uncertainty. Compare the same dimensions across candidates rather than rewarding similarity to the interviewer. A certificate may indicate study but does not alone demonstrate performance in your workflow. Where a gap is trainable, compare the cost and time of development with the need for immediate independent work.
Make competence usable after appointment
Provide access, context, a named reviewer and a clear definition of completion. Set early work at an appropriate risk level and review examples together. Maintain documentation and shared learning so the business does not become dependent on one expert’s memory. Review quality, turnaround and collaboration; high individual output can still create downstream rework.
From task to development
Keep each selection criterion traceable to a real work requirement.
Task
Identify an important output and its risk.
Capability
Describe the knowledge and judgement required.
Evidence
Choose a fair interview or work sample.
Support
Plan onboarding, feedback and review.
Comparison
| Requirement | Evidence | Support |
|---|---|---|
| Reconcile totals | Explain differences in a fictional dataset | Review the first reports with a mentor |
| Explain findings | Present one conclusion and limitation | Provide a stakeholder briefing template |
| Protect information | Identify unnecessary sensitive fields | Train on approved access and retention rules |
Choosing between immediate capacity and development
A hypothetical small business needs reliable monthly reporting. One candidate knows its software; another demonstrates stronger checking and explanation.
- The team scores both using the same fictional reconciliation task.
- It separates software familiarity from reasoning and quality control.
- It compares the training plan and review capacity before deciding.
The choice rests on evidence and support requirements. Neither a familiar tool nor an impressive presentation automatically determines the decision.
Practical checklist
- Link each requirement to an output.
- Define criteria before assessment.
- Use the same core questions fairly.
- Plan access, supervision and learning.
Common pitfalls
The impossible role
Combining several unrelated professions into one job can conceal an unrealistic resourcing plan.
Expertise without review
Specialists still need clear standards and a route to question assumptions.
Questions and answers
Must every specialist be hired permanently?
No. Consider the duration, continuity, confidentiality and supervision needs, then assess a suitable lawful resourcing arrangement.
Can a work sample replace every other assessment?
It shows only the capabilities represented by that sample. Combine proportionate evidence and explain what remains uncertain.
Sources and further reading
- Analyzing job requirementsU.S. Office of Personnel Management
Primary guidance on connecting tasks and required competencies.
- Structured interviewsU.S. Office of Personnel Management
Primary guidance for using consistent job-related questions and evaluation.
