Beyond the Resume: How to Design Performance-Based Assessments
If you followed our blueprint on skill-based hiring, you know that the “Work Sample Test” is the engine that makes the whole system run. But how do you actually build one?
Most companies fail here because they make their tests too theoretical or too long. To truly evaluate talent, your assessments must be “Job-Mirrors.” These are tasks that look exactly like a Tuesday morning at your company.
What is a Performance-Based Assessment? (Quick Definition)
A performance-based assessment is a practical evaluation where a candidate completes a specific task or project that mimics the actual work they would perform in the role. Unlike traditional interviews that ask, “Tell me about a time you did X,” these assessments require the candidate to actually “Do X” in a controlled environment.
The Three Rules of a Great Assessment
To ensure your test is objective and helpful, it must follow these three rules:
- Relevance: The task must be something the person will do at least once a week in the role.
- Time-Boxing: It should take no more than 60 to 90 minutes. Respect the candidate’s time.
- Objective Scoring: You must decide what a “Pass” looks like before you see the results.
Step-by-Step: Designing Your “Job-Mirror”
1. Identify the “Critical Incident”
Look at the role you are hiring for. What is the one task that, if done poorly, causes the most trouble?
- For a Customer Support Lead, it might be handling a high-tier escalation.
- For a Data Analyst, it might be cleaning a “dirty” dataset to find a specific trend.
- For a Sales Representative, it might be writing a cold outreach email based on a specific buyer persona.
2. Create the “Simulation Environment”
Provide the candidate with the same tools and information they would have on the job.
- The Brief: Give them a 1-page document explaining the goal.
- The Constraints: “You have 60 minutes and cannot use external AI tools for this specific part.”
- The Deliverable: Be clear about what they need to turn in. This could be a short video, a spreadsheet, or a written response.
3. Build the Scoring Rubric (The Anti-Bias Shield)
This is where the Bot Shreyasi approach differs from traditional hiring. You don’t “feel” like they did a good job. You measure it. Create a simple table:
| Criteria | 1 Point (Poor) | 3 Points (Average) | 5 Points (Excellent) |
|---|---|---|---|
| Accuracy | Multiple errors in data | 1-2 minor errors | 100% accurate |
| Clarity | Hard to follow logic | Logic is clear but wordy | Concise and actionable |
| Tone | Unprofessional | Professional but cold | Empathic and helpful |
Why “Adjacent Skills” Matter
Sometimes a candidate might not have the exact technical skill yet, but they show high “Learning Agility.”
During the assessment, give them a small piece of new information halfway through. For example, tell them a policy just changed and ask how that affects their answer. Their ability to pivot and incorporate new data is a better predictor of success than their initial knowledge.
Common Questions about Assessments
Should we pay candidates for these tests? If the assessment takes longer than two hours, yes. For shorter “Job-Mirrors” used in the early stages, it isn’t usually required. However, it is a great way to build your brand as a “talent-first” company.
How do we prevent cheating? Don’t test for things they can just Google. Test for judgment. Ask them to explain why they chose a specific solution. Cheating is hard when the right answer depends on the unique context of your business.
Can we use AI to grade these? You can use AI to help sort and highlight key points in their answers, but the final judgment should always be human. Skill-based hiring is about human potential, not just hitting checkboxes.
This guide is a deep-dive supplement to our primary guide on how to implement skill-based hiring.
