Suggested codes on routine encounters; denials get the specialist

The question this file answersWhat share of our claims bounce on first submission — and how long does a coder then spend on each one?

Fits: hospitals and groups where coders read every note and work denials in the same queue — the healthcare playbook gives no encounter count for this desk; the audit measures yours.

Not for: organisations that have already outsourced coding and denial management end to end — there is no in-house queue to reshape.

Typical day

What the desk looks like today

Typical, from the healthcare playbook — not a client's day. Every encounter produces coding work; the playbook gives no daily count, so the audit measures yours. A coder reads the clinician's note, assigns ICD and CPT codes, submits the claim, and — days or weeks later — gets the payer's query or denial, researches, resubmits, tracks payment. Routine visits and ambiguous notes share one queue, so the easy delay the hard. It hurts when denials come back in a batch and the coder must stop coding today's notes to argue about last month's.

What changes

What Monday looks like after

By the time the coder logs in, the overnight notes are queued with codes suggested and the supporting phrases highlighted; the coder confirms or changes each one — the decision is theirs, the reading from scratch is gone. Two claims are held with a note — "no laterality documented" — caught today, not in a denial next month. The denials that did arrive are grouped by reason, the routine queries have a drafted reply, and the appeal is with the specialist. Your revenue-cycle manager can see denials by reason and by payer. AHIMA (2023) measured assisted-coding accuracy on routine encounters — theirs, quoted above, not a promise about your coders.

Typical, not a measured client result. Every figure here comes from the playbook source named below.

~35–55%

of coding effort assisted — accuracy is still reviewed

Before: a coder reads every note and lives in the denial pile. After: AHIMA (2023) reports 85–90% suggestion accuracy on routine encounters — appeals and grey notes stay a specialist.

Where this number comes from

AHIMA (2023): AI-assisted coding achieves 85–90% accuracy on routine encounters; McKinsey (2023): revenue-cycle automation can reduce denial rates by 20–30% through better first-pass accuracy. Industry figures, not our measurement. Playbook range 35–55%. Not a revenue guarantee — every code is reviewed.

What we install

What we put in front of the systems you already run

Coding remains a coder's decision and the billing system remains yours — Waystar, Availity, Change Healthcare or the one you use, fed from Epic or Cerner. We add a preparation step and a sorting step:

each note is read and ICD/CPT codes suggested with the phrases that support them
before submission the claim is checked for the documentation payers ask for, a gap flagged while the clinician is reachable
returning queries and denials are sorted — a routine payer query gets a drafted reply for the coder, a patterned denial is grouped, an appeal goes to the specialist
claim status is read from the clearing house — no portal refreshing.

First scope: the visit types that carry most volume.

What stays human — and what this will not do

Complex coding decisions. Writing the appeal. Negotiating with a payer. Answering an audit. Every code is confirmed by a coder.

Not a revenue guarantee — every code is reviewed.

What can go wrong — and what we do about it

A suggested code for a thinly documented visit is a guess — hence a coder confirms every one, and the AHIMA 85–90% accuracy applies to routine encounters only. If your denial history is not coded by reason, the sorting has nothing to learn from at first; the manager sees noise before pattern. HIPAA means note access, logging and de-identification come before the first note is read, adding time. Appeals, payer negotiation and audit responses sit outside the playbook's 35–55%; a lower denial rate is a possible effect, not a promise.

How long it takes, and what we need from you

Audit, about two weeks (€1.5–3K): we pull a month of claims and denials, count first-pass denials by reason and payer, and check what your clearing house exposes and your EHR allows for notes. Pilot, 5–8 weeks — high complexity in the healthcare playbook, coding being regulated and every code reviewed (€10–20K): one specialty, suggestions and flags only, a coder confirms every code. Production: further specialties, then denial sorting and drafted replies. From you: de-identified sample notes, denial history, your compliance officer.

The path: free 60-second estimate → free 20-minute review → paid audit of this one process (€1.5–3K, typically two weeks) → pilot with your people in the loop (€10–20K, weeks, not quarters). No transformation programme. Prices are public, on the services page →

This is about you if…

Does a coder read every note from scratch, including the routine follow-ups?
Do denials and payer queries land in the same queue as today's coding?
Do you know your first-pass denial rate by payer — or would you have to build the report?

What does this mean in euros?

That depends on your volumes and wage costs — this page will not invent the number. The free 60-second estimate runs that calculation from your answers, with every multiplier sourced.

Get your free savings estimate 60 seconds · no sales call Or write first → Map a healthcare process like this one — free, 60 seconds →

Not a named Aperanda client. Process file · Healthcare.

Deep-dive process file. Volumes, weeks and sources come from the industry playbook; nothing here is a named client.

All process files