We test 8 voice agents on real calls, then set up the one that passes for your business

Why use someone neutral

• We don’t sell an agent

  Vendors recommend their own product. We set up whichever one passes on your script.

• Decisions from test calls

  The pick comes from recorded calls scored pass/fail, not a demo.

• Published method

  Our test battery and results are public, so you can check our work.

Proof: what round 4 found

We called 8 voice agents with the same receptionist prompt (a dental office), 3 calls per scenario. The same failure modes matter for any business that takes bookings by phone. See the full round 4 results.

• No agent read the callback number back correctly on all 3 tries. The best managed 2 of 3.
• Bland kept talking for 2.68 s after the caller cut in, over our 2.0 s limit.
• ElevenLabs was fastest, with a 1.06 s median reply.

Three calls per scenario is a small sample: treat these as patterns to check, not a final ranking.

What the agent handles, by industry

Examples of what we set up and test. Your script decides the details.

• Dental & healthcare

  • ✓ New-patient booking requests
  • ✓ Name and callback read-back
  • ✓ Urgent cases sent to immediate care
  • ✓ After-hours messages
  See the dental page →

• HVAC

  • ✓ Service-call lead capture with read-back
  • ✓ Booking requests for visits
  • ✓ No-heat or no-cooling calls transferred
  • ✓ After-hours answering

• Roofing

  • ✓ Estimate requests with address read-back
  • ✓ Inspection booking requests
  • ✓ Active leaks transferred
  • ✓ After-hours lead capture

• Contractors

  • ✓ Job leads with callback read-back
  • ✓ Site-visit booking requests
  • ✓ Urgent job-site calls transferred
  • ✓ After-hours messages

• Home services

  • ✓ Lead capture with read-back
  • ✓ Booking requests by service area
  • ✓ Emergencies transferred to on-call
  • ✓ After-hours answering

• Gyms

  • ✓ Membership and class enquiries
  • ✓ Tour and class booking requests
  • ✓ Urgent calls transferred to staff
  • ✓ After-hours questions and messages

• Salons

  • ✓ Appointment booking requests
  • ✓ Name and callback read-back
  • ✓ Calls for a stylist transferred
  • ✓ After-hours booking requests

• Law offices

  • ✓ New-client intake with read-back
  • ✓ Consultation booking requests
  • ✓ Urgent matters transferred to staff
  • ✓ After-hours intake

How it works

1. 1

   Audit

   We look at how your phones are answered today: missed calls, after-hours, common requests, and what your booking or CRM software can connect to.

2. 2

   Shortlist

   We pick the 2–3 agents from our benchmarks that fit your call mix, budget and integrations.

3. 3

   Private test on your script

   We run the shortlisted agents through our call battery using your hours, services, service area, pricing rules and emergency policy, with fictional caller data.

4. 4

   Setup

   We configure the winner on your own platform account: prompts, knowledge, call transfer, and booking integration where supported.

5. 5

   Go-live QA

   Before your number points at it, we re-run the tests on the final setup and only sign off when it passes.

Monthly QA

Agents change when platforms update models. Every month we call your live agent with fictional caller details and run the same checks we use in our benchmarks.

• ✓ Booking + read-back

  Books only real slots and reads back the caller’s name and callback number exactly. One wrong digit is a fail.

• ✓ Reschedule

  Offers only real availability; never invents a time.

• ✓ Unknown request + callback

  Doesn’t promise a service or plan you don’t offer; takes a callback or message correctly.

• ✓ Interruptions

  Stops talking within 2.0 s when the caller cuts in, then answers the new question.

• ✓ Emergency routing

  Follows your emergency policy, such as a live transfer or urgent dispatch, instead of booking a routine slot.

• ✓ Reply speed

  Median reply gap and p90, measured over the phone line.

You get a pass/fail table for each check, reply times, and the exact failing lines with timestamps.

FAQ