The new biller starts Monday. The person she is replacing left two weeks ago, the October ICD-10-CM update just landed, and the practice manager has blocked out "training" on the calendar with no plan for what it contains. By Thursday the new hire is posting payments from a stack of paper EOBs because that seemed like a safe thing to give her, and nobody has explained what an adjustment code means.

We have trained billers in-house, for client practices and through our live online courses, and the pattern is consistent: people who are given a sequence learn in a month; people who are given a pile take six and make expensive mistakes on the way. The fourth quarter makes this worse because there is no slow week to learn in. The plan below is built for that.

Key takeaways

  • Week one is understanding, not production: the anatomy of a claim, ten real claims followed end to end, and every adjustment code on their remittances explained aloud.
  • Rejections come first because they are specific and mechanical; posting comes third because its errors hide longest.
  • By week four the trainee owns one payer's AR completely. Owning one payer teaches more than touching every payer superficially.
  • Name the four predictable errors on day one and check for them specifically. It is faster and kinder than letting a trainee find them on a remittance.
  • Ready to work unsupervised means three consecutive clean weekly samples, usually between day 30 and day 45.

Before day one

Have the logins ready: practice management system, EHR (read access to notes), clearinghouse, the top five payer portals, and the shared drive where payer manuals and fee schedules live. Print or share three documents: the practice's payer list with filing and appeal limits, the write-off policy with adjustment codes, and a one-page map of who does what in the revenue cycle. A new biller who does not know who fixes eligibility problems will try to fix them herself.

Block the trainer's calendar. Two hours a day in week one, an hour a day in week two, thirty minutes a day after that. If the trainer is also working a full queue, the training will be squeezed out by the first payer call of the morning, and the trainee will be left with the pile. Tell the owner in advance what the trainer's production will look like for the month.

Week one: the claim, end to end

The goal of week one is not production. It is understanding what a claim is and where it goes. We spend the first two days on the anatomy of the 837P and its paper twin, the CMS-1500: the fields, what feeds them, and what breaks when they are wrong. Then the new biller follows ten real claims from the schedule through the encounter, the charge, the scrub, the clearinghouse acceptance, and the remittance. She reads the 277CA acknowledgements and the 835s for those ten claims and explains each adjustment code aloud. PR-1 is the deductible. CO-45 is the contractual write-off. CO-16 needs a remark code to mean anything. CO-197 means an authorization was expected. By Friday she should be able to look at a remittance and say, in plain words, what happened to each line.

Choose the ten claims deliberately. Include two that paid in full, two with a deductible, one with a secondary payer, one denied CO-22, one rejected at the clearinghouse and corrected, one with a modifier 25 visit and procedure, one Medicare claim with a crossover remark, and one that was appealed. Ten claims chosen this way cover most of what a biller will see in a year.

The evergreen tool for this week is a training environment where mistakes cost nothing. Practices without a sandbox in their own system can use a training EHR such as Revelrex EHR, where a trainee can create encounters, charges and claims against fictional patients and see what a rejection looks like before she causes a real one.

Week two: rejections and eligibility

Rejection review is the first task a new biller can own, because rejections are specific and the fix is usually mechanical. She runs the clearinghouse rejection report every morning, groups rejections by reason, corrects the ones with obvious fixes (subscriber ID format, missing rendering NPI, place of service mismatch), and brings the rest to the lead by 10 a.m. She also runs eligibility for the next day's schedule and learns to read a 271 response: active coverage, copay, deductible remaining, and the plan type, which decides whether a referral or authorization applies.

The rule for week two is that nothing is resubmitted unchanged. Every corrected rejection gets a one-line note saying what was wrong and what was fixed. By Friday the trainee should be able to say which three rejection reasons are most common in this practice and where each one starts, which is the beginning of thinking about process rather than claims.

