Beyond Volume: The Revenue Cycle Metrics Physician Group Leaders Should Watch Every Month

Many physician group leaders judge the month the way their bank account reports it: cash came in, cash went out, and this month was lighter than last. For physicians who built a group around clinical work, that’s often the whole scorecard. But to grow the group, you need to see what’s behind that number.

When cash dips, the drop is real. What matters is whether it’s expected and explainable, or a sign of a problem. To find out, I walk up a short ladder of questions, from the simplest to the most revealing. For each one, a leader needs to know four things: where the number comes from and whether it can be trusted, how it’s measured, what good looks like, and what to do when it’s off. That last one gets skipped most often. Most of these metrics end up pointing at a process problem or a provider-level problem, and each has a likely first place to look.

Step 1: Did the census fall?

Before looking for billing problems, check whether there was less work to bill. Compare the census at each hospital, or the visits on the schedule, with last month and with your trailing three-month average. Count by date of service rather than date posted, so a slow billing week doesn’t look like a slow clinical week. A seasonal dip, a lost facility contract or a provider on leave can lower volume without anything being wrong in the revenue cycle.

Many physician groups don’t own that census data. It sits in the hospital’s system, and getting a feed means waiting in the hospital IT team’s project queue. At sites with no feed, the group depends on providers to register every patient.

For sites that won’t send a feed, the face sheet can do the registration. At skilled nursing facilities, smaller hospitals and other sites without an interface, providers still get handed a face sheet or see one on a screen. 

Face Sheet Auto-Registration, coming in the November 16, 2026 Charge Pro release, allows you to quickly take a photo of the sheet, paper or on-screen, and the demographics populate a new registration automatically: name, date of birth, room and bed, and insurance. Before creating a new patient, the app checks for an existing match. From there the provider goes straight to the charge screen and their code shortcuts.

Providers at these sites work from their phones. A physician at an infectious disease group said:

“It has to be mobile, it has to follow you, it has to be there when you’re doing the charge, when you’re doing the note. This has to be right there and it has to be fast and easy.”

Two months later, after seeing the new workflow, he said: “It reads the face sheets. That’s better than what we have now.”

The back office gains as much as the provider. Each registration the app creates is an encounter the billing team knows about, so the group has a census to compare against even at sites that never send a feed. A revenue cycle executive at a national physician billing company described the registration side this way: “Even if it’s only half the completed record, it’s still better than doing it from scratch, plus it eliminates potential error.”

Where it comes from: the hospital census or ADT feed, or your group’s schedules, assignments and face sheet registrations where there’s no feed. A census counts patients in beds, not patients your group saw, so use your assignment list where you have one.

When it’s off: if the census falls, lower cash may simply reflect less work, and the conversation turns to staffing and growth rather than billing. If the census held and cash still dropped, the work was done and the gap is somewhere between the encounter and the bank. Move to Step 2.

Step 2: Did we capture every encounter, and how quickly?

The next question is whether every encounter became a charge, and how quickly. The first half of that is charge reconciliation: comparing the encounters you expected, from census, schedules, assignments, signed notes or face sheet registrations, against the charges actually received for the same dates of service, every day, by provider and by site. A group-wide total hides the gaps, and checking daily surfaces them while there is still time to bill. Reconciliation is only as complete as the list of expected encounters behind it.

The second half is charge lag, the time from the date of service to the charge being entered. HFMA’s benchmark is charges captured within three to five days of service, with late charges under 2% of the total. Measure it by provider and by site, and adjust for any intentional hold periods so the number isn’t misread. When one provider is an outlier, the cause is often end-of-day or end-of-week batching. Reviewing the day’s full patient list at wrap-up helps providers remember the service they almost forgot to bill.

A multi-state health system on Oracle Health uses Charge Pro specifically for reconciliation, because its EHR gave it no reliable way to see whether an encounter had been billed, or whether a physician had acted on a coder’s query.

Where it comes from: expected encounters come from census, schedules, assignments, signed notes or face sheet registrations. Charges and charge-entry dates come from your charge capture system. Good looks like no unexplained gaps.

When it’s off: if billed encounters fall short of the census or schedule, the group saw patients whose encounters never became charges. If one provider is consistently behind or has a long lag, look at when that provider enters charges. If one site is short, check whether its census feed or registrations are complete. If reconciliation is complete and every charge was captured and submitted to payers, move to Step 3.

Step 3: What are we collecting per wRVU, and where does it miss target?

Collections per work RVU is my favorite revenue cycle metric, and I look at it every month, even when the cash coming in looks great. It divides what the group actually collected by the work RVUs behind it, so it ties cash to the work providers did in a single number. I start with the group overall and its trend. If it’s low or trending down, I get more granular: by site, by provider type, such as physicians and advanced practice providers, by provider and by payer type. What I’m looking for is the unexpected number, the site, provider or payer whose dollars per wRVU doesn’t meet target. Every group should have a dollars-per-wRVU target based on its own payer mix.

Match the time periods carefully. Collections trail the work by a month or two, so compare collections with the wRVUs from the dates of service they paid for, or use a trailing three-month view. Payer mix affects timing too: MGMA notes that traditional Medicare typically pays a clean claim in 10 to 14 days, while Medicare Advantage plans take 30 to 45.

Where it comes from: collections by date of service come from your billing system. wRVUs come from the CMS fee schedule values applied to the CPT codes you billed, so use the same fee schedule year across periods. Good means meeting your payer-mix-based target, with a steady or rising trend.

When it’s off: if it’s low at one site, look at that site’s payer mix and denials. If it’s low for one provider, look at that provider’s denials and unpaid claims. If it’s low for one provider type, check how shared visits are billed and how payers pay advanced practice providers. If it’s low for one payer type, compare what that payer paid against your contracted rates and look at its denial pattern. If it’s falling for the whole group, look at payer mix, denial trends and contract rates.

