You’re Not Understaffed. You’re Misarchitected.

In the last 18 months, the conversation among COOs and VPs of Operations at medical groups we’ve spoken with has been remarkably consistent: we can’t hire fast enough, we can’t keep people, and the phones never stop.

The initial instinct is to add headcount. Post the roles. Raise the wages. Approve the FTEs. That gut feeling seems natural.

When patients can’t get through to schedule, you need someone to answer the phone. When prior auth backlogs are climbing, you need more staff to work them down. When front-desk turnover hits 40% annually, the immediate answer looks like more people in the seats.

But intuition and logic are not the same thing. And in the case of front-office staffing, acting on the wrong diagnosis is not a simple mistake.  It’s an expensive one that compounds every year if you don’t fix the underlying problem.

The problem is routing, not resource count

Call abandonment is the canary in the front-office coal mine. Industry data consistently shows that groups running abandonment rates above 8% are at the threshold where patient loss becomes measurable. Not just inconvenient, but a real risk to revenue and the patient experience.

Here’s what’s almost never true: calls are being abandoned because they don’t have enough people.

What is more likely is  that the calls are going to the wrong people in the wrong way. A patient calling to reschedule hits the same queue as a patient navigating a prior auth dispute. A medication refill request competes for the same staff attention as a new patient intake. Nobody designed that system intentionally. It evolved, one workaround at a time, until the architecture of the front office no longer resembles anything logical.

When you add an FTE to that system, you aren’t fixing the routing. You’re adding a person to a broken queue. The abandonment rate improves temporarily and then the next wave of call volume arrives and you’re right back where you started, except now you have a larger payroll and another person to train.

What you’re actually paying for when you add staff

The math is not in your favor. A single front-office FTE at a 50-100 provider group costs $45,000–$65,000 per year in loaded compensation and that’s before you account for recruiting, onboarding, and the productivity gap during the first 60 to 90 days of ramp. At 40% annual turnover (a conservative estimate for this role category), that investment resets more than once a year.

But the real cost isn’t the dollar figure. It’s the opportunity cost of what that headcount was supposed to fix. A group that adds three FTEs to address a 12% abandonment rate and gets to 9% has spent roughly $180,000 in annual loaded cost to remain above the threshold where patient loss is measurable.

They’ve also added three more people to a system that will keep generating the same pressure until someone changes the architecture.

The groups spending the most on front-office staffing are not the groups with the best patient access metrics. They are the groups with the most urgent unaddressed need for a structural fix.

What ‘architecture’ actually means, and what it doesn’t

Rearchitecting a front office is not a technology project. It starts with a single, explicit question that most groups have never formally answered: which interactions require human judgment, which can be resolved with automation, and which need a human in the loop?

Most groups, if asked, will guess. When you instrument the actual call volume; capturing handle time, call type, resolution rate, and escalation path, the picture that emerges is almost always the same: 55 to 70 percent of inbound call volume does not require the judgment of a trained front-office staff member.

Appointment confirmations. Basic eligibility questions. Directions. Prescription routing. Fax follow-ups.

That time is not recoverable with more people. Every hour a trained staff member spends confirming a Tuesday appointment that could have been confirmed automatically is an hour not spent on a prior auth that requires human intervention, a new patient intake that requires empathy and complexity, or a billing dispute that requires judgment and documentation.

The architecture fix is deciding explicitly, with data, what belongs to humans and what doesn’t. Until that decision is made, you’re staffing for a system you haven’t designed.

The compounding problem nobody talks about

There’s a second-order effect that doesn’t show up in the staffing budget but shows up everywhere else: burnout and turnover.

Front-office staff who spend the majority of their shift on low-judgment, high-volume interactions  that should be automated don’t stay.

Not because the pay is wrong (though it often is) and not because the culture is bad (though it sometimes is), but because the job as designed offers no room for the kind of work that makes a person feel like they’re contributing something meaningful. Answering the same eligibility question forty times in a shift is not a career. It’s a treadmill.

Groups that rearchitect their front offices (routing the automatable interactions away from human staff and focusing human attention on the interactions that actually require it) consistently report two things: better access metrics and lower turnover.

The staff who remain are doing harder, more engaging work. The patients who call are reaching someone who has the bandwidth to actually help them.

What this means for your operation right now

This is not an argument against hiring. Some groups are genuinely under-resourced and need more people in the near term. But for most groups in the 50-100 provider range (the groups where the staffing pressure is loudest and the operational complexity is highest)  the right question to ask before the next hire is a diagnostic one:

What percentage of our current inbound call volume actually requires a human to resolve?

If your team can’t answer that question with a real number — not an estimate, not a gut feel — you don’t yet have the information you need to make a good staffing decision. You have the information you need to start measuring.

The groups that will close 2026 with better patient access and lower overhead than they started with are not the groups that hired the most. They’re the groups that asked the harder question first and were willing to hear an uncomfortable answer.

If your call abandonment rate is above 8%, you’re not understaffed. You’re misarchitected. The fix isn’t a headcount decision. It’s an architecture decision. And it starts with measuring what you’re actually asking your staff to do.