Weeks three and four: posting, denials and the first real queue

Week three: payment posting and the denial queue

Posting comes third, not first, because it is the task where errors hide longest. An ERA posted with the wrong adjustment code looks fine until month end. We start with electronic remittances from one payer, checked line by line by the lead for three days, then expand. Balancing each posting batch to the deposit is not optional and is taught on day one of this week: the batch total in the system must equal the EFT or check amount before the batch is closed, every time, and an unbalanced batch is a stop, not a note.

The denial queue starts in week three as well, limited to two categories: coordination of benefits (CO-22) and duplicates (CO-18). Both are common, both are fixable without a clinical judgment, and both teach the difference between a corrected claim and an appeal. A CO-22 needs the right primary and a fresh claim; a CO-18 needs a decision about whether anything actually changed and, if so, a frequency code 7 with the payer's original claim number. Neither needs a letter.

Week four: follow-up and one payer's AR

In week four the new biller takes ownership of one payer's AR: every open claim over 30 days for that payer, worked oldest first, with a note on every touch. She learns the payer's portal, its claim status codes, its appeal form and its call center scripts. Owning one payer completely teaches more than touching every payer superficially, and it gives the lead a clean way to measure progress. Pick a mid-sized commercial payer for this, not Medicare and not the largest plan: enough volume to learn from, small enough that a mistake is recoverable.

What to check, and how often

WeekWhat the trainee ownsWhat the lead checksHow
1Nothing in productionExplanation of ten claims and their remittancesConversation, Friday
2Daily rejection report; next-day eligibilityEvery correction, first three days; then a 25% sampleClearinghouse resubmission log
3ERA posting for one payer; CO-22 and CO-18 denialsEvery posted batch balanced to deposit; every corrected claimBatch report plus corrected-claim list
4Full AR for one payerNotes on every touched claim; ten-claim sample weeklyFollow-up notes and status codes

The checks are the training. A lead who reviews the batch report and finds an unbalanced batch has a ten-minute teaching moment with a real example; a lead who finds it at month end has a two-hour reconstruction and a trainee who has been doing it wrong for three weeks.

The errors to expect

New billers reliably do four things. They change a diagnosis code to clear a rejection without checking the note. They post a payer's reduced allowed amount as a contractual adjustment when it was a denial that should be appealed. They send a corrected claim without the frequency code, producing a duplicate denial. And they resubmit a rejected claim without changing anything, hoping it will go through this time. Name these four on day one, and check for them specifically in weeks two and three. We think it is kinder and faster to tell someone the mistakes everyone makes than to let her discover them on a remittance.

The first one is the most serious, because it is a compliance problem and not just a billing one. A biller may fix a diagnosis code that was keyed wrong when the note plainly supports the correction; she may not choose a code the note does not support because it clears an edit. Say this on day one, in those words, and again in week two when the first rejection for a diagnosis edit arrives.

Questions we hear

Can we skip week one? We need production now.

You can, and you will pay for it in weeks five through twelve as the posting and adjustment errors surface. Two days on the anatomy of a claim is the highest-return time in the whole plan.

What if the trainer is also the only other biller?

Then the training is the trainer's main job for four weeks and the practice should expect AR follow-up to slip for that period. Say so to the owner in advance. The alternative, an untrained biller working the queue alone, costs more. Structured outside courses can carry the concept teaching while the in-house lead handles the practice-specific parts; that is exactly the split our training courses are designed for.

How do we know when the trainee is ready to work unsupervised?

When three consecutive weekly samples come back clean on the four expected errors, and when she brings the lead a rejection she has never seen before with the payer manual page already open. Both usually happen between day 30 and day 45.

What to do this week

  1. Request every login the trainee will need, and confirm each one works before Monday.
  2. Assemble the three documents: payer list with filing and appeal limits, write-off policy with adjustment codes, and the one-page map of who does what.
  3. Pick the ten teaching claims from last month's remittances, covering the mix above, and print or save their 837, 277CA and 835 records together.
  4. Block the trainer's calendar for the month and tell the owner what production will look like.
  5. Write the four expected errors on one page and put it at the trainee's desk on day one.