If you want to find the least efficient hour in a medical billing operation, look for the person with the prior authorization spreadsheet.
It is usually one person. It is usually a spreadsheet, or a shared inbox, or a stack of printed portal confirmations. It contains the authorizations that were submitted, the ones that were approved, the ones that were denied, the ones that are pending, and the ones that somebody thinks were submitted but cannot confirm. That last category is the expensive one.
Prior authorization gets discussed as a clinical burden, and it is. But operationally, the majority of the damage is not clinical disagreement. It is administrative: submissions that were never followed up, approvals that expired before the service was scheduled, and services that were rendered before anyone confirmed an authorization existed. Every one of those becomes a denial that is very hard to appeal, because the payer’s position is simple and defensible. You did not get authorization.
Why Prior Auth Breaks Down in Practice
The core problem is that prior auth has no single system of record.
Each payer has its own portal, its own form, its own documentation requirements, its own turnaround expectations, and its own way of communicating a decision. Some notify by portal message. Some fax. Some mail a letter. Some do nothing at all and expect you to check. A billing operation working fifteen payers is working fifteen incompatible processes simultaneously, and the only thing holding it together is a human maintaining a list by hand.
That list is fragile in predictable ways. It does not know that an approval issued in March expires in June. It does not alert anybody when a payer has been sitting on a submission for eleven days. It does not catch that a scheduled procedure next Tuesday has no authorization on file. It only knows what somebody typed into it, and only until that person is on vacation.
Meanwhile the volume keeps climbing. Payers have expanded prior auth requirements across imaging, specialty drugs, procedures, and durable medical equipment. Industry surveys have repeatedly found practices reporting many hours of staff time per physician per week on prior authorization, and those are industry level figures rather than a claim about your specific operation. Either way, the trend line is not moving in your favor.
The Tracking Layer Comes First
Before automating submission, automate knowing.
A prior authorization tracker should be the authoritative record of every auth request in flight, and it should populate itself. It reads the upcoming schedule from your practice management or EHR system and identifies services that require authorization based on payer and CPT. It knows which of those already have an authorization on file, which have a submission pending, and which have nothing at all. It knows each authorization’s approval date, expiration date, and how many units or visits remain against it.
Then it does the part humans forget. It escalates. A submission that has been pending past the payer’s stated turnaround gets flagged. An authorization expiring within two weeks with services still scheduled against it gets flagged. A procedure on next week’s schedule with no authorization gets flagged loudly, while there is still time to do something about it.
That alone eliminates most authorization related denials, because most authorization related denials are timing failures rather than clinical failures.
Automating Submission Across Payer Portals
Submission is more mechanical than it feels, because most of what goes into a prior auth request already exists in your systems.
Patient demographics, insurance information, ordering and rendering provider details, diagnosis codes, procedure codes, and the supporting clinical documentation are all sitting in the practice management or EHR system. The manual work is transcription: a staff member reading data out of one system and typing it into a payer portal form, then attaching documents, then submitting, then screenshotting the confirmation number.
Automation handles that transfer. Where a payer supports an electronic prior authorization transaction, the request goes through the clearinghouse. Where a payer only offers a web portal, and many still only offer a web portal, an automation navigates the portal, completes the form from the source data, attaches the documentation, submits, and captures the confirmation number back into the patient record.
The clinical content is not being generated by a machine. The clinical documentation comes from the chart, authored by the people who authored it. What gets automated is the moving of that information from one place to another, plus the record keeping around it.
Follow Up Is Where the Recovery Lives
Submission is the part everyone remembers. Follow up is the part that determines whether the authorization actually arrives.
Automated follow up checks the status of every pending request on a schedule, pulling from payer portals and electronic responses, and updates the tracker without anybody logging in. Approvals get written back to the patient record and to the scheduled service so the front desk and the biller both see them. Denials get routed immediately to the person who can respond, with the payer’s stated reason attached, because a prior auth denial has an appeal clock too and it is often shorter than a claims appeal clock.
Requests that go quiet get escalated on a timer instead of on a memory. That single behavior recovers a meaningful amount of revenue in most operations, because the pending request nobody chased is the most common way an authorization silently fails.
Reporting closes the loop. Once the tracker is authoritative, you can finally answer the questions that were previously unanswerable. Which payers take the longest. Which service lines generate the most authorization work. How often authorizations expire unused. Where scheduled revenue is at risk this week because authorization is not in hand.
Fitting It to the Systems You Already Run
Nobody is replacing their practice management system to fix prior auth, and they should not have to.
This work sits on top of what exists. It reads from your practice management or EHR system, transacts through your clearinghouse where payers support it, works your payer portals where they do not, and pushes results back into the record your staff already uses. Financial impact flows through to your accounting stack, including QuickBooks, once scheduled revenue stops falling out for missing authorizations. No rip and replace, no new interface for your staff to learn, no migration project.
Our approach to revenue cycle operations is detailed on the medical billing and revenue cycle page, and the broader automation model is under solutions.
A Sensible First Build
Pick the payer and service line combination that generates the most authorization volume, and build the tracker for it first. Not the submission automation. The tracker.
Once you can see every pending, approved, and expiring authorization for that segment in one place, with escalation running automatically, you will find out quickly where the actual losses are. Sometimes it is submission volume. Often it is follow up. Occasionally it is that nobody knew authorization was required for a specific code until the denial arrived. Build the submission automation second, against whatever the tracker proves is expensive.
We start with a free 30 minute Waste Audit to map where prior authorization is actually costing you, in staff hours and in scheduled revenue that falls out. If we build something, we take 15 percent of documented savings and you keep 85 percent. No savings, no fee. No platform license, no long contract, no replacing systems that work.
The authorizations are going to be required either way. The only thing you control is whether anyone is watching them.