Do You Really Need to Hire? Rethinking Healthcare's Staffing Reflex

A practice administrator posts a job requisition for a medical assistant on a Tuesday morning. Six weeks and a dozen interviews later, two promising candidates take other offers before the paperwork clears. Meanwhile, the work that req was supposed to solve, rooming patients faster, returning refill calls, chasing prior authorizations, keeps landing on the three MAs who are already there.

This is worth pausing on, not because hiring is the wrong move. Sometimes it is exactly right. But because the decision to hire often gets made before anyone asks a more basic question: what, specifically, is not getting done, and does fixing it actually require a new person?

MGMA's ongoing polling of medical group leaders shows why this reflex is so strong. As of an October 2024 poll, 53% of practice leaders named finding candidates as their single biggest staffing challenge, ahead of compensation and benefits at 29% and retention at 16%. By mid-2025, 65% of practice leaders said labor was the area driving the biggest increase in their operating costs. When candidates are hard to find and labor keeps getting more expensive, it makes sense that “hire” feels like both the obvious answer and an increasingly costly one.

What a staffing search is actually built to solve

A staffing search, whether run internally or through an agency, is built to answer one question: who can fill this seat? That is a legitimate question when the seat itself is well designed and the work genuinely requires a dedicated person.

But a staffing search does not ask a different, earlier question: does this seat need to exist in its current form? If a role has quietly absorbed tasks that do not belong together, some clinical, some administrative, some that could be automated, filling that role with a new hire just puts a new person into the same overloaded design. Six months later, the practice is often back where it started, this time with a frustrated new employee instead of an open req.

This is not a criticism of staffing agencies or recruiters. It is an observation about sequencing. Filling a seat and designing a seat are two different problems, and most practices only have tools for the first one.

The real question: capacity, not headcount

Capacity is not the same thing as headcount. A practice can add a person and still lose capacity if that person inherits a broken workflow. A practice can also gain meaningful capacity without adding anyone, by clarifying what work actually needs to happen, who should be doing it, and where technology can safely help.

That reframe matters most in the moments that trigger a hiring search in the first place. A provider says they cannot keep up with documentation and patients on the same day. A staff member gives notice and the remaining team scrambles to cover. A denial rate creeps up and someone traces it back to rushed, incomplete notes. Each of those moments feels like a staffing gap. Each one might also be a workflow problem, a role-design problem, or a documentation-burden problem that a new hire would not actually fix.

Documentation is a useful example because the data on it is unusually clear. Physicians have long spent roughly two hours on EHR and desk work for every hour of direct patient care, a pattern first measured in a 2016 time-and-motion study and still visible in more recent research. A 2024 Health Affairs study went a step further, finding that each additional hour spent on documentation reduced the share of a physician's patients whose outside medical records they actually reviewed that day by 7.1 percentage points. Documentation burden, in other words, is not just an after-hours problem. It visibly displaces other clinical work during the day, which looks a lot like a capacity problem because it is one.

Filling a seat and designing a seat are two different problems, and most practices only have tools for the first one.

A quick gut check before you post the next req

Before deciding to hire, it is worth asking what, specifically, is not getting done, and why. “We're short-staffed” is a symptom, not a diagnosis. Look for duplicate documentation, unclear handoffs, or manual work that could be redistributed or supported with the right tools before assuming the fix is a new person.

Revascent's Clinical Informatics Playbook lays out a fuller framework, work, workflow, workforce, technology, and outcomes, for working through that diagnosis step by step. The short version: only after ruling out workflow and delegation fixes does “who do we hire” become the right next question, and by that point, the hire is usually better defined and more likely to stick.

When staffing support still makes sense

None of this is an argument against hiring or against staffing support in general. Sometimes the work genuinely requires a dedicated person, on an ongoing basis, and no amount of workflow redesign changes that. A growing practice adding a new location, a specialty practice scaling patient volume, or a role that has been correctly scoped from the start are all legitimate reasons to hire.

