On this page
- Step 1: Set the scope
- Step 2: Gather the data sources
- Step 3: Map the patient-journey stages
- Step 4: Define what should happen, and by when
- Step 5: Pull what actually happened
- Step 6: Reconcile the two
- Step 7: Classify every open item
- Step 8: Validate with a clinician and a scheduler
- Step 9: Prioritize the work
- Step 10: Measure against a frozen baseline
- What should the assessment deliver?
Key takeaways
- A revenue leakage assessment reconciles expected events with actual events for the same service and counts what stayed open past its window.
- Narrow scope beats broad scope: one service line, two years of history and a handful of journey stages produce a result people trust.
- Classification and clinician validation decide whether staff believe the list, so do them before putting a value on anything.
- Prioritize by estimated value, urgency and confidence, then measure verified completion against a fixed baseline.
A healthcare revenue leakage assessment compares the care your practice expected to deliver with the care it can show was completed, then sizes and ranks the gap. You define scope, gather data, list the expected events at each stage of the patient journey, match them to actual events, classify and validate what stays open, then prioritize and measure. Done well, it produces a short list of patients and a baseline rather than a single headline number.
Some teams call this a revenue leakage audit. The steps below work with a spreadsheet, your EHR's reporting tools or dedicated software. If you want a rough sizing before you start, the revenue leakage calculator gives a first estimate from a few inputs. For background on the concept itself, see what healthcare revenue leakage is.
- ScopeChoose service lines, locations, providers and the time period.
- Data sourcesGather referrals, orders, recalls, appointments, encounters and optional claims.
- Expected eventsDefine what should happen at each stage, and by when.
- ReconcileMatch expected events to actual events for the same service.
- Classify and validateLabel each open item and have clinicians spot-check a sample.
- PrioritizeRank by estimated value, urgency and confidence.
- MeasureTrack verified completion against a frozen baseline.
An assessment reconciles what should have happened with what did, then turns the gap into a ranked, measurable list.
Step 1: Set the scope
Set scope narrowly enough to finish and broadly enough to matter. A good first assessment covers one or two service lines, every location that delivers them and about two years of history.
Write down:
- Service lines. For example, GI procedures and surveillance, or cardiac imaging and monitoring.
- Locations and providers. Include all sites, because variation between them is often the most useful finding. For a PE-backed platform, this is also where acquired practices with different habits show up; measuring revenue leakage in PE-backed practices covers the portfolio version.
- Time period. Two years lets you compare each month with the same month a year earlier.
- Owner. One person, usually the COO or a practice administrator, accountable for finishing.
Step 2: Gather the data sources
You need records that show what was intended and records that show what happened. Most already exist in the EHR and practice management system.
| Source | What it shows | Common gap |
|---|---|---|
| Referral orders and fax inbox | Who was sent and when | Faxes not filed to the chart or a queue |
| Procedure and diagnostic orders | What providers asked for | Orders left open after care elsewhere |
| Recalls and care-plan intervals | Who is due and when | Intervals written only in the note |
| Appointments | Booked, canceled and missed visits | Local appointment types that need mapping |
| Encounters and procedures | What was completed, where and by whom | Outside procedures not recorded |
| Claims and HIE feeds (optional) | Billing or outside confirmation of completed care | Not always linked to operational data |
Note every known gap now. It will explain odd results later.
Step 3: Map the patient-journey stages
Map the stages your scoped services move through: referral, visit, procedure, follow-up and longitudinal care. For each stage, write down the event that starts it and the team that owns the handoff. A GI service line might run from an inbound referral to a consult, a colonoscopy order, the procedure itself and a surveillance recall.
Step 4: Define what should happen, and by when
For every stage, write down the step that should follow the trigger and the window it should happen in. Your practice fills in the numbers; clinicians approve any window that touches clinical timing:
- Inbound referral, then first contact within [X] days and a completed visit within [Y] days.
- Procedure order, then the procedure inside the window set for that test.
- Canceled procedure, then a new date within [Z] days for that procedure type.
- Recall due date, then a completed visit or study for the same service.
Clinical intervals depend on findings and the practice's approved guidance, so do not copy windows from another group. The article on expected vs. actual events in healthcare explains this model in more depth.
Step 5: Pull what actually happened
Actual events are what the records show happened: appointments booked, visits completed, procedures performed, results received and care documented as done elsewhere. Pull them for the same period plus a trailing buffer, so late completions still count.
Step 6: Reconcile the two
Match each expected event to an actual event for the same service, inside the window. This is where most assessments go wrong. A patient with an open order for a sleep study who comes in for a medication check has had a visit, not the study. Match on service, not on any appointment, and map local appointment types and codes first.
