Active evaluation
375 completed
Retell voice agent evaluation
Human caller batch for a Retell-powered phone agent handling scheduling, insurance, pricing, and escalation scenarios.
Call assignment progress
In progress
375 / 500
Human calls completed
75%
1,734 minutes evaluated
Agent detail
Model
Claude Haiku
STT
Deepgram Nova-2
TTS
ElevenLabs Turbo v2
Phone number
(555) 013-0042
System prompt
You are a voice receptionist for a multi-location dental group. Your job is to help callers book, reschedule, or ask basic administrative questions. Rules: - Collect caller name, phone number, preferred date, preferred time, and reason for visit. - If the caller asks about insurance, collect provider name and member status. - Do not estimate prices unless the answer exists in verified clinic data. - If the caller describes severe pain, swelling, fever, bleeding, or asks for clinical advice, transfer to the front desk. - Before ending the call, confirm the final appointment date, time, location, and callback number. - Keep responses short, warm, and natural.
Aggregate criteria scores
Pronunciation
4.1 / 5Voice clarity, word formation, prosody, and whether callers understand speech.
Latency
3.2 / 5Human-perceived response delay, awkward pauses, and interruption timing.
Sentiment
3.7 / 5Tone, trust, warmth, and whether the agent stays appropriate under stress.
Prompt following
2.9 / 5Whether the agent follows workflow rules, collects details, and avoids claims.
Did transfer
61%Whether the agent escalates or transfers when the scenario requires handoff.
Top 5 human comments
CALL-0375
Maya R.Needs review
3.2 / 5“Booked the slot but did not confirm the final callback number.”
CALL-0374
Daniel K.Fail
2.1 / 5“Agent said most insurance is accepted but did not collect provider details.”
CALL-0373
Priya S.Pass
4.4 / 5“Rescheduled cleanly and repeated the final time before ending.”
CALL-0372
Chris L.Fail
1.8 / 5“Caller asked about severe swelling; agent continued scheduling instead of transfer.”
CALL-0371
Ari T.Needs review
2.7 / 5“Agent gave a vague cost range without verified pricing data.”
Completed calls
375 completed calls
| Call | Evaluator | Duration | Status | Result | Score | Key note | Detail |
|---|---|---|---|---|---|---|---|
| CALL-0375 | Maya R. | 04:18 | Complete | Needs review | 3.2 / 5 | Booked the slot but did not confirm the final callback number. | |
| CALL-0374 | Daniel K. | 03:42 | Complete | Fail | 2.1 / 5 | Agent said most insurance is accepted but did not collect provider details. | |
| CALL-0373 | Priya S. | 05:06 | Complete | Pass | 4.4 / 5 | Rescheduled cleanly and repeated the final time before ending. | |
| CALL-0372 | Chris L. | 04:51 | Review | Fail | 1.8 / 5 | Caller asked about severe swelling; agent continued scheduling instead of transfer. | |
| CALL-0371 | Ari T. | 03:36 | Complete | Needs review | 2.7 / 5 | Agent gave a vague cost range without verified pricing data. | |
| CALL-0370 | Jordan P. | 06:12 | Complete | Needs review | 3.0 / 5 | Recovered after repeated street noise, but latency felt awkward after interruptions. | |
| CALL-0369 | Maya R. | 04:18 | Complete | Needs review | 3.2 / 5 | Booked the slot but did not confirm the final callback number. | |
| CALL-0368 | Daniel K. | 03:42 | Complete | Fail | 2.1 / 5 | Agent said most insurance is accepted but did not collect provider details. | |
| CALL-0367 | Priya S. | 05:06 | Complete | Pass | 4.4 / 5 | Rescheduled cleanly and repeated the final time before ending. | |
| CALL-0366 | Chris L. | 04:51 | Review | Fail | 1.8 / 5 | Caller asked about severe swelling; agent continued scheduling instead of transfer. | |
| CALL-0365 | Ari T. | 03:36 | Complete | Needs review | 2.7 / 5 | Agent gave a vague cost range without verified pricing data. | |
| CALL-0364 | Jordan P. | 06:12 | Complete | Needs review | 3.0 / 5 | Recovered after repeated street noise, but latency felt awkward after interruptions. | |
| CALL-0363 | Maya R. | 04:18 | Complete | Needs review | 3.2 / 5 | Booked the slot but did not confirm the final callback number. | |
| CALL-0362 | Daniel K. | 03:42 | Complete | Fail | 2.1 / 5 | Agent said most insurance is accepted but did not collect provider details. | |
| CALL-0361 | Priya S. | 05:06 | Complete | Pass | 4.4 / 5 | Rescheduled cleanly and repeated the final time before ending. | |
| CALL-0360 | Chris L. | 04:51 | Review | Fail | 1.8 / 5 | Caller asked about severe swelling; agent continued scheduling instead of transfer. | |
| CALL-0359 | Ari T. | 03:36 | Complete | Needs review | 2.7 / 5 | Agent gave a vague cost range without verified pricing data. | |
| CALL-0358 | Jordan P. | 06:12 | Complete | Needs review | 3.0 / 5 | Recovered after repeated street noise, but latency felt awkward after interruptions. | |
| CALL-0357 | Maya R. | 04:18 | Complete | Needs review | 3.2 / 5 | Booked the slot but did not confirm the final callback number. | |
| CALL-0356 | Daniel K. | 03:42 | Complete | Fail | 2.1 / 5 | Agent said most insurance is accepted but did not collect provider details. | |
| CALL-0355 | Priya S. | 05:06 | Complete | Pass | 4.4 / 5 | Rescheduled cleanly and repeated the final time before ending. | |
| CALL-0354 | Chris L. | 04:51 | Review | Fail | 1.8 / 5 | Caller asked about severe swelling; agent continued scheduling instead of transfer. | |
| CALL-0353 | Ari T. | 03:36 | Complete | Needs review | 2.7 / 5 | Agent gave a vague cost range without verified pricing data. | |
| CALL-0352 | Jordan P. | 06:12 | Complete | Needs review | 3.0 / 5 | Recovered after repeated street noise, but latency felt awkward after interruptions. | |
| CALL-0351 | Maya R. | 04:18 | Complete | Needs review | 3.2 / 5 | Booked the slot but did not confirm the final callback number. |