Templates

Interview scorecard for call center agents, with anchors for each competency

On this page
  1. Which competencies should a call center agent scorecard cover?
  2. The scorecard: weights and anchors for 1, 3 and 5
  3. The copy-ready scorecard
  4. Anchors for 2 and 4, and how to test each competency
  5. Two work samples to use in the interview
  6. A filled example
  7. How should the scorecard change by seat and level?
  8. What mistakes are specific to scoring call center agents?
  9. Where does this scorecard fit in the hiring process?
  10. Questions people ask

An interview scorecard for call center agents should score six things a generic form misses: call control and resolution, systems accuracy while talking, policy and disclosure adherence, empathy and tone under pressure, coachability, and reliability. Score each on a 1–5 scale with written anchors, weight them before the first candidate, and keep two items out of the math as pass/fail gates: availability for the training class and verifying the caller's identity before discussing the account. Most of the scores should come from a role-played call, not from interview answers about past calls.

The five points mean what they mean in the 1–5 interview rating scale: 3 is a real pass, 5 is rare. The form layout follows the interview scorecard template. The loop that produces the evidence (a systems test, a scored call, a short structured interview) is in the interview guide for call center roles; this page is the scoring sheet that loop reports into, with an anchor for every point.

Hiring for a role that mixes phone, email and in-person service rather than a queue-based contact center? Our customer service representative scorecard weights the same skills differently.

Which competencies should a call center agent scorecard cover?

  • Call control and resolution. Does the candidate steer a frustrated caller toward what can be fixed and end with a clear next step?
  • Systems accuracy while talking. Can they find, update and note the right record without long dead air?
  • Policy and disclosure adherence. Do they say what the call flow requires and say no to an out-of-policy request without inventing an exception?
  • Empathy and tone under pressure. Is the acknowledgment specific to this caller, and does the tone hold when the caller interrupts?
  • Coachability. Can they see what went wrong in their own call and change it on a second try?
  • Reliability. Is there evidence they keep to a fixed schedule and handle conflicts through the proper channel?

Gates, not scored: the training dates and the shift (confirmed on the call center agent phone screen), identity verification per the call flow during the simulation, a private and connected workspace for remote seats, and any license the seat needs, such as a state insurance license for licensed sales seats (see how to verify an insurance license). A gate is pass or fail; it never adds points.

The scorecard: weights and anchors for 1, 3 and 5

Example weights for an inbound customer service or billing seat, the same split the interview guide uses. Lock yours at intake.

CompetencyWeight1 looks like3 looks like5 looks like
Call control and resolution25%Lets the caller run the call, talks over them or goes silent; the call ends with no next step.Acknowledges, asks two or three targeted questions, steers back to what can be fixed, closes with a next step and a timeframe.Does all of the 3 and resolves the hard turn (a refused request, a supervisor demand) with an allowed alternative, without needing to transfer.
Systems accuracy while talking20%Stops talking for long stretches to type, or enters the wrong record or field.Keeps the conversation going while navigating; required fields correct; a note another agent could act on.Correct entries and a complete note with no dead air, and catches the planted mismatch between the screen and what the caller says.
Policy and disclosure adherence15%Skips a required disclosure or invents a policy ("I'll waive that for you").Says the required disclosures, follows the policy sheet, and declines the out-of-policy request clearly.Follows policy, explains the reason in plain words, offers what the policy does allow, and notes the refusal on the account.
Empathy and tone under pressure15%Defensive or sarcastic, or repeats a stock line with nothing behind it.Acknowledges the caller's specific problem, keeps a steady tone through the interruption, blames no one.The role-player notes the caller calmed because of what the candidate said, and the tone stays warm through the refusal.
Coachability15%Disputes the feedback or cannot name anything they would change.Names one real improvement in the out-of-role debrief and tries it on the second attempt.Self-assessment matches the quality analyst's on the main points, and the second attempt shows the change clearly.
Reliability10%No example of keeping a fixed schedule, or describes missed shifts as other people's fault.A specific job with set start times, how they kept to it, and a conflict handled through notice or a swap.Two such examples, one where they solved a coverage gap for the team, and a reference confirms it.

If your team uses a 1–4 scale, keep the 1 and 3 anchors as they are and merge the 4 and 5 anchors below into one top point.

