Front desk phone scripts that convert

The cheapest person on your payroll controls the most expensive number in your practice. Dental offices pay receptionists a median of $20.34 an hour and medical secretaries $22.28, and there are 175,520 of them answering phones in this industry, according to the Bureau of Labor Statistics. That workforce converts the average new-patient phone inquiry into a booked appointment 23% of the time, while practices with strong phone skills book 70% to 90% of the same calls, per the ADA's analysis of a Viva Concepts study covering more than 10,000 calls.

Here's our call, and plenty of owners will hate it: the script is the smallest part of the fix. Scripts don't convert, permissions do. The reason your front desk ends a call with "let me check with the doctor and get back to you" isn't vocabulary, it's that the person holding the phone can't actually give away a time slot, can't quote a fee, can't shorten an appointment, and can't override your block schedule. We'd take your current staff and your current wording and hand them authority over the first 30 minutes of chair time before we'd send anyone to another phone seminar.

Twenty-Three Percent Is a Permissions Problem

Two credible datasets disagree about how bad this is, and the disagreement is instructive. The ADA's number is 23% of new-patient inquiries. Resonate's call data says average practices convert 50% to 53% of answered calls into appointments, with the top decile at 75%, and separately that average practices miss roughly 20% of new-patient inbound calls while top performers answer 95% or more.

Both can be true at once. One counts every inquiry including the ones that hit voicemail at 11:40 on a Tuesday; the other counts only the conversations that happened. Multiply them and you get close: four out of five calls answered, half of those booked, and you're in the low 40s before no-shows. The gap between 23% and 90% is not one failure. It's a stack of them, and only the bottom one is about what your staff says.

The bottom one is availability. The ADA's patient intake guidance is blunt about the threshold: answer every incoming call by the third ring, and use standardized scripts for the highest-frequency call types, meaning scheduling, rescheduling and canceling, and emergencies. Third ring is roughly 12 seconds. If your one front-desk person is checking in a hygiene patient, posting a payment, and holding a line for an insurance verification, the third ring is a fantasy, and no script in the world gets recited to a dial tone.

The 84-Office Study Says Stop Pitching

The best field evidence we have on what front desks actually say is old and still devastating. Researchers called 84 dental offices as prospective new patients and coded the behavior on the other end. Only about 40% of staff invited the caller to share their questions or concerns. Thirty-one percent instead launched into a promotional pitch about the practice, and the staff who pitched kept callers on hold longer, per the study in Health Communication.

Read that again, because it's the whole problem in one sentence: a third of front desks respond to a buying signal by selling. The caller already dialed. They don't need to hear that you have digital X-rays and a gentle touch. They need to be asked what's going on and then handed a day and a time.

This is also why so much phone training backfires. Owners buy a program, the program supplies value-proposition language, and the staff dutifully converts a 90-second booking conversation into a three-minute commercial. The measured consequence in that study wasn't just lower service orientation. It was longer hold times, because pitching takes time the desk doesn't have.

Our view: delete every sentence in your current script that describes your practice. Replace it with one question and two time slots.

Twelve Patients a Month, $64,035 a Year

Run it on a single-doctor practice taking 40 new-patient calls a month, which is 480 a year.

At average performance, you miss 20%, so eight calls a month never reach a human. That leaves 32 answered. Convert those at 52% and you book 16.6 new patients a month.

Now run top-decile mechanics on identical call volume. Answer 95% and you reach 38 of the 40. Convert at 75% and you book 28.5. That's a gap of 11.9 new patients every month, or 142 a year, on the same marketing spend and the same phone number.

Put a conservative $900 of first-year production on each new patient, which is roughly an exam, images, and two hygiene visits with a modest restorative rate. The full gap is worth $128,070 a year. We don't believe any real practice captures all of it, so halve it for scheduling friction, no-shows, and the fact that some of those callers were never going to show up anywhere: $64,035.

Against that, one full-time front-desk hire at the BLS median costs $42,307 in wages before taxes and benefits. The discounted half of your phone gap pays for one and a half of them.

And at 480 annual calls, each single percentage point of combined answer-and-book rate is worth 4.8 patients, or $4,320 a year. That's the real unit of account here. One point. Every staffing decision, every hold-music decision, every "we'll call you back" is priced in points.

