Medical Practice KPIs: 12 Numbers Worth Watching Every Week
Medical practice KPIs that show where money, calls and patients get stuck: 12 measures, how each one is worked out, and how to read it week to week.
The short answer
- 01Twelve numbers worth a weekly look across billing, the phones and remote monitoring
- 02Compare your practice against its own earlier weeks, not a generic target
- 03Read how a KPI is worked out before you act on a change
Medical practice KPIs are the few numbers that tell you, before the month closes, where money, patients and work are getting stuck. This list covers 12 worth a look every week: what each one measures, how it is usually worked out, and what a change in it tends to mean. There are no target values here on purpose. The right number depends on your specialty, your payers and how you define the measure, so the comparison that matters is your practice against its own earlier weeks.
How to read a KPI before you act on it
A number on a dashboard is only useful if everyone reads it the same way. Five habits prevent the usual misreadings.
- Know whether it is a flow or a snapshot. A flow counts what happened in a period (claims sent this week). A snapshot says where things stand right now (claims waiting today). Widen the date range and a flow grows; a snapshot does not.
- Compare like with like. Put this week beside last week, or this month beside the same length of time before it. A 30-day count will beat a 7-day count without anything having improved.
- Look at the count behind every rate. One extra denial out of five claims moves a rate a long way. Read the rate and the raw count together.
- Write each definition down once. Clean claim rate, for example, is calculated differently from one billing team to the next. Pick one definition and keep it, or the trend means nothing.
- Up is not always good. A rising collection rate is welcome; a rising denial rate is not. Read the direction with the meaning of the measure.
Revenue cycle KPIs
These seven show whether the work your providers do turns into money, and how fast.
1. Days in accounts receivable
Days in A/R is the average number of days it takes to collect what you bill. A common way to work it out: take your total outstanding A/R and divide it by your average daily charges over a recent period, such as the last three months. When it rises, money is arriving more slowly. Split it by payer to see whether one payer is the cause or the whole process has slowed.
2. Accounts receivable by age
Group what you are owed by how long it has been outstanding: 0 to 30 days, 31 to 60, 61 to 90, 91 to 120 and over 120. The total matters less than the movement between groups. Balances sliding into the older groups are harder to collect, and some will pass a payer's timely filing limit, which differs by payer and contract.
3. Clean claim rate
The clean claim rate is the share of claims accepted on the first submission with no edits, rejections or requests for more information. It can be measured at the billing partner's front-end edits or at the payer. Either works if you stay consistent. A drop usually points to a data problem upstream: a wrong member ID, a missing modifier, a provider detail that does not match. Our guide to claim rejections and denials explains where each kind of failure happens.
4. First-pass resolution rate
First-pass resolution is the share of claims paid, or otherwise finally settled, on the first submission. It is not the same as the clean claim rate: a claim can pass every edit and still be denied. Every claim that misses first pass costs staff time to work again, so a falling first-pass rate shows up as a busier billing team before it shows up as lost money.
5. Denial rate and top denial reasons
Denial rate is the number of claims denied divided by the number submitted in the period. You can also track it by charge value, since a few large denials can matter more than many small ones. The rate alone tells you something is wrong; the reasons tell you what. Group denials by reason code each week and work the top few. Our list of habits that reduce claim denials covers the usual fixes.
6. Net collection rate
Net collection rate compares what you collected with what you were entitled to collect: payments divided by charges minus contractual adjustments. Gross collection rate (payments divided by full charges) mostly reflects your fee schedule, so the net rate is the one to watch. When it falls, money is being left behind: write-offs that should not have happened, denials nobody appealed, or patient balances nobody followed up.
7. Signed encounters not yet billed
Count the encounters waiting on a provider signature, and the signed ones that have not become a claim yet, with the age of the oldest. The related measure is charge lag: the days from the date of service to the day the claim is sent. A growing backlog here delays everything downstream, and it is often fixed by a reminder rather than a process change.
Front desk and phone KPIs
8. Eligibility issues open
This is the number of patients whose coverage check came back with a problem (coverage inactive, plan changed, member not found, details that do not match) and is still unresolved. Track the count and the age of the oldest one. An issue found before the encounter is a phone call; found after the denial, it is rework and sometimes a write-off.
9. Calls answered and calls missed
Compare inbound calls answered with calls missed, and break both down by hour of day. Add the callbacks still waiting and how long they have waited. The hourly view shows whether a missed-call problem is a staffing gap at certain hours rather than a general one. Our phone workflow for practices goes through the fixes.
Remote monitoring and device KPIs
10. RPM patients reaching the required reading days
Medicare pays for remote monitoring device supply based on how many days a patient sends readings in a 30-day period. For years the device supply code has required 16 days, and CMS added a code for a shorter range of days starting in 2026. Rules change each year and other payers set their own, so confirm the current requirements. The KPI is the number of patients on track for the period, the number at risk of falling short while there are still days left to act, and the number with no reading at all.
11. Time to first response on urgent readings
When a reading falls outside the limits the provider set for a patient, how long does it take your team to act on it? Use the median time from the reading arriving to the first action, and look separately at the slow ones. A good median can hide a handful of readings that waited overnight. What counts as urgent, and how fast the practice responds, are clinical decisions for the provider and your protocols, not for a dashboard.
12. Shipments with issues
Count the device shipments with an open problem: a delivery exception, a wrong address, a package returned to sender or lost. Beside it, count devices delivered but not yet active. Both mean a patient who is enrolled but not monitored. Our guide to RPM device logistics covers how to keep both lists short.
Setting up a weekly KPI review
The numbers help only if someone looks at them on a fixed day and acts.
- Give each number one owner, the person who explains it and fixes what it shows.
- Review on the same day each week, on one page, in the same order.
- For each number, note the change from last week and one action, or "no action".
- When a number moves, open the rows behind it before deciding what caused it.
- Revisit the definitions every quarter, and when a payer or process changes.
If your practice books appointments, no-show rate and appointment lead time belong on the same page; they come from your scheduling system.
Frequently asked questions
What is the difference between a KPI and a metric?
Every KPI is a metric, but only a few metrics are KPIs. A KPI is a number you have chosen to watch because a change in it calls for action, and it has an owner. Everything else is detail you open when a KPI moves.
How many KPIs should a practice track?
Few enough that one person can review them in a short weekly meeting. A dozen is plenty for a practice with billing, phones and a monitoring program. If a number never leads to an action, drop it.
Should we compare our KPIs with published benchmarks?
Carefully. A benchmark is only comparable if it uses the same definition, the same kind of practice and the same payers, and that is rarely clear. Your own trend over several months is the more reliable guide.