Step 4: Are we billing at the level the work supports?

Collections per wRVU has one blind spot. When a provider codes too low, wRVUs and collections drop together, so dollars per wRVU can look normal. That’s why I also track average work RVUs per encounter and CPT distribution by provider, month over month. In my experience, many providers choose a moderate E/M level because it feels safe, even when the documentation supports more. With patient acuity rising, hospital-based groups are more likely to be under- than overbilling.

Read these numbers carefully. A provider below peers may have a different patient mix or schedule, so compare providers with similar patients. A provider well above peers is a reason for a compliance review.

Where it comes from: billed CPT codes come from your charge data. CMS publishes Medicare billing data by provider and specialty, which gives you an outside comparison for E/M level distribution alongside your own peers.

When it’s off: review the level distribution and a sample of documentation for the outliers, and start with education and documentation feedback. The next post in this series goes deeper on CPT distribution and wRVU trends for hospitalists.

Step 5: What the other revenue cycle metrics are telling you

These four numbers explain most of what Steps 1 through 4 leave open. Each has a published benchmark, and each needs a check on how it’s measured before you act on it.

1. Net collection rate. The share of collectible dollars you actually collected. HFMA puts the minimum at 95%, with 97% to 99% as optimal.
2. Days in accounts receivable. HFMA’s benchmark is 30 to 40 days, with less than 10% of AR older than 90 days.
3. Initial denial rate. HFMA puts the industry average at 5% to 10%, with less than 5% as optimal. In an MGMA poll, 60% of medical group leaders said their denial rates had risen compared with the year before. In Experian Health’s 2025 State of Claims survey, 41% of providers said more than 10% of their claims are denied, up from 30% in 2022, and 54% said claim errors are increasing.
4. Clean claims rate. HFMA’s target is 98%.

Where it comes from: all four numbers come from your billing system, clearinghouse or RCM partner. Ask how they calculate net collection rate, because how contractual adjustments are counted varies, and whether “initial denial rate” counts claims or dollars, since that changes the number.

When it’s off: if the net collection rate drops, look at denials and write-offs first. If days in AR rise, break them out by payer and by age before drawing conclusions. A shift toward slower-paying plans raises days in AR even when nothing in your process changed, while a growing share older than 90 days usually means claims aren’t being followed up. If the denial rate rises, sort denials by reason. Registration and eligibility issues alone account for nearly 27% of denials, according to MGMA, and those are fixed at intake, not on appeal. If the clean claims rate is below target, look at which edits claims are failing and whether those edits can run before charges are released to billing. Most of these problems cost far less to prevent on the front end than to fix after the claim comes back, which is the focus of the first post in this series [LINK: Blog 1, “You Captured the Charge. Did You Capture the Revenue?”].

Step 6: Are we keeping up with coding and Medicare rule changes?

Coding rules change constantly. CPT codes are updated every January, ICD-10-CM codes every October, and CMS updates its NCCI edits every quarter, on top of payer policy changes throughout the year. A change that isn’t built into your workflow and taught to your providers shows up later as denials or lower payment. Medicare usually changes the rules first, and commercial payers tend to follow.

The Medicare program that reaches a physician group’s own revenue most directly is the Quality Payment Program, specifically MIPS, the Merit-based Incentive Payment System. MIPS adjusts your Medicare Part B payments by as much as 9% in either direction, two years after the performance year, so a weak year turns into a payment penalty long after the work is done. If you’re finding out where you stand months later, you’re managing blind.

Where it comes from: code set updates from the AMA and CMS, quarterly NCCI edit updates from CMS, your payers’ policy bulletins, and your MIPS feedback reports in the QPP portal.

When it’s off: if denials jump in January or October, check whether the new code set was loaded into your systems and whether providers were trained on the changes. If your MIPS feedback shows you trending toward a penalty, adjust the measures you report while there’s still time in the performance year.

I’ll cover what’s changing in the Quality Payment Program for hospitalist groups, and what their leaders should do about it, in a webinar on October 14. 

Watch our On-Demand Webinar: MIPS Is Changing: What Hospitalist Group Leaders Need to Know Now

Starting a group? Set up these six numbers on day one

Habits and systems set in quickly. If you’re building a group, start tracking these from the first month:

1. Census or scheduled volume, by site
2. Encounters billed versus expected, by provider and by site
3. Charge lag, by provider
4. Collections per wRVU, overall and by provider
5. Work RVUs per encounter, by provider
6. Initial denial rate, sorted by reason

Each one gives you a baseline to compare next month against.

Where the numbers live, and which ones Charge Pro reports

For most groups, these numbers sit in different systems, and pulling them together takes work. The owner of a cardiology group described what that looks like day to day: “The [heart caths] are entered separately, the visits are entered by a separate person, and then it has to be recompiled. It takes several hours of work for an employee to get it done on a daily basis.”

medaptus Charge Pro covers the charge side of the ladder. Its reconciliation view shows expected, charged and uncharged encounters by provider, location and group. It reports charge lag by provider and group, including whether a charge is waiting on the provider or the coder, along with charges on hold by reason, how many times coders touch an encounter before it goes out, and CPT distribution and work RVUs per encounter by provider. Collections, days in AR and payer denials come from your billing system or RCM partner after the claim leaves Charge Pro.

Where to start

Pull last month’s census, billed encounters, charge lag, collections and wRVUs for your largest site, and walk up the ladder once. The step where the numbers stop lining up is where to spend next month’s attention.

See how Charge Pro supports physician groups across every facility where your providers work.