Your Answering Service Costs More Than It Saves

An answering service will cut about $900 a month off your front-desk labor line and quietly cost you three or four new patients in the same thirty days. That trade only works if you believe a voicemail is a schedule. It isn't. An answering service doesn't answer your phones; it produces a polite written record of the money that left.

Our view: for any practice with a functioning new-patient flow, the answering service is not a labor substitute. It's a message-taking subscription that gets benchmarked against the wrong alternative, and the comparison owners run in their heads is almost always rigged in the vendor's favor. Plenty of smart operators will disagree, because the invoice math is genuinely attractive and the losses never show up on a P&L line. That's precisely the problem.

The Pitch Is Honest About Cost and Silent About Yield

The sales math isn't fraudulent. Vendor pricing published by Viva AI puts human answering services at roughly $600 to $1,200 a month, or $0.75 to $1.50 a minute, against a fully loaded in-house receptionist at $3,646 to $6,187 a month once you stack 25 to 35 percent benefits on a $35,000 to $55,000 salary. Those are vendor figures, so treat them as directional pricing color rather than gospel, but they're not far off what a practice actually quotes out.

The hard floor underneath them is public. The U.S. Bureau of Labor Statistics puts the median wage for receptionists and information clerks working specifically inside offices of dentists at $20.34 an hour, or $42,460 a year, across 83,950 people in the role as of May 2023. That's the real market price of a human who picks up your phone, and it's an industry-specific number rather than a generic clerical average.

So the delta is real. A service at $800 a month runs $9,600 a year. A full-time dental receptionist at the BLS median, loaded at 28 percent, runs about $54,349. Five and a half times the cost. Any owner looking at those two numbers and nothing else will sign the service contract, and most do.

What the pitch never quantifies is yield: how many of the calls it touches turn into a patient sitting in your chair. The service bills you for handling calls. It does not bill you for booking them, which tells you everything about what it's actually selling.

Coverage Plan B Saves $10,707 and Loses $35,493

Run it on a real shape of practice. Two general dentists, suburban, open 8 to 5 Monday through Thursday, 420 inbound calls a month. The phones ring well outside staffed hours: the 7:30 to 8:30 warm-up, the lunch hour nobody covers, the 5:00 to 6:30 window when working patients finally get a minute, and all day Friday. Call it fifteen hours a week of ring time with nobody on it.

Option A, the service. $800 a month, all in, including modest overage. $9,600 a year.

Option B, buy the hours. Fifteen hours a week is 780 hours a year. At the BLS dental-office median of $20.34, that's $15,865 in wages. Load it at 28 percent for payroll taxes and benefits and you're at $20,307.

Option B costs $10,707 more per year. That's the number the vendor wants in your head, and it's correct.

Now price the yield. Of 420 monthly calls, assume 22 percent land in the uncovered windows: about 92 calls. Assume 12 percent of those are new-patient inquiries, which is conservative for a practice running any marketing: eleven new-patient calls a month into the gap.

A trained front desk with the schedule open on the screen books maybe 65 percent of those: 7.2 patients. An answering service takes a message, and someone calls back the next business day into voicemail, and some fraction of those people have already booked with the practice that answered. Call it 30 percent: 3.3 patients. Those conversion rates are our working assumptions, not published figures, so swap in your own if you track them. Most practices don't, which is the second problem.

The gap is 3.9 new patients a month, or about 46 a year. Dental Economics pegs first-year production per new patient at roughly $850 to $1,300, with lifetime value north of $8,000; that's the author's own industry-benchmark modeling rather than a named dataset, so use it as a frame and not a load-bearing figure. At the midpoint of $1,000, the gap is worth $46,200 a year in first-year production alone.

Net: Option A saves $10,707 in labor and forfeits $46,200 in production. The service costs you $35,493.

Here's the version that matters even if you hate every assumption above. The service has to not cost you eleven new patients a year to break even. Eleven. Less than one a month. If you think a message-and-callback workflow loses fewer than one new patient a month against a live human who can book, keep the contract. We don't think that, and we'd drop the plan.

