It is the most common opening line a dental practice hears. About one in four new patient callers leads with some version of “do you take my plan,” and how the next ten seconds go decides whether that person becomes a patient or a competitor's patient.
Most offices lose the call right there. Not because the answer is no, but because the answer is just no. The caller hears one word, says thanks, and hangs up. Nobody asked what they needed, nobody found out they had been putting off a cracked molar for four months, and nobody mentioned that the practice files out-of-network claims and that the difference on their plan would have been about forty dollars.
Short answer: A dental office should never answer the insurance question with a flat yes or no. The right response confirms the plan name, tells the caller the practice files claims for their plan whether in network or out, explains that exact coverage depends on their specific benefits and will be verified before the visit, then immediately asks what brought them in and offers two appointment times. The goal of the call is the appointment, not the coverage determination, and the coverage gets confirmed after the caller is on the schedule.
What follows is the script, the reason it works, what these calls are worth, and the harder problem underneath all of it: the front desk cannot answer the phone because the front desk is on hold with an insurance company.
Why “no, we are out of network” loses a patient who would have booked
When a caller hears no, that is genuinely all they hear. The rest of the sentence does not land. They assume out of network means they pay full price with no help, which is usually wrong, and they move down the Google results.
Two facts make this expensive. First, commonly cited dental phone benchmarks put the share of new patients who call the office before their first visit at roughly 98%, so the phone is the entire front door. Second, the average practice converts only 50% to 60% of new patient calls into booked appointments, while practices that train and structure their phones run above 80%. That 20 to 30 point gap is not a marketing problem. It is a conversation problem, and the insurance question is where most of it happens.
The value at stake per call is real. Industry estimates commonly put a new dental patient at roughly $850 to $1,300 in first-year production, and $15,000 to $25,000 in lifetime value once you count recurring hygiene, restorative work, and family members who follow them in. Losing one call to a one-word answer is not losing a cleaning. It is losing a decade of a household.
The answer that keeps the caller on the line
Here is the shape of it. Confirm, reframe, redirect, book.
“We do work with Delta Dental. We are out of network on that specific plan, but we file the claim for you either way, so you are not paying anything up front out of pocket beyond your portion. On most Delta plans the difference between us and an in-network office ends up being pretty small, and I can verify your exact benefits before you come in so there are no surprises. What brought you in, is this a cleaning or is something bothering you? I have Thursday at 3:00 or Monday at 8:30.”
Count what that does. It never says no. It tells the truth about network status, which matters for trust. It removes the fear underneath the question, which is not “are you in network,” it is “am I about to get a bill I cannot pay.” It promises verification, which is a concrete commitment rather than a brush-off. Then it turns the conversation toward the clinical need and offers two specific times, because “would you like to schedule” invites a no and “Thursday at 3:00 or Monday at 8:30” invites a choice.
The script is not the hard part. Every practice management consultant in the country teaches some version of it. The hard part is that nobody picks up the phone to say it.
Why the front desk cannot answer the phone, and why it is not their fault
Look at what insurance work actually costs your front desk in hours.
A single manual verification runs 12 to 13 minutes per patient on average, and regularly stretches to 18 minutes or longer when the payer routes you through an IVR maze, asks you to authenticate twice, transfers you to a benefits specialist, and drops the line. For a practice with 20 patients on tomorrow's schedule, that is two and a half to three hours of someone sitting on hold, reading back frequency limitations and deductible status. Insurance and billing questions also make up roughly 12% to 18% of all inbound calls, and those calls run longer than average.
So the same person who is supposed to answer the new patient call is on another line, on hold, verifying benefits for a patient who is already booked. The phone rings. It rings again. It rolls to voicemail.
The results show up in the miss rate. Industry benchmarks put dental missed call rates at 30% to 38% of incoming calls during business hours, and some locations run far higher. That is during business hours, with staff in the building, doing their jobs correctly. It is a math problem, not a performance problem, and hiring a second front desk person only moves it if the second person is free at the exact moments the first one is not.
How does an AI receptionist handle the insurance question?
This is a well-bounded problem, which is what makes it a good fit for a bilingual AI receptionist rather than a risky one.
You configure it once. You load the plans you are in network with, the plans you are out of network with but still file for, the plans you do not work with at all, and the exact language you want used for each. You set the rule that it never quotes a coverage percentage, a deductible, or an estimated patient portion, because those depend on the individual's remaining benefits and nobody should be guessing at them on a first call. You give it your new patient exam fee and your policy on payment plans and third-party financing.
From there the call runs the way your best treatment coordinator would run it. Sol answers on the first ring, in English or Spanish depending on how the caller starts speaking. It takes the plan name and subscriber details, gives your configured answer about network status, makes the verification promise, asks what brought the caller in, and books them into an open slot in your calendar. It confirms by text before hanging up and leaves the plan information and the chief complaint in your notes so your coordinator can run verification before the visit with everything already in hand.
Anything outside the configured boundaries gets handed to a person. A caller asking what a crown will cost on their specific plan, a billing dispute, a clinical question about pain that needs triage judgment, all of it routes to your team with the context attached. Sol backs up the front desk. It does not make coverage determinations and it does not practice dentistry.
What about Spanish-speaking callers?
The insurance conversation is exactly where a language gap turns into a lost patient. Network status, claim filing, deductibles, and estimated portions are confusing in your first language and close to impossible in your second.
In most US markets a meaningful share of new patient calls come from Spanish-speaking households, and very few practices have bilingual coverage at the front desk, let alone after hours. Sol handles the whole conversation in Spanish, including the network explanation and the booking, and follows a caller who switches between languages mid-sentence the way bilingual families actually speak. Your coordinator gets the notes in English.
What the math looks like for a typical practice
Take a two-doctor practice taking 400 inbound calls a month. At a 32% miss rate, that is roughly 128 unanswered calls. Say a quarter of those are new patients, which is conservative for a practice running any kind of marketing. That is 32 missed new patient calls.
Even if your team would only have converted half of them, you lost 16 new patients in a month. At $850 in first-year production, that is about $13,600. Carried out to lifetime value on the ones who would have stayed, the number stops being a rounding error and starts being the reason the schedule has holes in it.
Against that, 24/7 AI coverage costs a fraction of one part-time front desk salary. It does not go on hold, it does not take lunch, and it does not stop answering the main line because it is stuck in a payer's phone tree.
The objections worth answering honestly
“Insurance is too complicated for an AI to handle.” Which is why it should not handle all of it. Network status and claim filing policy are fixed facts you already know. Coverage estimates are not, and Sol is configured not to touch them. The line is drawn on purpose.
“Patients will be annoyed it is not a person.” Less annoyed than by a voicemail box. Of callers who reach voicemail, the large majority never call back and most simply dial the next practice on the list. A helpful answer in ten seconds beats a callback tomorrow every time.
“We already have an answering service.” Answering services take messages. They do not know which plans you file for, they cannot see your schedule, and they hand you a callback list in the morning that you work through while today's calls go unanswered.
“Our team is great on the phone.” They probably are. This is not about the calls they take. It is about the calls that ring while they are on hold with Cigna.
Stop losing patients at the first question
The insurance question is the cheapest new patient you will ever get, because that person already found you, already decided to call, and is already telling you they want to come in. All they want is permission to stop worrying about the bill.