The difference this framework makes is timing. Running through work, workflow, workforce, and technology before defaulting to a hire means that when a practice does bring someone on, whether as a direct employee, a staffing placement, or specialized support like a medical scribe, the role is more likely to be designed around real, sustainable capacity rather than a symptom that will resurface in another six months.

A hypothetical worth working through

Imagine a multi-provider practice where the front desk has been asking for a second scheduler for a while. Visits have grown, phones ring constantly, and the current scheduler is stretched thin. The obvious move is to hire a second scheduler.

But a closer look at the role might show that a meaningful share of the scheduler's day goes to manual, repetitive work: re-entering information that already exists elsewhere in the system, or answering the same handful of routine questions over and over. None of that requires a second full-time hire to fix. It requires fixing the workflow.

Hiring a second scheduler would still help in that scenario. It would also lock in the inefficiency for another employee's entire tenure. Fixing the workflow first might mean the practice needs a second scheduler eventually, but not yet, and not to do the same broken version of the job.

This is a hypothetical, not a specific client outcome. But the pattern it describes, an open role that looks like a headcount problem and turns out to be a workflow problem, shows up often enough that it is worth checking for before any job goes live.

Where this leaves practice leaders

The instinct to hire when something feels short-staffed is understandable. It is also, often, the most expensive and slowest way to solve a capacity problem, especially in a labor market where 53% of practice leaders already say finding candidates is their top challenge.

A more useful starting question is not “who can we place” but “where are we actually losing capacity, and what combination of people, workflow, and technology closes that gap.” That question takes a little longer to answer. It also tends to lead to fixes that hold up.

That same hiring reflex can also mask a retention problem with the people already on staff, a theme Why Good Employees Leave Medical Practices, and How to Fix It explores in more depth.

FAQ

Do I need to hire another employee for my medical practice?

Not automatically. Start by identifying exactly what work is not getting done and why before deciding a new hire is the fix. Workflow redesign, delegation, technology, or specialized outsourced support like a medical scribe may solve the same problem more directly.

How can a medical practice increase capacity without hiring?

Map where time is actually going, identify work that does not require a licensed provider, and look for workflow friction or documentation burden that technology or delegation could address before adding headcount.

How do healthcare practices manage staffing shortages?

By treating the shortage as a capacity question first: what work isn't getting done, why, and who or what is actually the right fix, rather than defaulting straight to a new hire.

What's the difference between a staffing problem and a capacity problem?

A staffing problem means a role genuinely needs a dedicated person. A capacity problem means the work isn't getting done for reasons that might include poor workflow, misallocated tasks, or documentation burden, none of which a new hire automatically fixes.

When does hiring actually make sense for a medical practice?

When the work is genuinely growing in volume, not just poorly distributed, and the role requires ongoing, dedicated capacity, for example a growing practice adding a location or a specialty practice scaling patient volume.

What should I check before posting a new job requisition?

Look for duplicate documentation, unclear handoffs, or manual work that could be redistributed or supported with the right tools. If the role has absorbed tasks that don't belong together, fix that first.

Is a staffing agency the same thing as Clinical Informatics?

No. A staffing agency is built to answer who can fill a seat. Clinical Informatics starts with a different question: where is capacity actually being lost, and what combination of people, workflow, and technology fits the work.

What is workflow redesign, and how does it reduce the need to hire?

It means fixing the process behind a task, removing duplicate entry, unnecessary approvals, or unclear handoffs, so the same work takes less time and fewer people. It's usually the lowest-cost fix and compounds over time.

How can I tell if a role in my practice was poorly designed?

Signs include a new hire in the role struggling within months, staff doing work well outside their training or license level, and the same complaints resurfacing after previous hires.

Can technology reduce how often a practice needs to hire?

In specific cases. Technology works best on repetitive, well-defined tasks like documentation or scheduling, and only when paired with a clear plan for who reviews the output.