Step 7: Classify every open item
Every unmatched item needs a label, because not every gap is leakage. A simple scheme:
| Class | Meaning | Action |
|---|---|---|
| Open opportunity | Expected event missing, patient reachable | Goes to the work queue |
| In progress | Booked inside the window | Monitor; reopen if canceled |
| Completed elsewhere | Care happened at another facility | Close with a reason; useful for network reporting |
| Excluded | Clinically canceled, deceased or moved away | Close with a reason |
| Data issue | Mapping or filing problem | Fix the rule, not the patient |
Step 8: Validate with a clinician and a scheduler
Have a clinician and a scheduler review a sample of flagged patients from each stage. If the sample includes patients who should not be there, fix the rule and rerun before anyone makes a call. This is the step most teams skip, and it decides whether staff trust the list.
Step 9: Prioritize the work
Rank open opportunities by estimated value, urgency and confidence. Value uses your own conservative reimbursement per completed service. Urgency reflects how far past the window the patient is and how time-sensitive the service is. Confidence reflects how clean the match was. When a patient has several open items, combine them into one row with one next action; that is the idea behind patient opportunity intelligence, and how Nuvory works shows the same ranking run every night.
Illustrative example. An assessment finds 200 open procedure orders and 600 overdue recalls. Assuming round values of $1,500 per procedure and $150 per follow-up visit, the estimated value at risk is (200 × $1,500) + (600 × $150) = $390,000. These inputs are hypothetical. The point is the method: the procedure orders carry most of the value, so they lead the queue even though the recalls are three times as numerous, unless a clinician marks some recalls as time-sensitive.
Step 10: Measure against a frozen baseline
Freeze the baseline, then report monthly. Count completed care after detection, not bookings or call attempts, and report it as correlation with outreach rather than proof of cause. Mark every rule change on trend charts. The guide to revenue leakage metrics and KPIs covers which ratios to track.
Revenue leakage assessment checklist
- Scope written down: service lines, locations, providers, time period and owner.
- Data sources listed, with known gaps such as fax referrals or outside procedures noted.
- Journey stages mapped from referral to longitudinal care for the services in scope.
- Expected events and completion windows defined and signed off by a clinician.
- Actual events pulled and matched on the same service, not just any appointment.
- Every open item classified, with exclusions recorded and a reason attached.
- A sample of flagged patients reviewed by a clinician and a scheduler.
- Open opportunities valued conservatively and ranked by value, urgency and confidence.
- Baseline frozen and monthly measurement of completed care agreed.
What should the assessment deliver?
A finished assessment should hand leadership three things, each one short enough to read in a meeting:
| Deliverable | What it contains | Who uses it |
|---|---|---|
| Baseline | Conversion and open counts by stage, location and month, with definitions attached | COO, finance, service-line leaders |
| Open list | Validated patients with the reason each is open, an owner and a rank | Schedulers, referral coordinators, practice administrators |
| Value range | A low and a high estimate of value at risk, with the assumptions written out | Leadership, owners, investors |
A one-time assessment shows the size and shape of the problem. Keeping the gap closed means running the same reconciliation every day, flagging items as windows pass and closing them as care is completed. Teams that do not want to rebuild the spreadsheet monthly usually move to healthcare revenue leakage software at this point, and how to choose revenue leakage software covers the evaluation; the rest of the revenue leakage series covers what to watch once the program is running. If you'd rather run the assessment on your own data with software doing the reconciliation, a 30-day revenue leakage pilot produces the baseline and a validated list of opportunities before you commit to a subscription.
Frequently asked questions
How long does a revenue leakage assessment take?
It depends on data access and scope. A narrow assessment of one service line with clean EHR data moves much faster than a multi-site, multi-EHR review. The slowest steps are usually mapping local appointment types and getting clinician sign-off on windows, so start those early.
Who should be involved in the assessment?
An operations owner such as the COO or practice administrator, someone who can pull and join data, a clinician to approve windows and review samples, and schedulers or referral coordinators who know how the workflows actually run. Revenue cycle can supply service values and claims data.
Do we need claims data to run an assessment?
No. Referrals, orders, recalls, appointments and encounters are enough to start. Claims and HIE feeds, where available, make completion matching more precise and help identify care that happened at another facility.
What is the difference between an assessment and a calculator estimate?
A calculator applies averages to a few inputs to give a rough order of magnitude. An assessment works patient by patient from your own records, validates the results with clinicians and produces a list that staff can act on.