What a front desk hire actually costs a primary care practice

Before I built MedPhone I assumed the hardest objection from clinic owners would be that they do not trust AI to talk to their patients. I was wrong.

The objection I actually get is far simpler. We already have a front desk. What is the return?

That deserves a real answer rather than marketing arithmetic, so I sat down with the numbers.

The number nobody sees

A front desk hire at a US primary care practice does not cost $35,000 a year. That is the salary line. The real figure is closer to $52,000. Here is the stack.

• Base salary: about $35,000 for a medical secretary or receptionist, per BLS occupational data. Urban New York and California run 20 percent higher, the rural south about 15 percent lower
• Payroll taxes: roughly 8 percent on top. FICA at 7.65 percent plus SUTA and FUTA depending on state. Call it $2,800
• Benefits: where it balloons. Health insurance is the big one, $6,000 to $9,000 a year with an employer contribution. Retirement adds $1,000 to $1,750 if you offer it. Carrying PTO is about $1,400
• Equipment and training: workstation, phone, EHR seat license, software, training time. Amortized, $2,000 to $3,000 a year
• Turnover: this one hurts. Medical front desk turnover runs about 25 percent annually, and each event costs one and a half to two months of the departing salary in lost productivity, recruiting and training. Amortized, another $2,000 to $3,000

Total, $50,000 to $54,000 depending on region and practice. I use $52,000 in every conversation because it is the middle of that range, and because it matches what practice owners tell me they are actually paying once they run the same stack themselves.

How the number varies by region and practice size

The $52,000 is a national middle. Where you are and how big you are both move it, and they move it in opposite directions often enough to be worth separating.

The table below applies the regional adjustments in the BLS occupational data to the same cost stack above. It is derived rather than surveyed, so treat it as the shape of the variation rather than a quote for your market.

Practice size cuts the other way. A single provider practice pays the same fully loaded cost for one person as a five provider practice does, but spreads it across a fifth of the patient panel. That is why the arithmetic on any front desk decision looks very different at one provider than at five, and why a per seat priced alternative is a particularly bad fit for the larger one.

One thing that does not vary much is turnover. Front desk turnover runs around a quarter of staff annually in most markets, and in a tight rural hiring pool the cost per event is higher rather than lower because the vacancy stays open longer.

What that buys, and what it does not

Forty hours a week. One caller at a time. Sick days, vacation and turnover.

Here is the gap most owners underestimate. Your front desk is not on the phone for forty hours. It is more like twenty five to thirty. The rest goes on check in, insurance verification, confirming appointments, the fax machine that somehow still exists, managing the waiting room, and lunch.

During those other ten to fifteen hours, the phone rings and nobody picks up.

Then there is everything outside opening hours. One hundred and twenty eight hours a week when the front desk is off. Weekends. Holidays. The 7:30am rush when patients are already dialing and your staff are not in yet.

The invisible cost sitting next to it

This is the number that shocked me the first time I ran it for a customer.

A typical primary care practice takes about 800 inbound calls a month. Abandonment at a busy practice runs 15 to 25 percent, whether that is a ring out at three minutes, a hang up in the queue, or a busy signal. Around 15 percent of those missed calls are new patient inquiries, which is the MGMA benchmark and a figure you can check against your own call log.

Here is the chain, and it matters that every step is in it. Most vendor calculators stop after the second one, which is how they arrive at numbers three times too big.

1. 800 calls a month at 20 percent abandoned is 160 missed a month, or 1,920 a year
2. At 15 percent new patient inquiries, that is 288 inquiries lost
3. At a 50 percent inquiry to booked conversion, 144 of those would have become patients
4. Around 65 percent of unreached callers never try again, so 94 are gone for good
5. At $3,000 primary care lifetime value, that is about $282,000 a year

The two steps in the middle are the ones that keep this honest. Not every inquiry converts, and not every unreached caller is lost, because a third of them do ring back. Skip those and you get a number you cannot defend in front of a practice owner.

Nobody sees it because there is no line on the P&L that reads "revenue we lost because nobody picked up". So it does not get counted, and what does not get counted does not get fixed.

The alternative, honestly compared

MedPhone starts at $150 a month and is billed on the calls it answers, so a typical practice lands at roughly a third of what a hire costs. Some competitors charge per seat, which means five providers can cost you considerably more than that.

When this is the wrong answer

I would rather say this plainly than have you find out after signing.

• Concierge medicine and high touch specialty. If your model depends on patients hearing one specific person when they call, no AI fits, however good it gets. Some concierge practices deliberately pay for two front desk hires, because the personal touch is the product
• Very small practices where the front desk is also the practice manager. If that person also runs billing, scheduling, credentialing and ordering, taking the phones off them does not save you a hire. It gives back part of their day. That can still be worth it, but the return is much smaller
• Practices with genuinely unusual clinical call flows. Some specialties have patterns specific enough that the configuration time is hard to justify at small scale

For most primary care, family medicine, dental and mid sized specialty practices, the arithmetic works.

The worked example

A practice taking 800 calls a month at 20 percent abandoned, run through the full chain above: 94 patients a year lost for good, at $3,000 lifetime value.

• Front desk hire: $52,000
• Missed new patient lifetime revenue: about $282,000
• Front desk hire plus missed revenue: about $334,000

MedPhone runs at a fraction of the hire, and on a conservative 80 percent capture it recovers a large share of that missed lifetime revenue. Nobody I have sold to has let a receptionist go, so the honest framing is not net savings against a hire. It is that the recovered revenue covers the cost many times over, and the return you actually feel is the hours back and a phone that gets answered.

That is the arithmetic I use in sales conversations, and it is the arithmetic a family medicine practice we work with actually ran before they signed. Every assumption in it is on the page above, so disagree with any of them and redo it.

The one thing to do next

You do not need to buy MedPhone. You do not need to buy any AI phone agent.

What you should do is run this on your own practice. Log calls for one full week and note total inbound, abandoned, and new patient inquiries. Your phone system may already have it, so ask. Then multiply out the annual revenue at risk and set it against both the cost of a hire and the cost of an AI alternative.

If the number surprises you, it should. It surprised me the first time. If it does not, and you already run your practice this way, you are ahead of most of the country and you probably already know whether this fits you.

Either way, put a real figure on the missed call revenue. That figure changes decisions.

Run it on your practice

Questions people ask

Is $52,000 specific to primary care?

It is the primary care benchmark. Specialty practices often pay more for the credentialing knowledge, rural practices less, urban more. $52,000 is the median I see.

What if we do not take 800 calls a month?

It scales linearly. At 400 calls, halve everything. The break even threshold moves with it.

What does MedPhone actually cost?

It starts at $150 a month and is priced on the calls handled rather than per seat, so a practice with unusually heavy volume pays more and a quiet one pays less. Adding a provider or a receptionist never moves the number. There is a calculator on the pricing page that will size it against your own call volume.

Is 25 percent turnover on front desk staff really normal?

Unfortunately yes. It is a high stress, moderate pay role with a clear ceiling. Turnover is one of the largest hidden costs practice owners carry.

How much of the missed call revenue can this actually capture?

At Synergy Medical we resolve an estimated 65 to 75 percent of calls without them reaching staff, up from just over half at launch. The first few weeks of any deployment are the noisiest while it learns your patterns.