How to Evaluate Candidates When Their Best Work Is Confidential

Employers can evaluate candidates whose strongest work cannot be shared publicly by defining job-relevant criteria, accepting several forms of substitute evidence, asking structured questions, and scoring each candidate against the same standards. A missing public portfolio is not evidence of weak performance. It simply means the employer needs another way to examine the candidate’s decisions, contribution, execution, and outcomes without requesting protected information.

The goal is not to reconstruct a confidential project. It is to collect enough relevant evidence to make a responsible hiring decision while respecting the candidate’s existing obligations. Employers should explicitly tell candidates not to disclose confidential, proprietary, client-identifying, regulated, or trade-secret information. Questions about a specific agreement or disclosure risk should be referred to qualified counsel or the appropriate internal team.

Define What the Role Requires Before Asking for Proof

Start with the work the person will perform—not the portfolio format you expect them to provide. Translate the role into a short set of observable capabilities, decisions, behaviors, and outcomes.

For example, a product design role might require the ability to:

These criteria provide the basis for the assessment. A polished public case study is only one possible source of evidence for them.

Separating skill from artifact availability matters because candidates work under different disclosure conditions. One candidate may have public projects because a former employer encouraged publication. Another may have done equally relevant work for a private company, regulated organization, or client that prohibited sharing. Portfolio access and job capability are different questions.

Before an interview, decide what evidence would be sufficient for each criterion. Also define what a weak, acceptable, and strong response would look like. Explicit criteria help interviewers focus on the role rather than interpreting confidentiality as evasiveness or filling information gaps with assumptions.

Build an Evidence Menu That Does Not Require Protected Work

Give candidates more than one acceptable way to demonstrate relevant ability. Standardize what you assess while allowing the evidence format to vary.

Possible options include:

Redaction does not automatically make a document safe to share. A file may still reveal protected strategy, customer information, internal methods, metadata, or a recognizable project. Employers should not ask candidates to upload source code, internal dashboards, private communications, customer data, unreleased designs, strategy documents, or other materials from a former employer.

Explain the boundaries before requesting evidence. A useful instruction is: “Please do not share names, files, screenshots, nonpublic metrics, customer information, proprietary methods, or anything restricted by an agreement. We are interested in how you approached the work, not in obtaining the underlying material.”

MeeBoss supports direct conversations between employers and job seekers. Chat to Apply can provide an early channel for discussing which evidence formats are acceptable before a formal assessment begins. It does not verify, protect, or evaluate confidential work; the hiring team and candidate still need to establish appropriate boundaries.

Use Structured Questions to Understand the Candidate’s Contribution

A structured walkthrough can explore the candidate’s reasoning without asking them to reveal the protected project itself. Give every candidate the same core questions, then use limited follow-ups to clarify relevant evidence.

Begin by reminding the candidate to stay at a level they are permitted to discuss. Useful questions include:

Listen for concrete reasoning, not confidential detail. A candidate may be able to describe a decision sequence, identify the categories of evidence used, explain a tradeoff, and distinguish personal work from team output without naming the organization or disclosing internal results.

Interviewers should avoid treating greater detail as automatically more credible. A candidate who respects confidentiality may provide less contextual color than someone with publicly shareable work. Rate the response according to the preselected criteria, and record any unresolved uncertainty rather than pressuring the candidate to cross a disclosure boundary.

Compare the Tradeoffs of Walkthroughs, Exercises, and References

No substitute evidence method answers every question. Select a proportionate combination based on the role, the decision being tested, and the burden placed on the candidate.

Evidence methodWhat it may help exploreImportant limitationsAppropriate use
Sanitized walkthroughProblem framing, decisions, contribution, and communicationRelies partly on self-report and may omit essential contextEarly or mid-stage structured interview
Permitted redacted artifactExecution quality and how ideas were organizedRedaction can remove context and does not itself establish permission to shareWhen the candidate independently confirms disclosure is allowed
Fictional take-home exerciseApproach to a defined role-relevant taskCreates candidate burden and may reward extra available timeShort, clearly scoped assessments with transparent criteria
Live simulationReasoning, questions, and communication in the momentTime pressure may not resemble normal working conditionsRoles where interactive decision-making is genuinely relevant
Reference conversationResponsibilities, collaboration, and observed behaviorThe reference may have limited visibility or disclosure constraintsCorroborating selected claims rather than authenticating an entire project

A layered approach is often more useful than relying on one method. For example, a walkthrough may identify how the candidate thinks, a fictional exercise may show how they approach a new problem, and a reference may provide limited corroboration of their responsibilities. Each source contributes partial evidence; none should be treated as conclusive proof.

