The calls you don't make

If you sit at the front desk of a busy specialty clinic for an hour, here is what you will see. The phone rings, and someone answers it. A patient walks up to the window, and someone helps them. A fax tray fills up, and someone files it. A portal message buzzes, and someone replies. The work is reactive by design, because the work that walks in the door has a face, and the work that needs to be done at some point today does not.

This is the unspoken hierarchy of every front desk in healthcare. Inbound wins. Outbound waits.

The trouble is that outbound is where the practice lives or dies. The reminder calls that prevent no-shows, the rescheduling calls that recover canceled slots before they cost the day, the recall calls that bring lapsed patients back through the door. None of those tasks have a face standing at the counter. None of them ring with the same urgency as a patient who is already on the phone. So they slip. They slip a little on Monday, more on Tuesday, and by Friday the list of calls that should have been made has quietly become the list of revenue and outcomes that were lost.

I want to put a number on what that costs, because the cost is much larger than most clinic owners realize.

The revenue side

The average independent physician practice loses around $150,000 a year to no-shows alone. That figure has been remarkably stable for years across MGMA-tracked data. Some specialties suffer worse. Sleep clinics see no-show rates around 39 percent. Dermatology and pediatrics hover near 30 percent. Ophthalmology and oncology sit around 25. Even the better-attended specialties like endocrinology lose 14 percent of their appointments to people who simply don't show up. Across the U.S. healthcare system, the annual cost of patient no-shows has been estimated at $150 billion.

Inside any single clinic, the math gets more visceral. If your average visit is worth $200 and you run 2,000 appointments a month, a 15 percent no-show rate is $60,000 walking out the door every month. Multiply that across a year and you are looking at three quarters of a million dollars of unrealized revenue inside a practice that is, by every other measure, fully booked.

What separates the practices that survive this from the ones that don't is whether someone called. The Memorial Hospital at Gulfport case, well-documented in the operations literature, is a useful illustration. They built a system around multi-channel reminders and proactive rescheduling, dropped their no-show rate by 28 percent, and recovered just over a million dollars a year. Not because they bought new equipment or hired new providers. Because somebody picked up the phone on the right day at the right time.

Reactivations are the same story told at a different scale. Most clinics have a few hundred to a few thousand patients who used to come in and just stopped. They moved, they got new insurance, they had a bad experience, they forgot. The dental world has studied this most rigorously, and the numbers are sobering. The typical practice loses 15 to 20 percent of its active patient base every year to attrition. At an average lifetime value of $5,000 to $15,000 per patient depending on the specialty, a 2,000-patient practice is bleeding somewhere between $1.5 million and $6 million in compounding revenue every year, just to the people who used to show up and don't anymore.

A well-run reactivation campaign brings 15 to 25 percent of those patients back. The math is not subtle. Reactivating a lapsed patient costs roughly 20 percent of what it costs to acquire a new one, and the patient already trusts you. It is, by a wide margin, the highest-ROI marketing activity any clinic can do. And almost no one does it consistently, because consistent reactivation requires hundreds of phone calls a week, and the front desk is busy answering the phone.

The patient safety side

If the revenue case were the whole story, this would still be worth fixing. But the deeper case for outbound is clinical, and I want to spend a moment on it because in my experience this is where the real urgency lives.

A landmark study published in BMC Medicine tracked 824,374 patients in Scotland over three years. The finding that should sit in the back of every clinic owner's mind is this: patients who repeatedly miss general practice appointments have markedly higher mortality than patients who attend. Not slightly higher. Markedly. Around 5 percent of patients with mental and physical comorbidities who miss two or more appointments in a year die within twelve months. Missing appointments turns out to be one of the strongest behavioral predictors of premature death that we have.

The same pattern shows up in specialty care. A 2024 study of stage I lung cancer patients found that missing a greater proportion of oncology follow-up appointments in the first year after diagnosis was independently associated with increased mortality. A meta-analysis of 34 cancer studies showed that every month of treatment delay corresponds to roughly a 10 percent increase in mortality. The HIV literature, the vascular surgery literature, and the chronic disease management literature all tell the same story in different vocabularies. The patient who does not come back is, very often, the patient who is sickest.

