~4% Of expected revenue never arrives
A 2.7% final denial rate on top of 1.3% bad debt. Both are recoverable in part, and both are treated as the cost of doing business.
Healthcare
You are spending nineteen billion dollars a year reworking claims you already earned.
Denials are not a billing problem. They are a documentation problem that surfaces as a billing problem.
Book a 20-minute call Free. No demo, no slides.
$48.4B
of hospital revenue lost to denials and uncollected bills in one year, up 25 percent.
Source: Kodiak Solutions, State of the healthcare revenue cycle
What is happening
The claim was right the first time. It just could not prove it.
$57 Per reworked claim, and you rework a lot of them
US hospitals spend roughly $19.7 billion a year overturning denials. The cost is almost entirely staff time.
+12-14% Denied claim values are climbing
Average denied inpatient claim value rose 12% and outpatient 14% in one year, alongside rising audit volume.
The evidence exists but is not where the claim is
Medical necessity lives in physician narrative and nursing documentation. Your claims system holds structured fields. Bridging the two is manual, every single time.
Intelligence, plumbed in
A denial letter and a chart are both text. Both can be read.
The work is HL7 and FHIR plumbing, one agreed meaning per field, and an eval set of appeals your team already won.
How we make AI survive real data-
Connectors -
A semantic layer -
Evals you can check
What changes
Judge us on this, not on what we built.
Denials Appeal packets assembled, not researched
The clinical evidence pulled from the record, matched to the payer's stated reason, formatted to that payer's requirements.
Prevention Denial risk flagged before submission
Patterns from your own denial history, applied to claims that have not gone out yet. The cheapest denial is the one that never happens.
Visibility Denial performance by payer, service line and physician
So the conversation with a payer is evidence-led, and so you know which fights are worth having.
Research
We wrote a full page on each of these, with the sources.
Each one carries its own figures and the citations behind them. Start with whichever sounds most like your week.
However hard, whatever it is
Denials are one example. Bring the one that is actually costing you.
This page happens to be about appeals. If your problem is theatre scheduling, bed flow or a report nobody trusts, we start in exactly the same place.
-
01 We sit with you
Days where the work happens, not a workshop in a meeting room. We watch the job get done and write down the shortcuts nobody wrote down.
-
02 We read everything
Your data, your rules, your vendors and their documentation, and the published research on your sector. We report what is actually in there.
-
03 We break it to first principles
Not which tool fixes this. What is actually causing it, taken apart until we reach the piece that cannot be divided further.
-
04 Then we build
Weeks, not quarters. By this point we are not guessing what to build, and guessing is the thing that makes projects long.
Where we sit
Your EHR stays. We build the work that happens around the chart.
4
What you get
- Denials Appeal packets assembled, not researched
- Prevention Denial risk flagged before submission
- Visibility Denial performance by payer, service line and physician
3
Built new for you — none of this exists in your stack today
- An appeal pack that assembles itself from the chart
- A denial board the team stands at every morning
- A bedside tag that says when a check last happened
2
The revenue-cycle layer Reads the chart. Writes nothing into it that you have not approved.
Connectors, one agreed meaning per field, and a model reading what no
field holds. Accuracy measured on your own records.
1
- Epic
- Cerner / Oracle Health
- Meditech
- Athenahealth
- Waystar
What you already run — unchanged, and still yours
If it is in the chart, we can reach it. HL7 v2, FHIR, an interface engine, a scanned consent form. We have built against all four.
-
No API -
No documentation -
A terminal from 1994 -
It arrives as paper -
The vendor said no -
It reports nothing
Not a list of limits. Name yours on the call.
And once we can reach it, a model can read it. Most of the value here is in the sources nobody ever structured — the note, the letter, the screen.
Who this is for
The people who feel this first
-
VP Revenue Cycle -
Director of Patient Financial Services -
CFO -
Chief Medical Information Officer -
Director of HIM
Straight answers
The questions you would ask on the call
-
We already have a revenue cycle vendor. Where do you fit?
Underneath them, usually. Billing platforms and RCM vendors are built to submit claims and follow up on status. The work that does not get done is arguing a denial, because that means reading the medical record rather than querying a database. That is the gap we build into.
-
Do you replace our EHR?
No, and you should be suspicious of anyone who suggests it. We build the operational layer on top of the system of record you already have.
-
How do you start?
A free call, then a proper look at your data. We come back with your real denial and appeal numbers, what you can recover, and whether a build is worth it. If it is not, we say so.
-
Can AI draft appeals from the chart?
It assembles the evidence and drafts a first version. A person signs every one. We score the drafts against appeals your team already won.
Also in healthcare
Healthcare
Skilled nursing & post-acute care
Your team is not wrong about the denials. They just do not have four hours per appeal to prove it.
HealthcarePhysician groups & clinics
Ninety-two percent of practices have hired or reassigned staff purely to keep up with it.
HealthcareHome health & hospice
Caregiver turnover is at 79%. The software you bought was designed around the chart, not the caregiver.
Next step
Start with your own numbers.
-
We talk
20 minutes. Free.
You tell us what is not working. We ask how the work really gets done.
-
We look at your data
A few weeks.
We read your systems, including the notes and letters no field holds. You get what is really in there, what it costs, and the accuracy we can hit.
-
We build
A few months.
Only if step 2 says it is worth it. Fixed price, agreed before we start.