Keep exercises proportionate. Tell candidates why the task is being used, what will be assessed, how much time it should require, and whether any part resembles current business work. Use fictional or employer-provided information rather than asking candidates to reproduce a former employer’s process or create an undisclosed real deliverable.

Apply One Scorecard Across Different Forms of Evidence

Consistency should come from the criteria, not from forcing every candidate to provide an identical artifact. A candidate using a verbal walkthrough and a candidate presenting a public case study can still be evaluated against the same job requirements.

A simple scorecard might cover:

Add behavioral anchors to each rating. For instance, a strong problem-framing response might define the objective, identify competing constraints, and explain how the problem changed as new information appeared. A weak response might remain vague about the objective or the candidate’s role.

Ask interviewers to record ratings and supporting observations before discussing the candidate as a group. Where evidence is incomplete, mark the gap explicitly. A numerical score does not eliminate judgment or uncertainty, but documented reasoning makes it easier to see whether the decision rests on job-relevant evidence or portfolio access.

Example: Evaluating a Candidate Without Requesting the Original Work

Consider a hypothetical employer hiring a senior product designer. The candidate says their most relevant project involved an unreleased product and cannot be shown publicly.

  1. Define the criteria. The hiring manager selects problem framing, research interpretation, design tradeoffs, cross-functional collaboration, and validation as the core areas to assess.
  2. Confirm boundaries. The recruiter tells the candidate not to share the product name, screenshots, customer data, internal metrics, strategy, or files. The candidate says they can discuss the process at a high level and show a permitted, sanitized diagram they created for interviewing.
  3. Conduct a structured walkthrough. The candidate explains the general user problem, their responsibilities, the types of constraints involved, and the alternatives considered. When a follow-up approaches protected details, the candidate declines and describes the decision category instead.
  4. Use a fictional exercise. The employer provides a made-up product brief and asks the candidate to outline discovery questions, identify tradeoffs, and sketch a validation plan. The task has a defined time expectation and is scored against the same criteria established at the start.
  5. Check a reference. With the candidate’s permission, a former manager confirms the candidate’s general level of ownership, collaboration responsibilities, and decision-making role without discussing the confidential project itself.
  6. Document uncertainty. The team records that it did not inspect the original deliverable or independently verify the project’s nonpublic outcomes. It then considers whether the combined evidence is sufficient for each criterion.

The employer has not recreated or authenticated the confidential project. Instead, it has assembled several relevant evidence sources while respecting the candidate’s boundaries.

Put the Evaluation Process Into Practice

Turn the approach into a repeatable hiring workflow:

  1. Define the capabilities and outcomes required for the role.
  2. Create behavioral anchors before reviewing candidates.
  3. Offer several proportionate evidence methods.
  4. Tell candidates what must not be disclosed and what the assessment will involve.
  5. Brief interviewers on the same questions, criteria, and confidentiality boundaries.
  6. Record ratings before group discussion.
  7. Compare evidence across methods and document unresolved gaps.
  8. Escalate questions involving confidentiality agreements, trade secrets, regulated data, or employment law to qualified counsel or the appropriate internal team.

Review the process as well as individual hiring decisions. Useful operational measures can include assessment completion rate, candidate withdrawal rate, estimated exercise time, differences among interviewer ratings, and the frequency of unresolved evidence gaps. These indicators can identify excessive burden or inconsistent execution, although they do not by themselves prove fairness or predict job performance.

Candidate communication is also a practical checkpoint. Explain why evidence is requested, what format options are available, how the information will be assessed, and how much time the process should take. A respectful refusal to disclose protected information should not be treated as a negative signal.

MeeBoss is designed to make hiring more conversational through profiles, recommendations, and direct employer-candidate communication. When portfolio constraints need clarification, Chat to Apply can help both sides discuss acceptable evidence formats before investing time in a formal exercise. AI can support recommendations and workflow, while interview design, evidence interpretation, and hiring decisions remain human responsibilities.

Join MeeBoss and start a real hiring conversation.