What Most Owners Get Wrong: Grading Tone Instead of Outcomes

The specific mistake is this: owners listen to call recordings for manners and greeting compliance, then grade the call on whether their staff sounded warm. We've watched owners play a recording, nod approvingly at a friendly, professional, thoroughly pleasant four-minute conversation that ended with no appointment, and conclude the phones are fine.

The owners who do this most are the ones who just hired a marketing agency. They're six weeks into a new spend, the dashboard shows call volume climbing, and the agency reports calls as the deliverable. Volume goes up, bookings don't, and the owner's conclusion is that the leads are low quality. The leads are fine. One in five of them is hitting voicemail and half the rest are getting a brochure read to them.

Multi-location owners make a second version of the same error. They buy a call-tracking platform, pull reports on ring time and abandonment, and never once listen to a call that was answered and lost. Ring time is a plumbing metric. Booked-or-not is the only outcome.

There's a third group, and this is the one closest to the permissions argument: owners who guard the schedule so tightly that saying yes is above the front desk's pay grade. If a new patient can only go in a specific block, and that block needs doctor approval to open, and the doctor is in a crown prep, the honest script is "we'll call you back." Then a competitor two miles away who answered by the third ring gets the patient. You didn't lose on words. You lost on org chart.

The Four Scripts Worth Standardizing

The ADA names scheduling, rescheduling and canceling, and emergencies as the call types that need standard language. We'd add the fee question, because it's where most lost calls actually die. Four scripts, each short enough to live on one index card.

• New patient, no urgency. Open with a question, not a pitch: "Happy to get you in. What's going on with your teeth right now?" Listen, then close with two options and nothing else: "I can do Thursday at 8:10 or Monday at 3:40. Which is easier?" Never ask "when would you like to come in?" An open question hands the scheduling problem back to a stranger who doesn't know your schedule.
• Pain, swelling, or trauma. This one is a protocol, not a conversation. Any mention of those three words gets a same-day or next-morning slot with no internal debate. The script is: "That needs to be seen today. Can you be here by 2?" If your team has to ask the doctor whether a swelling patient can be seen, you don't have an emergency script, you have an escalation path.
• Cancellation and reschedule. The losing move is "okay, call us when you know your schedule." The winning move is never ending the call without a date: "No problem, I'm taking you off Tuesday. Let's put you somewhere now. I have Thursday at 10 or next Wednesday at 4." A canceled appointment with no replacement date is a lost patient with extra steps.
• The fee question. Most desks either dodge it or recite a number and hang up. Do both halves: "A new-patient exam with images and a cleaning runs about $X here. Do you want me to check what your plan covers, or should I hold you a time and we'll sort the plan before you come in?" The second option is the one that converts. Quote the fee, then put a time on the table in the same breath.

Note what's missing. No practice history. No technology list. No "we're a family practice that treats you like family." The 84-office study already told us what that language does to hold times and service orientation.

Three Changes We'd Make Before Hiring Anyone

We'd give the front desk standing authority to open any slot for a new patient inside the next 72 hours without asking. We'd route overflow calls to a second answerable extension before we spent another dollar on ads, because reaching 95% answer rate is arithmetic, not skill. And we'd grade calls on one binary: did the caller end the conversation with a date and a time? Tone is coachable later. Booked is the number.

We would not buy another phone training course this quarter. At $4,320 a point, the course has to move your combined rate by more than a point to break even, and courses that teach value-proposition language reliably move it the wrong way.

Do This Before Friday

Pull ten answered new-patient calls from last week and score each one booked or not booked, ignoring everything else. Count how many ended with the staff offering two specific times, and how many ended with "call us back." Then write the four scripts above on index cards, tape them to the phone, and tell your front desk in writing that they can open any slot in the next three days without asking. Measure the same ten-call sample again in two weeks.

What we're tracking next: whether the AI answering services now flooding this market actually close the 20% miss rate without wrecking conversion on the calls they pick up. Early signs are that they answer beautifully and book badly, which would make them a plumbing fix sold as a growth fix. We'd change our position on the permissions argument if we saw a cohort of practices with full schedule authority at the desk still stuck near 50%. So far we haven't.

Front-desk payroll is one of the largest controllable lines in your overhead, and it's the line that decides whether 480 calls a year turn into 200 patients or 350. Before you add another salary to it, see where your administrative cost actually sits against collections and what a point of phone conversion is worth in your numbers. Run your own numbers on our free overhead calculator and find out whether your phone problem is a staffing problem or a permissions one.