A patient who wants an implant calls several clinics, often in the same hour, from the same sofa, and the shortlist gets made out of whichever calls went well. That is the part most clinics never see: the comparison is happening on the phone, and the phone is usually the least managed thing they own. So here is how to test your own line the way that patient would, and what an AI voice agent on the same number does and does not change.
How to read this
This is a method. The scorecard is our own: eight criteria, each scored zero to two, applied to a test call you place to your own line. We quote no score or timing from any test call, because the only line worth scoring is yours. The few published figures below come from named surveys, each attributed in the sentence that uses it.
The price question: what does a turnkey implant cost
Start with the question an implant patient needs answered before anything else: what does the whole thing cost, start to finish, with nothing left to add later.
Before it can even be asked, somebody has to pick up. On a busy line the caller can wait for a person, then wait again while that person goes to find someone who can talk about prices. An AI voice agent on the same number picks up on the first ring and has the price list in front of it from the start.
Then listen to the answer. A front desk without the full price list in front of it can only quote the part it remembers, and that part may be just the first line of the total. Ask a second time and the same partial figure can come back, because asking again puts the list in front of nobody.
A partial figure is worse than a high one. A patient who hears only the first line puts the clinic at the top of the shortlist, arrives, finds out the abutment and the crown are separate lines, and leaves feeling handled. The clinic wins the call and loses the patient, which is the most expensive way there is to lose one.
The agent answers the same question from the price list itself. If the clinic's price list sets out what a turnkey implant is made of, the agent can name each part, say plainly that the final figure depends on an examination, and offer to book the examination.
What the caller is really doing
They are not asking for a number. They are asking for a number they can compare. An incomplete answer is not a cheaper answer, it is an unusable one, and the caller finds that out at the next clinic on the list rather than at yours.
The access question: can a caller even reach a coordinator
A specialist line, such as a fertility centre's, can fail one step earlier: the caller has trouble reaching anyone who can answer.
The first thing the caller measures is how long it takes before a human voice says anything. Queues, hold music and transfers each add to that wait, and some calls end before anyone who can answer the question has picked up.
When a call does land, every transfer carries a second cost: the caller explains the same situation again to the next person. An AI voice agent takes the intake in one unbroken conversation and hands it on in writing, so nobody has to ask the same thing twice.
A line that cannot answer on the spot may offer a callback. The trouble shows up on the next call: if nothing from the first one was written down, the second starts from nothing, and the caller has to begin again from the top.
The cheapest thing a clinic can do on the phone is remember the last call.
The scorecard
The scorecard covers eight things a caller notices whether or not they could name them:
- How fast anything at all happened.
- Whether the person on the line established who they were talking to.
- Whether anyone asked what the caller actually needed.
- Whether the answer given was complete.
- Whether the price was stated in a form the caller could compare.
- Whether the obvious objection was handled or deflected.
- Whether the call ended with a next step.
- Whether the caller's contact details were captured.
Each one scores zero, one or two, so every call finishes somewhere out of sixteen. Most of the eight are about how the call is run.
The uncomfortable part
A coordinator who answers every clinical question well can still score low, because speed, the next step and the caller's details are scored too.
Three failures to listen for
All three come from how the line is set up:
- Speed. The clock runs until the caller hears a human being. ContactBabel put the average across UK contact centres at 116 seconds in 2024, and a Velaro survey back in 2012 found around 60% of callers will not hold longer than a minute. A clinic line that makes callers queue for a person is exposed to both.
- Completeness. A front desk needs the full price list in front of it, because a price quoted from memory may be just the first line of the total.
- Memory. If nothing said during a call is written down, nothing survives the call. Every later conversation restarts, and the caller reads that as being a stranger to a business they have already spoken to.
None of the three gets fixed by a training session, because none of the three is a knowledge problem. They are all what happens to a person under load with no reliable place to put things.
What an AI voice agent changes
The neuro-agent picks up on the first ring, quotes every line of the price structure, captures the caller's details, and puts what it has learned into the record before the call ends.
It is worth being precise about why, because it is not intelligence. The agent has the price list open on every single call, and it cannot decide that this particular caller probably only wants the headline figure. It does not get tired at the end of a shift. It does not forget to write the note, because writing the note is not a separate task it has to remember to do.
An agent has the price list open on every call.
What it does not do
It does not diagnose anything and it does not try. It cannot give a final price either, because a final price needs an examination. It says exactly that, in one sentence, and lists the parts the final figure will cover.
It does not replace the coordinator either. The clinical conversation belongs to the coordinator, and a good coordinator handles it better than anything an agent will do this year. What changes is that the coordinator picks up a call with the intake already taken. That is the arrangement we argue for everywhere: the agent takes the part that is arithmetic and the human takes the part that is judgement.
What to do with your own line
You can run this test on your own line this week. It costs a few phone calls, and it lets you hear the line the way a patient does.
- Call your own main number from a phone nobody there recognises, at a normal busy hour, and time the wait before a human voice says anything.
- Ask for the total price of your most expensive service, then ask again in the same call. Compare both answers against your own price list.
- Call back the next day, mention the first call, and see whether anybody knows what you are talking about.
- Score all of it out of sixteen on the eight criteria above, so the result is a number your team can argue with instead of an opinion they can dismiss.
- Fix in this order: answer time, then completeness, then memory. Answer time is the only one your callers are actively timing.
Start with the line you are most confident about, and score it the way a patient would hear it.