Days in A/R: Formula, Benchmark, and the Four Levers

Days in A/R = Total Accounts Receivable / Average Daily Charges. Average Daily Charges = Total Charges over the trailing measurement period (typically 90 days) / Number of days in that period (90). The reason for using a trailing 90-day window for the denominator rather than a single month is to smooth the month-over-month variance from holiday weeks, seasonal volume, and payer cycle effects. A practice with $900,000 in trailing-90-day charges has $10,000/day in average daily charges; if total A/R is $350,000, days in A/R = 35. This is the standard formulation of the metric. HFMA MAP Keys publishes the definitions and equations for revenue cycle KPIs, and MGMA collects days in A/R in its annual Cost and Revenue Survey; note that neither publishes a free target value — MGMA's medians sit inside licensed DataDive products. For a freely published comparator, the AAFP states days in A/R should be "below 50 days at minimum; however, 30 to 40 days is preferable."

The formula divides billed receivables by average daily charges, so it is blind to money that has not become a claim yet and to money that has stopped being one — because the numerator counts open, billed balances and the denominator moves whenever charge volume moves. Three boundaries decide what your number actually means, and the two freely-readable methods each take a position on all three.

Do credit balances belong in the numerator?

No. Both published methods remove them. AAFP's calculation subtracts credits received from charges before dividing, and its own guidance is to subtract credits from receivables "to avoid a false, overly positive impression of your practice." HFMA's MAP Keys define the net days in A/R numerator as the net patient receivable "net of credit balances," and give credit balances a separate key rather than folding them into the A/R measure. Overpayments and unapplied credits left sitting in the numerator net your A/R down and make the practice look faster than it is — which is why a days in A/R number that improves in a month with heavy recoupment activity is usually an artifact, not a result.

Are unbilled charges in the numerator at all?

No, and this is the boundary that most changes what your number means. HFMA's aged A/R keys measure against billed A/R and count only active, open billed accounts, so a charge that has been posted but not yet submitted sits outside the metric entirely. The consequence is counter-intuitive: a charge-entry or coding backlog makes days in A/R look better, because the delayed charges are neither in the numerator nor collectable yet. Track charge lag as its own number alongside this one — days in A/R cannot see it, and a practice reading only this metric will mistake a billing backlog for a collections win.

What happens to your number in a month with unusually low charges?

It rises, without anything about your collections having changed. Average daily charges is the denominator, so a quiet month, a closed week, or a provider on leave shrinks the divisor and inflates the result. That is the argument for a trailing window rather than a single month: AAFP's steps allow the period to be three, six or twelve months, and this page's formula uses a trailing 90 days for the same reason. Two further movements have the same character — accounts sent to a collection agency are written off current receivables, and patients moved onto payment plans extend the collection horizon by agreement. AAFP's guidance on both is to calculate the metric with and without them rather than to pick one treatment silently. Whichever convention you adopt, hold it constant; a mid-year change to the numerator makes your own trend line, the most reliable comparator you have, unreadable.

For a physician practice, the published benchmark is AAFP's: days in A/R should stay below 50 days at minimum, with 30 to 40 days preferable. That is a practice-management guidance range, not a measured median — AAFP publishes no sample or denominator behind it.

What no free source publishes. There is no freely-available observed median days in A/R for physician practices. MGMA collects the figure in its Cost and Revenue Survey, but the medians sit inside licensed DataDive products. HFMA MAP Keys publishes the equation for net days in A/R, not a target value. Figures circulating as "the MGMA benchmark" of 35 or 47 days could not be traced to any MGMA page, and specialty-level days in A/R breakdowns could not be traced to any authoritative source at all — so we do not publish them.

Reading your own number. Because the benchmark is a range rather than a per-specialty median, interpret your figure against your own trend and payer mix rather than against a specialty average you cannot source. A number that is rising quarter over quarter is a more reliable signal than a number that sits above or below someone else's published range. Specialties carrying heavy pre-authorization or case-rate adjudication — surgical lines, DME, home health — structurally collect more slowly than primary care, but no free source quantifies that gap credibly.

Days in A/R is a single rolled-up number; the aging bucket distribution under it reveals the underlying health. Standard buckets are 0-30, 31-60, 61-90, 91-120, and 121+ days. A healthy practice keeps roughly 65-75% of total A/R in 0-30 days, 12-18% in 31-60, 6-10% in 61-90, and under 12-15% combined in 91+.

Collectability in the 90+ bucket runs around 50% and keeps falling the longer balances age. A practice with the same 38-day average but 25% of A/R sitting over 90 days has a denial rework backlog masked by recent charge volume — the average looks fine because new claims are paying, but the aged tail stays unrecovered without dedicated accounts receivable follow-up on the 90+ bucket. Conversely, a practice with 44 days but only 8% over 90 has a slower payer mix (more Medicaid, more secondary COB) but no actual collection failure.