The copy-ready scorecard

Candidate: [name]                 Seat: [inbound / outbound / collections / tier 2]
Interviewer: [name]               Stage: [systems test / call simulation / interview]

GATES (pass / fail)
Training class and shift confirmed:            [ ]
Verified identity before account details:      [ ]
Remote workspace meets the stated minimum:     [ ] (remote seats only)
Required license (licensed seats only):        [ ]

Competency                          Weight  Score (1-5)  Evidence (what they said or did)
Call control and resolution          25%    [ ]          [ ]
Systems accuracy while talking       20%    [ ]          [ ]
Policy and disclosure adherence      15%    [ ]          [ ]
Empathy and tone under pressure      15%    [ ]          [ ]
Coachability                         15%    [ ]          [ ]
Reliability                          10%    [ ]          [ ]

Weighted total: sum of (weight x score) = [ ]
Not assessed (and why): [ ]
Recommendation: [Strong yes / Yes / No / Strong no]

Anchors for 2 and 4, and how to test each competency

The table gives the 1, 3 and 5. These are the in-between points, so no interviewer has to guess.

Call control and resolution

Tested by: the scored call. 2: reaches a fix, but only after the caller repeats the problem or the role-player steers. 4: regains control after an interruption without sounding scripted and reads the agreed fix back to the caller before closing.

Systems accuracy while talking

Tested by: the systems exercise and the scored call. 2: fields mostly correct, but the note says something like "customer upset, fixed" and nothing a second agent could use. 4: accurate entries, and when they need silence to search, they tell the caller what they are doing and for how long, as the call flow requires.

Policy and disclosure adherence

Tested by: the scored call, with a one-page policy sheet given five minutes before. 2: says the disclosures late or rushed and hesitates when the caller pushes on policy. 4: handles the out-of-policy demand and the supervisor request exactly as the sheet says, without blaming "the company."

Empathy and tone under pressure

Tested by: the scored call. 2: polite but flat; the acknowledgment could apply to any caller. 4: slows down and changes wording when the caller escalates, and names the caller's actual concern ("you were charged twice and you need that money this week").

Coachability

Tested by: the out-of-role debrief ("How did that go? What would you change?") and a 90-second second attempt after one piece of feedback. 2: accepts the feedback, but the second attempt sounds the same. 4: names the same top issue the quality analyst found before being told.

Reliability

Tested by: the structured interview ("Tell me about a job where you had to be in a set place at a set time every day") and references. 2: says they are always on time but gives no example. 4: a specific example plus how they handled a schedule conflict early, through the supervisor. Ask about their record and habits only; the lawful attendance wording, and what not to ask, is in the phone screen guide.

Two work samples to use in the interview

Scenario A: the double charge (8 minutes, role-played). An experienced agent plays a caller who says they were charged twice for a monthly plan, is calling from the car and wants the money back today. The candidate's mock screen shows one posted charge and one pending authorization. The example policy sheet says pending authorizations drop off on their own and refunds of posted charges are issued by a supervisor only. Halfway through, the caller demands a supervisor.

What to listen for: identity verified before any account detail (the gate); reading the screen correctly instead of agreeing there are two charges; explaining pending versus posted in plain words; a timeframe rather than a promise; handling the supervisor request the way the sheet says; and a note on the account a second agent could act on.

Scenario B: the address change while talking (6 minutes). The candidate updates a shipping address in a mock record while the role-player keeps asking about delivery; the ZIP code the caller gives does not match the city on file.

What to listen for: narrating each pause, confirming the mismatch instead of typing what they heard, and reading the address back. Score the record they leave behind, not typing speed.

Recording the simulation helps a second analyst check scores, but the candidate is a party to the call: get consent first and see two-party consent states. Never put a candidate on a live customer call.

A filled example

A made-up candidate for an inbound billing seat, scored after Scenario A and the structured interview:

Gates: training class confirmed; verified name and the last four digits on file before discussing the charges. Pass.

Call control and resolution — 4. When the caller interrupted, said "Let me make sure I fix the right thing" and asked which charge date they saw. Read back the outcome before closing. Did not offer an allowed alternative for the refund, so not a 5.

Systems accuracy while talking — 3. Found the pending authorization and explained it; note was complete. Twelve seconds of silence while searching with no explanation to the caller.

Policy and disclosure adherence — 3. Declined to promise a same-day refund and transferred the supervisor request per the sheet.

Empathy and tone under pressure — 4. "You need that money this week, so let me show you why only one of these will actually come out." Role-player noted the caller calmed.

Coachability — 3. In the debrief named the silent search as the thing to fix; second attempt narrated the search.

Reliability — 3. Two years in a warehouse role with a 6 a.m. start; described swapping a shift through the supervisor two weeks ahead for a known conflict.

