On this page
- Five competencies cover most voice roles: communication, listening, customer handling, process following and availability.
- Role-play questions show more than "tell me about yourself".
- A 1-to-4 rubric with examples keeps scores consistent across recruiters.
- Score what was said. HireVue dropped facial analysis in 2021, and accent is not a skill.
What should you assess in a call centre candidate?
| Competency | How to test it | A strong answer |
|---|---|---|
| Spoken communication | Answer in the process language for 60 to 90 seconds | Clear, structured, easy to follow |
| Listening and comprehension | Play a short customer message, ask them to summarise it | Gets the facts and the request right |
| Customer handling | Role-play an upset customer | Acknowledges, apologises, explains next step and timeline |
| Process following | Describe a short policy, then a scenario that tests it | Applies the policy without promising extras |
| Availability | Ask about shifts, notice and location | Clear yes or no, with dates |
What questions should you ask?
- "A customer says they were charged twice and is angry. Respond as you would on the call."
- "Listen to this 30-second message. What is the customer asking for, and what will you do?"
- "Our policy allows refunds within 7 days. A customer asks on day 9. What do you say?"
- "Tell us about a time you had to stay calm with someone who was upset."
- "Which shifts can you work, including nights and weekends, and when can you join?"
- "Explain how to reset a password to someone who has never done it." (tests clarity)
How do you score answers fairly?
| Score | Customer handling (example rubric) |
|---|---|
| 4: Strong | Acknowledges feelings, apologises once, confirms details, gives a clear next step and timeline, stays within policy |
| 3: Acceptable | Polite and clear, but misses one element such as the timeline |
| 2: Weak | Defensive or vague; no clear next step |
| 1: Not ready | Argues with the customer or promises what policy doesn't allow |
Write a rubric like this for each competency, agree it with the process lead, and calibrate by having two recruiters score the same ten answers. Structured interviews like this are among the most valid selection methods in the research (.42 in Sackett and colleagues' 2022 meta-analysis), well ahead of unstructured interviews (.19).
What should AI not score?
Faces, eye contact, accents, "emotion" or "sentiment". These are not job skills, they raise fairness and legal risks, and HireVue, one of the best-known vendors, dropped facial analysis from its assessments in 2021. Candidates are also wary: only 26% trust AI to evaluate them fairly (Gartner, 2025). Score the content and clarity of the answer, flag integrity issues for a person, and keep the decision human.
Video or voice: which is better for call centre roles?
For voice processes, an AI voice screening call is close to the real job and works on any phone; Zena, Hirin.ai's voice agent, runs these in English, Hindi and 22+ Indian languages. One-way video suits chat, email and client-facing roles where written or on-camera communication matters. Many teams use a voice call first and video for the shortlist.
Assess your next batch
One-way video interviews scored against your rubric with quotes, and AI voice screening for voice processes.
Frequently asked questions
What makes a good call centre agent?
Clear communication in the process language, careful listening, calm handling of upset customers, following process, and reliability on shifts.
How long should a call centre AI interview be?
Four to six questions and 15 to 25 minutes is enough for a first round.
Can AI judge empathy?
It can score whether an answer contains the elements of an empathetic response (acknowledging, apologising, next steps) against a rubric. It shouldn't try to read emotion from faces or voice tone.
The competencies, questions and rubric were written for this guide from how Indian BPO teams assess voice and chat agents. Research figures were checked against their sources on 12 October 2026 and are not Hirin results.
- Sackett, P. R., Zhang, C., Berry, C. M. and Lievens, F. (2022), Revisiting meta-analytic estimates of validity in personnel selection, Journal of Applied Psychology.
- SHRM, HireVue discontinues facial analysis screening (3 Feb 2021).
- Gartner, Just 26% of job applicants trust AI will fairly evaluate them (31 Jul 2025).
- Langer, M., König, C. J. and Krause, K. (2017), Examining digital interviews for personnel selection, International Journal of Selection and Assessment.
Rajni Bansal
Rajni Bansal is a seasoned HR leader with 15+ years of experience driving people strategy across global tech and services organizations. She brings deep expertise in talent management, digital HR transformation, and AI adoption in recruitment. As a contributor to Hirin.ai, Rajni shares practical insights on how HR teams can leverage emerging technology to build agile, future-ready workplaces.