Lever 1: Clean claim rate. Every percentage point of CCR improvement removes about 0.3-0.5 days from the average because rework adds 30-60 days to a touched claim's collection cycle. Lever 2: Days from charge to submission. The HFMA target is under 5 days; many practices run 8-12. Reducing charge-to-submission lag from 10 to 4 days takes 6 days off Days in A/R directly. Lever 3: First-pass appeal velocity on denials. Denials over 60 days old recover at half the rate of denials worked within 14 days — the operating argument for dedicated denial management services. Lever 4: Self-pay collection process. Self-pay balances over 90 days drag the average up sharply because the underlying recovery rate is in the 18-22% range without active collection. A patient statement cycle issued on day 5 post-EOB with a 30-day follow-up cadence reduces self-pay aging materially.

A fifth lever sits on the receipt side rather than the claim side: EFT enrollment as a days-in-A/R lever removes the mail float and manual reconciliation that every payer still paying by paper check adds to the cycle.

Average payment turnaround varies by payer class, and that variation sets the floor on achievable Days in A/R. The one figure that is authoritative here is Medicare's: CMS may not pay a clean electronic claim before the 14-day payment floor. Commercial and Medicaid turnaround figures circulate widely but could not be traced to a primary source, so we do not publish per-payer averages — measure your own turnaround by payer from your remittance data instead, which is both sourceable and actionable. Payer-specific filing deadlines, which do materially affect A/R, are in our timely filing database.

Pattern 1: A/R is rising, charges are flat. This indicates a denial backlog or appeal queue not being worked. Compare 90+ day A/R as a percentage of total A/R against the prior period — a rising 90+ percentage with flat charges is a denial recovery failure. Pattern 2: A/R is rising in proportion with charges. This is volume growth, not a process problem; the absolute dollar number rises but the days metric stays flat. Pattern 3: Days in A/R drops sharply month-over-month. Usually indicates either large lump-sum capitation or risk payment, large recoupment offset, or large write-off cleanup — none of which reflect operational improvement. Pattern 4: Days in A/R rises after a payer policy change (e.g., new prior authorization requirement). Drill into denial codes — a CARC 197 spike confirms the cause.

Three common errors distort reported Days in A/R. First, using a 30-day denominator rather than a 90-day rolling window — this makes the metric whipsaw on monthly volume changes and produces meaningless month-over-month comparisons. Second, including credit balances (negative A/R from overpayments awaiting refund or recoupment) in the numerator without netting them out — credit balances are not collectable A/R and including them understates the true number. Third, using gross charges rather than net charges as the denominator. The HFMA-aligned method uses gross charges in both numerator and denominator (so the ratio is consistent), but some systems mix gross charges in the denominator and net A/R in the numerator, distorting the result by 5-15%.

A days-in-A/R number tells you a problem exists. It does not tell you which repair to buy. Practices that stall here usually stall because they treat a staffing problem as a process problem, or the reverse.

Three questions separate the two, and they are worth answering before you price anything:

1. Is the cause procedural or structural? If claims are going out clean and payers are simply slow, that is procedural and an internal fix is realistic. If the same denial reason recurs monthly, or follow-up stops whenever one person is out, the cause is structural and more effort from the same team will not move it.
2. Do you have coverage? In most small practices one person owns charge entry, submission, posting and follow-up. When that person takes leave, the cycle stops and the timely-filing clocks keep running. A single point of failure is not a workflow you can optimise your way out of.
3. Is the backlog aging faster than you can work it? This is the question with a deadline attached. Any balance that reaches 75% of its timely-filing limit is one that will be written off if it is not worked now — and unlike a slow payment, a missed filing deadline is unrecoverable. If the over-90 bucket is growing month over month while the team is already at capacity, the decision has effectively been made for you.

What each path costs. Internal billing costs a salary plus benefits, software, clearinghouse fees, training and the turnover risk that comes with a single-biller operation. Outsourcing to MedPrecision is a published percentage of what we actually collect — 7.0% for solo practices and 6.0% for groups, with no setup fee, no per-claim charge and no long-term contract. Neither is automatically cheaper; the comparison only becomes real once you load every internal cost, which is what the billing cost calculator is for.

Where to take the decision next. If you want the structural comparison, outsourced vs in-house medical billing works the all-in math both ways and names the cases where in-house still wins. If you have already decided and the worry is the transition, how to switch billing companies without revenue loss covers what happens to the existing A/R backlog during a handover — the part practices most often get wrong. And if the immediate problem is the aged balance rather than the operating model, A/R follow-up services is the engagement that works the backlog directly, sequenced by filing deadline rather than by age.