Five billing census checks.
These are the five ways a manual billing census comes apart, found in a real five-center network. They are the detail behind leak 1, care delivered but census incomplete. You can check four of them yourself this week, with one facility and one month.
Three records that have to agree.
A service is delivered. It is documented. It is billed. Revenue is correct only when all three say the same thing.
Revenue is correct. There is nothing to find.
Money leaves, and no alarm goes off.
Nothing in a manual process forces them to agree. Attendance lives in Kipu, Sunwave, or a paper log. Money lives in the census workbook. A person carries it across. One person compares a daily log against a chart against a spreadsheet, hundreds of rows, once a week, while doing four other jobs. These five are not a list of mistakes. They are the five ways those three records come apart.
Each check, and how to run it.
One service, from the room it happened in to the money arriving. Each error branches off at the exact point it happens.
Identity mismatches
The log has a name but no record number, so the service is matched to the wrong episode, or to a client discharged last month. Wrong client. The hardest one to defend.
Open the attendance log template. Look for an MRN column. Adding it lets every check match by record number. At the network in the case study, every log matched once it was added.
Unbilled attendance
Delivered and documented, but it never gets entered into the census workbook. No claim is ever made, so nothing downstream triggers a warning. Full rate, gone.
Reconcile facility log mark counts against billed entries for one completed month.
Billing above documentation
A line the chart cannot support. It bills cleanly and no denial ever fires. Returns later as a clawback rather than a denial.
Pull twenty random billed lines. Trace every one to a signed clinical note.
Threshold days
An IOP day under 3 hours, a PHP day under 5. In the census sheet it looks identical to a full day. At the network we studied, its rate table pays a short day at roughly a tenth of full rate, while the day cost the same to deliver. Your own contracts set your figure.
Sort last month by hours delivered. Read the bottom of the list, then read the names on it.
Level-of-care conflicts
A step-up recorded in the clinical record, with an authorization that never caught up. Often denied months later, when it is hard to fix.
Check every client whose LOC changed. Verify the effective date in all three systems.
The correction window
Every error above is findable. What decides the money is when you find it. How often errors happen matters less than when you find them.
Same week, it is a correction. Ninety days in, it is often a denial that is hard to recover. Years later, it can come back as a takeback. A payer audit can still ask for the money 12 to 24 months later.
Found by a weekly check
The note can still be written, the authorization still chased, the claim still corrected before it goes out. It costs an hour of someone's time.
Found on a billing company report
The same error, the same size, ninety days older. Often all that is left is an appeal or a write-off.
That is the argument for running these checks weekly, upstream of your billing company.
Run it on your own numbers.
I will not tell you what you are losing. Put your figures in. It computes in your browser and is sent nowhere.
The $391,176 is what these example inputs produce. It is not a finding about your business. This is arithmetic on figures you typed. It leaves out clawback exposure entirely, because that follows audit risk rather than volume. Treat it as a reason to look, not a result.
Go find the real number.
The one page checklist that turns this into an afternoon of work. Print it and hand it to your operations lead. No email, no gate.
A person can find these. A person cannot find all of them, every week, forever.
Give someone this page, one facility and one completed month, and they will learn more in a day than in a quarter of dashboards. What nobody can do by hand is run it across every facility, every week, before the billing company opens the sheet. That is the only version that lands inside the correction window, and it is the only reason the engine exists.
An illustration of coverage, not client data. Dashed = a partial check at month end.
Find out what yours is costing.
Start with a free 20-minute call. Walk me through how the work becomes money and I will tell you where I would look first. The paid Ops Audit comes after that, and only if it is worth it. The Ops Engine, our billing census automation, checks all five on every line.