How much context should entry-level candidates receive in a work sample?

Entry-level applicants should receive the company-specific context needed to understand the responsibility: the objective, starting information, relevant terms, available resources, constraints, authority, and expected output. Do not supply the judgment you want to observe. Do not withhold procedures or knowledge the company normally teaches after hiring and then score applicants for not knowing them.

The useful question is not “How little can we tell them?” It is “Which part of the difficulty belongs to the work?”

Separate four kinds of knowledge

Before writing instructions, sort every expected piece of knowledge into one of four buckets.

Knowledge type	What to do in the exercise	Example
Required on entry	It may be reasonable to assess it if it is genuinely necessary for the level and role.	Writing a clear update for a communications role.
Company-specific	Provide it or make it available as a resource.	The company’s refund authority or escalation route.
Learned after joining	Do not quietly turn it into a selection requirement. Either provide it, simplify it, or choose different work.	An internal code, legacy workflow, or undocumented exception.
Irrelevant to the responsibility	Exclude it.	Internal history that does not change the applicant’s decision.

OPM’s guidance makes a similar distinction: competencies required on the first day may be appropriate for selection, while competencies intended to be learned through experience or training should not become selection requirements (OPM, Job Analysis FAQ).

The buckets must be defined for the actual role. An entry-level laboratory role, hotel operations role, and software role will not have the same entry requirements.

Give enough context to act, not enough to prescribe the decision

A useful exercise supplies the environment around the decision while leaving the decision open.

Applicants usually need:

• A clear responsibility: what they own in this piece of work.
• A starting state: what has happened and what is currently known.
• Relevant records: the information a person in the role could reasonably access.
• Terms and definitions: company language that would otherwise be insider knowledge.
• Constraints: time, policy, quality, safety, budget, or authority limits that genuinely apply.
• Available resources: instructions, reference material, people or channels they may consult.
• Decision rights: what they can do, what needs approval, and what must be escalated.
• Expected output: what should be written, decided, produced, or handed over.
• Rules for uncertainty: whether to ask a question, state an assumption, pause, or continue.
• Practical conditions: expected effort, deadline, submission route, and support contact.

Do not provide a script that dictates every action if the point is to explore judgment. Do not omit the refund limit if acting within authority is part of the responsibility.

An illustrative example: the hidden escalation rule

The following example is fabricated and does not report customer research.

Weak version

An entry-level guest-services applicant receives a complaint from a guest whose room is unavailable. The prompt says, “Resolve the situation.” The reviewer expects the applicant to know that only a duty manager can approve relocation above a certain nightly rate and that one partner hotel should never be used after midnight.

The applicant may fail because the company withheld two internal rules. The exercise cannot cleanly distinguish weak judgment from missing employer knowledge.

Better-context version

The applicant receives:

• The guest’s booking and stated needs.
• A short relocation policy with the applicant’s approval limit.
• A list of available partner properties and current rates.
• The duty manager’s contact route.
• A note that transport time and accessibility needs must be considered.
• A clear output: communicate with the guest, record the decision, and leave any pending obligation visible.

The exercise still leaves meaningful questions open. Which options will the applicant investigate? What will they say before every fact is known? Will they escalate, and when? How will they balance speed, suitability, authority, and the guest’s needs?

The added context does not give away the answer. It makes the source of difficulty more relevant.

Missing information can be deliberate

Real work often contains incomplete information. Removing every uncertainty can make an exercise less representative. But an applicant needs a legitimate way to respond to what is missing.

Deliberate uncertainty can reveal whether someone:

• Notices a gap.
• Checks an available source.
• States an assumption.
• Requests clarification.
• Escalates a decision outside their authority.
• Pauses when proceeding would create unacceptable risk.
• Leaves a clear handover for the next person.

Accidental uncertainty is different. A broken link, undefined acronym, unavailable record, or hidden reviewer assumption adds noise without creating useful evidence.

The entry-level context checklist

Use this before an applicant receives the exercise:

• We have named the capabilities genuinely expected on entry.
• We have separated them from knowledge taught after hiring.
• The responsibility and desired output are clear.
• Company-specific terms and procedures needed for the work are available.
• The applicant’s authority and escalation routes are explicit.
• Every referenced record or resource is accessible.
• Missing information is intentional, and the applicant has a reasonable way to handle it.
• The time and effort expectation matches what was communicated.
• The instructions have been checked by someone who did not help write them.
• Applicants receive comparable core conditions.
• Reviewers know which evidence is observable and which conclusions require discussion.
• The invitation explains how to request an adjustment or accommodation.

Why instructions do not automatically make an exercise fair

Providing context solves one design problem. It does not resolve every question about a selection procedure.

An exercise can contain clear instructions and still:

• Measure something unrelated to the job.
• Demand more unpaid time than is reasonable.
• Use conditions that disadvantage some applicants unnecessarily.
• Be administered differently across candidates.
• Depend on an inaccessible format.
• Be reviewed through vague or shifting standards.
• Produce conclusions stronger than the evidence supports.

Research summarized by SIOP links applicant perceptions of fairness to factors including job relatedness, consistent procedures, explanations, opportunity to perform, and two-way communication (SIOP, Applicant Reactions to Selection). That is useful design guidance, not proof that any specific exercise is fair.

Legal duties vary. For example, UK government guidance describes reasonable adjustments during recruitment, while U.S. EEOC guidance explains reasonable accommodations under the ADA (GOV.UK; EEOC). Employers should apply the rules and professional standards relevant to their jurisdiction.

Test the instructions without testing an applicant

Ask a colleague who did not design the exercise to read the material and answer:

1. What responsibility do I own? 2. What information can I use? 3. What am I allowed to decide? 4. What should I do if information is missing? 5. What must I produce or leave behind? 6. Which parts test general capability, and which depend on company knowledge?

If two informed readers answer differently, revise the environment before using it in hiring.

Where Calavion fits

In Calavion, a company submits selected work and Calavion prepares the environment with the company for its approval. That preparation identifies missing context, asks targeted questions, and places approved context and resources inside the version applicants work through.

That process can make hidden assumptions easier to find. It does not automatically establish fairness, accessibility, validity, or equal treatment. Those remain design and employer responsibilities that require appropriate review.