This is the part that should be harder to live with. A missed appointment is not a billing event. It is a clinical signal. And the call you do not make to a patient who has missed two visits is not just a lost slot on Tuesday. It is a moment where a human being, often the most vulnerable kind, has slipped out of the system that was supposed to be watching for them.

In oncology this is the patient whose tumor was caught early and is now growing again. In cardiology it is the post-MI patient who never came in for their three-month review. In ophthalmology it is the diabetic who is going blind on a timeline they cannot see. In behavioral health it is the patient whose suicidal ideation went undocumented because the intake never happened. The call that nobody made on Wednesday afternoon, because the front desk was buried, becomes the bad outcome that nobody can quite explain on the morbidity and mortality round.

The front desk is not failing

I want to be careful here, because it is easy to read the numbers above and conclude that front desk staff are not doing their job. The opposite is true. The front desk is doing the impossible job of triaging fifty fires at once with no margin for error and a phone that will not stop ringing. They are not failing. The math is failing them.

A specialty front desk in a clinic of any size is asked to be a switchboard, a concierge, a billing analyst, a clinical translator, an emotional buffer, an insurance lawyer, and a sales rep, all at the same time, all day, in real time. The reason outbound work loses is not that the team does not care. It loses because outbound work is calm and quiet and uninterrupted, and the front desk is none of those things. It is the cognitive equivalent of asking a chef to prep next week's menu in the middle of dinner service. It will not happen, no matter how skilled the chef.

This is the structural problem. And it does not get solved by hiring another person, because that person also has to answer the phone.

Where AI agents change the math

For the past few years the conversation about AI in clinic operations has mostly been about chatbots that handle inbound chat or automated text reminders that go out on a schedule. Useful, but not transformative. A reminder text that nobody answers is just noise. A bot that hands off to a human at the first sign of complexity is just a slower IVR.

What changes the math is an agent that can actually conduct an outbound conversation. One that calls the patient, confirms or reschedules them, handles the objection, finds the next slot, books it, and writes the note back into the EHR without anyone at the clinic having to lift a finger. One that runs reactivation campaigns at scale, making three hundred personalized calls in an afternoon to patients who lapsed eighteen months ago, in a voice that sounds like a member of the team, with a follow-up plan for everyone who didn't pick up. One that triages the patients who missed two appointments in a row and surfaces them to the clinical team, instead of letting them disappear into the silence of an unworked recall list.

This is what we have built Taxo to do. It is not a replacement for the front desk. It is the staff member that does the calls the front desk never has time to make. Reminders, reschedules, recalls, post-visit check-ins, recall list cleanup, insurance verification follow-ups, the long tail of outbound work that quietly compounds in either direction depending on whether someone, anyone, picks up the phone.

What the front desk gets to do instead

The most interesting thing that happens when a clinic delegates outbound to an AI agent is not the saved revenue or the recovered patients. Those are real, and they are quickly visible. The more durable change is what the front desk team starts doing with their time.

When the calls that need to happen are happening, the human in the front office stops being a switchboard. They start being the person who takes the patient who is crying in the waiting room into a private space and figures out what is actually wrong. They start being the person who notices that Mrs. Patel has lost weight since her last visit and flags it to the nurse before she leaves. They start being the person who builds a real relationship with the regulars, who knows the new dad by name, who pulls the provider out of the room to handle a complaint before it becomes a Google review. They start doing the work that humans are uniquely good at and machines are not.

The clinics I have spent the most time inside, going back to my training and my years as an investor and now as a founder, all share a quiet truth. The ones that feel good to be a patient in are the ones where the front desk staff have time. Time to look up. Time to listen. Time to be human. The reason they don't have that time, in most clinics, is that they are buried under outbound work that nobody else can do.

We can change that. The phone calls that need to be made today will be made today. The patients who need to come back will be brought back. The appointments that need to be rescheduled will be rescheduled before they become a hole in tomorrow's schedule. And the people at the front desk, finally, will get to do the job they actually trained for.

The calls you don't make are costing you a quarter of your revenue and, in a meaningful number of cases, costing your patients their lives. There is no longer a good reason not to make them.