Speed of Answer Is Half the Variable

There's a defensible case that answering services beat the alternative on the one thing patients notice first, which is whether anyone picks up at all. The evidence that this matters is better than most owners assume. A peer-reviewed study of 252,145 patients across 285 VHA medical centers, published in the American Journal of Managed Care, found that patients calling facilities in the slowest average-speed-of-answer quartile, between 107.7 and 351.3 seconds, had significantly lower odds of reporting they could get an urgent appointment (OR 0.85, p<.01) or get one within a day (OR 0.84, p<.001).

Read that carefully, because it cuts both ways. Phone handling measurably moves what patients believe about access, which most owners file under soft variables and ignore. But the study measured speed to pick up, not whether the person picking up could do anything. An answering service wins the first half of that equation and structurally cannot compete on the second. Fast pickup plus no booking authority is a better experience than a ringing phone and a worse outcome than a human with your schedule open.

Buying speed without authority is buying the measurement, not the result.

What Most Owners Get Wrong: Benchmarking Against a Full-Time Hire

The specific mistake, and it is made overwhelmingly by solo and two-doctor owners who are also the practice's de facto office manager, is comparing a $800-a-month service to a $54,000 full-time receptionist. That comparison is a fiction. You were never going to hire a second full-time front-desk person for fifteen hours of gap coverage. Nobody does.

The honest comparison is the service against marginal hours: a part-timer, a cross-trained assistant with phone duty, a current team member moved onto a split shift, or a remote scheduler paid hourly with write access to your practice management software. Against a $54,000 hire, the service looks like a steal. Against $20,307 of targeted part-time coverage that can actually book, it looks like what it is, which is a $10,707 discount on a 55-percent reduction in conversion.

The second mistake follows from the first: nobody audits the service. Owners buy it to stop thinking about the phones, which means the one thing the purchase guarantees is that the owner never looks at the call log again. Ask most owners who use a service how many new-patient calls it handled last month and how many of those became appointments and you'll get a shrug. The service doesn't report it because the service isn't measured on it.

When We'd Keep the Service

We're not arguing these things have no use. There's a real structural reason the category is growing. Per ADA News, citing ADA Health Policy Institute polling from late 2024, roughly three in five dentists named recruiting and retaining staff as a top expected challenge for 2025, and fewer than half of practices planned to add staff at all. The answering service is a rational response to a genuine hiring market, not a scam.

We'd keep one for after-hours emergency triage, where the job is routing a swollen face to the on-call doctor and nothing else. We'd keep one through a maternity leave, a resignation, or a holiday closure, as a bridge measured in weeks. We'd keep one as true overflow, catching call four when three lines are already live, because a message beats a busy signal every time.

What we wouldn't do is let a bridge become architecture. If the service has been handling your 8 a.m. and 5 p.m. new-patient calls for eighteen months, you didn't solve a staffing problem. You capitalized one.

Four Contract Terms We'd Refuse to Sign

• Per-minute billing with no spam filter. If you're paying $1.50 a minute for a supply rep and a Medicare robocall, you're funding the vendor's worst incentive.
• No call recordings released to you. If you can't listen to how your practice sounded, you can't manage it. Non-negotiable.
• No write access to the schedule. A service that can't book is a service that can't convert. If the vendor won't integrate, price the contract as marketing waste, not labor savings.
• No per-call outcome reporting. Calls handled is a vanity metric. Demand new-patient calls received and appointments booked, monthly, or assume both numbers are bad.

Call Your Own Office Before Friday

Dial your main line at 7:45 a.m., 12:30 p.m., and 5:30 p.m. this week from a number nobody recognizes, and act like a new patient with a cracked molar. Time the pickup, then time how long until someone calls you back with an actual appointment offer. Pull last month's call log from the service, count the new-patient inquiries, and match them against your practice management software to see how many got scheduled. Then price fifteen hours a week of live coverage at $20.34 and compare the two numbers honestly.

Front-desk labor is one of the most mispriced lines in a dental P&L, and it almost never gets isolated from the rest of overhead. Pull your real staffing percentage, drop in the coverage hours you're actually paying for versus the ones you need, and run your own numbers on the overhead calculator before the next contract renewal.

What we're watching: the AI answering tools now writing directly into practice management schedules at $300 to $500 a month. If one of them produces audited book-rate data on live new-patient calls that matches a trained human, the math in this piece flips and we'll say so. So far every vendor we've asked for that number has sent back calls answered instead. Until someone shows the booking rate, assume there's a reason they're showing you the other metric.