E1 · Patient access
Eligibility and authorization check
Leak: services delivered without active coverage or a valid authorization.
Lumbar MRI booked 03/12 · no authorization on fileheld at scheduling · prior auth request drafted · $1,850 protected
OTLEN / Online products
Software that closes revenue leaks in healthcare.
Digital products for health systems, revenue cycle teams, private clinics and med spas. They find where revenue escapes, work it back, and keep it from leaking again.
Revenue leaks through ordinary work: a charge that is never entered, a denial no one appeals, a payment below the contracted rate, a package session used and never drawn down. OTLEN’s product work starts there: identify the repeated leak, understand the workflow around it, and remove the steps that let it happen.
Show examples for
Where revenue leaks · the revenue cycle, patient access to patient balanceWhere revenue leaks · a clinic day, booking to provider payout
Each leak links to its product area
No.01
Where revenue leaks
Revenue leaks at specific, repeatable points: before the visit, at the point of care, at the claim and after payment. Each product area targets one of them, for health systems and revenue cycle teams, and for private clinics and med spas.
E2 · Point of care
Charge capture reconciliation
Leak: documented services that never become charges.
Infusion documented 2 h 10 min · charged: first hour onlyadd-on hour missing (CPT 96366) · routed to coding
E3 · Coding
Pre-submission claim scrubber
Leak: denials for errors that could have been caught before the claim left.
CPT 99214 + 20610 same day · no modifier on the E/Mmodifier 25 likely required · claim held before submission
E4 · Claims
Denial worklist and appeals
Leak: denials that are never worked before the appeal window closes.
2,140 open denials · $3.86M at stake · sorted by date receivedranked by dollars and days left · $1.12M closes within 14 days, worked first
E5 · Payment
Contract underpayment detection
Leak: payments below the rate the payer contracted to pay.
CPT 73721 · paid $412.00 vs. contract $468.00 · same gap on 7,400 lines this year$414,400 underpaid · appeals queued with the contract terms
E6 · Patient balance
Patient balance resolution
Leak: patient responsibility billed late, billed wrong, or never billed.
3,120 patient balances past 45 days · $1.86M outstandingestimate vs. actual explained · payment plans offered · statements sent
C1 · Booking
No-show and late-cancel recovery
Leak: missed appointments with a policy fee that is never charged.
212 no-shows and late cancels this month · $150 policy fee$31,800 chargeable under policy · 47 card requests drafted
C2 · Treatment
Charted-versus-billed check
Leak: units, syringes and add-ons given but not rung up.
Charted 44 units, billed 24 · same gap on 312 visits this quarter6,240 units unbilled · $74,880 at $12 a unit (example price)
C3 · Packages
Package and membership ledger
Leak: sessions used without being drawn down, credits that never expire.
1,140 open packages · 418 out of balance$96,140 in sessions used, never drawn down · ledgers corrected after review
C4 · Checkout
Point-of-sale to EMR match
Leak: visits that close in one system and not the other.
4,860 signed notes this month · 4,791 paid tickets69 visits with no ticket · $31,050 unbilled, listed before close
C5 · Payments
Deposit, refund and discount audit
Leak: deposits never applied, refunds without approval, discounts outside policy.
Discount 40% on a filler visit · policy maximum 20%flagged for manager approval · $38,400 outside policy this quarter
C6 · Payouts
Provider compensation check
Leak: commission paid on revenue that was refunded or never collected.
Commission on $1.24M collected · $86,500 later refunded$10,380 overpaid at 12% (example rate) · adjustment queued
Inputs and outputs are specimens written to show each leak. They are not records from a live system or from any client.
01 · Revenue at risk
A leak is a rate. Volume turns it into money.
revenue at risk per year
Volume × 12 × leak rate × average value; recoverable = that × recoverable share. The defaults are an example sized for a mid-to-large health system or a six-location med spa group, not a benchmark: enter your own figures. The bars hold everything fixed except the leak rate.
No.02
AI inside the workflow
AI is most useful when it handles ambiguity that would otherwise require a person: classifying inbound information, extracting fields, summarizing context, matching records, drafting a next action, or deciding which exception needs attention.
Inbound · remittance advice · example
Commercial payer · claim 4471-0932
DOS 03/12 · CPT 72148 MRI lumbar spine w/o contrast Billed $1,850.00 · allowed $0.00 · paid $0.00 Adjustment CO-197 · remark M62 Payer note: “Precertification absent. Submit a corrected claim with the authorization number within 60 days.”
Inbound · end-of-day record · med spa · example
Visit 0318-14 · one patient, one provider
Treatment note (signed): neuromodulator, 44 units, glabella and forehead Checkout ticket: neuromodulator, 24 units, paid Package: laser series, 3 of 6 used · ledger shows 2 Ticket closed 17:42 · note signed 18:05
Example output
Without it
No.03
One job at a time
Each product should have a clear promise. The interface should make that promise obvious in seconds. Complexity can stay inside the system; the user should not have to understand the architecture to get the outcome.
What the revenue team sees
Load this week’s remittances. See every claim paid below contract.
ERA files, 03/10–03/164,812 lines
Payer contracts14 loaded
$957,840a year below contract, at this week’s rate37 claims · $18,420 this week
Open the 37 claims
What stays inside · 8 jobs behind 1 promise
| Parse | 835 remittance files from every clearinghouse |
| Load | payer contracts, fee schedules and carve-outs |
| Price | every line at the contracted rate |
| Separate | patient responsibility (PR) from contractual (CO) |
| Compare | allowed amount against expected |
| Group | variances by payer, code and reason |
| Log | every price and every variance |
| Queue | appeals with the contract terms attached |
Screen figures are an example; the yearly figure is this week × 52. The revenue team sees one promise and one result; the contract logic stays inside.
What the clinic manager sees
Close the day. See every service given but not charged.
Signed notes, 6 locations238
Paid tickets227
$1,029,600a year unbilled, at today’s rate11 services · $3,960 today
Review the 11 services
What stays inside · 8 jobs behind 1 promise
| Read | signed treatment notes from the EMR |
| Read | tickets and payments from the point of sale |
| Match | by patient, visit and provider |
| Convert | units and syringes into billable items |
| Check | package and membership balances |
| Price | at the clinic’s own menu |
| Log | every match and every miss |
| Hold | tickets edited after close |
Screen figures are an example; the yearly figure is today × 260 working days. The manager sees one promise and one result; the matching stays inside.
No.04
Product experiments
Some tools will become products. Some will stay internal. Others will prove that the problem is not important enough to solve. That is useful information too. A product lab should optimize for learning speed as much as launch speed.
Becomes a product
Revenue teams keep using it after the pilot.
Stays internal
It runs inside one organization’s workflow, with no market behind it.
Not worth solving
The leak cost less than the fix.
All three are results.Learning speed ≈ launch speed
No.05
From prototype to product
When a tool survives repeated real-world use, the next job is hardening: permissions, security and compliance, billing, monitoring, documentation, reliability, onboarding, and support. That is when a clever prototype becomes software people can depend on.
Hardening
Example state: three of eight done. Solid squares are hardened; dots are still prototype.
A focused product that recovers real revenue is the fastest route to a bigger system.
This is an editorial starting point, designed to grow with original articles, projects, research, tools, examples, and case studies specific to this property.