Weighted total (example arithmetic): 0.25×4 + 0.20×3 + 0.15×3 + 0.15×4 + 0.15×3 + 0.10×3 = 3.40. Recommendation: Yes. Gates passed, no score below 3, and the one weak spot (silence during searches) is coachable in nesting.

A sample decision rule, to adapt: every gate passed, no score below 3 on call control or policy adherence, and a weighted total of 3.0 or higher. Set your own before the first class and apply it to every candidate.

How should the scorecard change by seat and level?

Seat or levelWhat to change
Entry level, no call center experienceKeep the anchors; take the evidence from the simulation and the second attempt rather than past metrics. Raise coachability to 20% and lower reliability to 5%, then confirm reliability with references.
Outbound salesReplace part of empathy's weight with objection handling, anchored the same way: a 3 answers the objection with a reason, a 5 turns it into a question that moves the call forward. Licensed seats keep the license gate.
Third-party collectionsWeight policy and disclosure adherence highest. The Fair Debt Collection Practices Act requires debt collectors to disclose that they are attempting to collect a debt in the first oral communication and to identify themselves as a debt collector after that (15 U.S.C. 1692e(11), checked November 2026); have compliance write the script the candidate must follow.
Bilingual seatsRun the same scenario in the second language with an assessor fluent in it and score the same anchors; see how to interview bilingual candidates.
Senior agent or team leadAdd a seventh line, coaching others: the candidate listens to a recorded practice call and gives the agent feedback, scored like the coachability anchors in reverse.

What mistakes are specific to scoring call center agents?

MistakeWhy it failsFix
Marking down an accentThe EEOC's national origin guidance (2016, on eeoc.gov as of November 2026) allows a decision based on accent only if it "interferes materially with job performance," which takes evidence, not an impressionScore whether the role-played caller got clear, correct information
A typing-speed cutoff nobody can justifyThe Uniform Guidelines "apply to tests and other selection procedures which are used as a basis for any employment decision" (29 CFR 1607.2(B)); see the four-fifths ruleScore accuracy in the mock system while talking, as above
Asking about childcare, health or how they get to work to judge reliabilityThose questions reach family status and disability, not the jobState the schedule and attendance rule and ask if they can meet it
No accommodation route for the systems testA candidate who uses a screen reader or other assistive tool is tested on the tool, not the job; the ADA counts changes to the application process as reasonable accommodation (29 CFR 1630.2(o)(1)(i))Say in the invitation how to ask for an accommodation, and score the same anchors
Analysts scoring without calibrationOne analyst's 3 is another's 4, and the class ranking becomes a scorer lotteryScore three practice calls together before the class, as in interview calibration

These points were checked against the linked sources as of November 2026 and are not legal advice.

Where does this scorecard fit in the hiring process?

The phone screen settles the gates, and the call center agent job description template maps each must-have to a row here. Score systems accuracy at the systems test, the call competencies at the simulation, and reliability in the interview and references; each scorer fills in only their rows before the debrief. After the hire, the 30-60-90 day plan for customer service representatives picks up the weak spots the scorecard found. If you run the structured interview in Interview Signal (we make it), you can enter these six competencies as scorecard areas on the kit screen, each with a weight of 1, 2 or 3; it scores each area from 1 to 10 against three rubric lines and keeps only quotes that appear in the candidate's side of the transcript word for word, so rewrite the 1, 3 and 5 anchors above as its rubric lines.

Questions people ask

What should an interview scorecard for a call center agent include?

Score six things on a 1 to 5 scale with written anchors: call control and resolution, systems accuracy while talking, policy and disclosure adherence, empathy and tone under pressure, coachability and reliability. Keep training-class availability and identity verification on the call as pass or fail gates, and score most of the competencies from a role-played call rather than from interview answers alone.

Can I score a call center candidate down for their accent?

Not for the accent itself. The EEOC's national origin guidance says an employment decision may be based on accent only if the accent interferes materially with job performance, which requires evidence that spoken English is needed for the job and that this person's accent actually stops callers from understanding them. Score whether the caller in the simulation got clear, correct information, not how the candidate sounds.

How do I score a call center candidate with no call center experience?

Use the same anchors but score the simulated call and the coachability debrief rather than past metrics. A first-time agent who verifies the caller, gives a clear next step and applies feedback on a second attempt can earn a 3 or 4 without ever having worked a queue, which is the evidence that predicts how they will do in training.

Should typing speed be a scored competency for call center agents?

Only if the job requires it and you can show the cutoff is related to the work, because a typing test is a selection procedure under the federal Uniform Guidelines. Most teams get better evidence by scoring accuracy in a mock system while the candidate keeps talking, which is what the systems competency on this scorecard measures.