Stop Asking Candidates to Describe the Job. Let Them Do It.
Every interview contains the same quiet flaw. You ask a candidate to tell you about a time they did something difficult, and they tell you a story. A good story. Possibly a true one. But you have no way to know whether the story reflects their ability or their storytelling.
There's a simple fix, and it's not a cleverer question. It's a different mode entirely: stop asking candidates to describe the work, and let them actually do a version of it. This is the core of work sample testing, job auditions, and hiring simulations—and it's one of the most reliable upgrades a hiring process can make.
The logic is almost too obvious to argue with. If you want to know whether someone can do a job, the most direct evidence is watching them do something very close to it. Everything else is an inference.
QUICK ANSWER
Work sample tests and job auditions ask candidates to perform tasks that closely mirror the actual role, producing direct evidence of ability instead of a verbal description of it. Research consistently ranks them among the strongest predictors of job performance—more predictive than interviews or resumes—because they replicate the work itself. They also give candidates a realistic preview of the role, which improves fit and retention. Hiring simulations (R2R Select) apply this principle systematically: build the task from the real role, then measure how candidates perform it.
Key insight : "Work sample tests are roughly 29% more predictive of job performance than interviews—and about 3x more effective than CVs at identifying the best candidate."— Work-sample research, as cited by Aakash G. and X0PA
Why "Describe It" Fails and "Do It" Works
When you ask a candidate to describe how they'd handle something, you're testing three things at once:
Whether they've done it.
Whether they can articulate it.
Whether they've rehearsed a good answer.
...and you can't tell which is which. A brilliant performer who interviews poorly looks the same as a weak performer who interviews brilliantly.
When you ask a candidate to actually do a representative task, that ambiguity collapses. You see their real approach: how they structure the problem, what they prioritize, where they get stuck, and how they recover. The output is evidence you can compare fairly across candidates, side by side, against the same standard.
"Job auditions as a standalone method can predict more than 30% of performance—versus roughly 14% for an unstructured interview."— Vervoe, on job audition validity
The Hidden Second Benefit: Candidates See the Job Too
There's a benefit to this approach that has nothing to do with assessment. When a candidate completes a realistic task, they get an honest preview of what the role actually involves. Someone who thought they wanted the job but discovers the day-to-day isn't for them can self-select out—before they're hired, before they become an early-attrition statistic.
This makes the process fairer and more transparent for candidates, and it protects the organization from the expensive kind of mismatch where someone accepts a role that was never going to fit. Evidence-based hiring improves both the decision and the mutual fit behind it.
Visual: Describe vs. Do
Method | What You Learn | Candidate Outcome |
Ask Them to Describe It | You learn how well they interview. | Candidate gets a rehearsed conversation. |
Ask Them to Do It | You learn how they'd actually perform. | Candidate learns whether the job actually fits. |
Doing It Well: From Ad-Hoc Task to Systematic Simulation
A work sample can be as simple as a marketing candidate drafting a campaign outline or a support candidate handling a role-played complaint. But ad-hoc tasks have a weakness: without a consistent design and a clear scoring standard, they can drift back toward subjective judgment—the very thing they were meant to replace.
This is where a systematic approach matters.
A hiring simulation built on the R2R Select method doesn't just hand candidates a task—it starts from the real role, identifies the specific things that role requires someone to do, builds a realistic scenario around those, and scores the resulting behavior against a defined rubric.
The result is evidence that's not only direct but consistent and comparable across every candidate. As covered in "Your Interview Predicts 9% of Job Performance," the goal is to replace proxies with evidence—and doing that reliably takes structure, not just a good exercise.
Frequently Asked Questions
What is a work sample test in hiring?
A work sample test asks a candidate to complete a task that closely mirrors the actual work of the role—for example, drafting a campaign, solving a representative problem, or handling a role-played customer situation. Instead of describing their ability, candidates demonstrate it, giving employers direct evidence of performance.
Are work sample tests better than interviews?
For predicting job performance, generally yes. Research consistently ranks work sample tests among the strongest predictors—more predictive than unstructured interviews or resumes—because they replicate the actual work rather than relying on proxies or self-description.
What is a job audition?
A job audition is a simulation or trial of the role a candidate has applied for—a work test, skills assessment, or case study where they perform realistic tasks. It lets employers assess candidates on real-world scenarios and is significantly more predictive of performance than an interview alone.
Do work samples improve retention?
They can. Because a realistic task gives candidates an honest preview of the role, people who wouldn't enjoy or suit the work can self-select out before being hired. This improves mutual fit and reduces the early attrition that comes from roles not matching expectations.
How is a hiring simulation different from a one-off work sample?
A one-off work sample is a single task, often judged subjectively. A hiring simulation built on a framework like R2R Select is designed systematically—derived from the real role, structured around its actual requirements, and scored against a defined rubric—so the evidence is consistent and comparable across all candidates, not just a good exercise.
The R2R Select explainer breaks down why interviews reward interviewing—and how hiring simulations show what candidates can actually do.
Curious how this applies to your roles?
R2R Select is how we design hiring simulations that produce real evidence of candidate readiness—mapped to the specific roles you're hiring for. We walk organizations through how it works.
Comments