Here is an email that lands in someone’s inbox every single day in this industry:
“We have outstanding invoices dating back to October. Your team can’t tell us what was used, what was billed, or what’s still pending. If this isn’t resolved by Friday, we’re pulling your devices from our facility.”
That’s a hospital administrator — at one of your busiest accounts — telling you they’re done. Not because your implant failed. Not because a competitor offered a better price. Because your billing operations are so broken that they’ve lost confidence you can manage the basic commercial relationship.
I’ve seen versions of this email more times than I can count. The hospital name changes. The dollar amount changes. The root cause never does.
This guide is for the operations leader who wants to fix the mess before the email arrives — or who just got one and needs a plan that goes beyond “let’s schedule a call with the hospital.”
The Anatomy of the PO Mess
The PO mess isn’t one problem. It’s three problems that feed each other, and they almost always show up together.
Problem 1: Usage Reporting Lag
A surgeon implants a device on Tuesday morning. The rep is in the room. The rep knows what was used. But the information about that case doesn’t reach the people who bill for it until… when, exactly?
In most medical device commercial operations, the answer is “it depends.” It depends on whether the rep remembers to text the coordinator. It depends on whether the coordinator checks that text before leaving for the day. It depends on whether the usage details are complete — lot number, catalog number, patient info, PO reference — or whether someone has to chase down the missing fields.
For consignment cases, this lag averages 2–4 weeks. For bill-only POs, it can stretch longer. Every day of lag is a day your hospital partner doesn’t have an invoice, and a day your finance team can’t recognize the revenue.
The lag isn’t caused by laziness. It’s caused by the fact that the person closest to the usage event — the rep — has no structured way to report it. They’re between cases. They’re driving. They’re prepping for the next surgery. So they text a coordinator who’s managing 15 other reps, and the information enters a queue that nobody audits until something breaks.
Problem 2: PO-to-Invoice Matching Chaos
Even when usage gets reported on time, converting that information into a correct invoice is its own obstacle course.
The coordinator receives a text that says “used 2 SP019s at Memorial, Dr. Chen’s case.” Now they need to match that to the right PO — which means finding the PO number in an email from three weeks ago, confirming the contract pricing, checking whether this hospital uses bill-only or consignment terms, and entering all of it into the ERP without transposing a digit.
One wrong catalog number and the invoice gets rejected. One wrong price and the hospital’s AP department sends it back. One missing PO reference and the invoice sits in a queue until someone calls to ask about it. Each of these errors adds 5–30 days to the collection cycle and requires staff time to investigate, correct, and resubmit.
Healthcare organizations lose an estimated 4–5% of revenue to billing-related leakage — credits, rebills, and write-offs caused by errors in the invoicing process [1]. For a medical device distributor doing $20M in annual revenue, that’s $800K–$1M walking out the door because of data entry mistakes and process gaps.
Problem 3: Reconciliation Drift
This is the quiet killer. Even when individual invoices go out correctly, the aggregate picture drifts over time.
Your records say Memorial Hospital has 47 units on consignment. Their records say 52. Nobody noticed the discrepancy until the hospital ran an audit and found five devices they were never billed for. Now they’re questioning whether your other consignment counts are accurate — and whether they should bother auditing the rest, or just switch to a competitor who can keep track of their own inventory.
Reconciliation drift happens because each case is processed in isolation. Nobody is comparing the running total of what’s been implanted against what’s been invoiced against what’s sitting on the shelf. The spreadsheet that’s supposed to track this gets updated by three different people who each have their own version. When the numbers don’t match — and they never match — nobody can tell which version is right.
Why the PO Mess Gets Worse, Not Better
If you’re reading this and thinking “we have this problem but it’s manageable,” I’d push back on the word “manageable.” Here’s why.
More reps means more handoffs. Every rep you add creates another communication channel that your coordinator has to monitor. If you have 10 reps texting usage to one coordinator, that coordinator is processing 10 different message formats, 10 different levels of detail, and 10 different response times. The error rate doesn’t scale linearly with headcount — it compounds.
More product lines means more PO types. When you carried one product family, your team knew the catalog numbers by heart. Now you carry three families with 70+ SKUs each. Your coordinator needs a reference sheet just to confirm whether “the small one” means the 19mm or the 21mm. Multiply that ambiguity across hundreds of cases per month and you start to see where the invoice errors come from.
More hospital systems means more billing portals. Hospital A accepts POs by email. Hospital B uses a vendor portal. Hospital C still faxes purchase orders. Hospital D requires everything through their supply chain management system. Your billing team is logging into four different systems, each with its own format requirements, each with its own approval workflow, each with its own timeline for payment. Standardization is a fantasy when your customer base can’t agree on how to send you a purchase order.
More volume means the spreadsheet breaks. The Excel workbook that tracked everything when you had 50 cases a month doesn’t work at 200. It’s too slow. It has too many tabs. It crashes when two people have it open. But nobody has time to build a replacement because everyone is busy processing this month’s orders — in the spreadsheet.
This is the doom loop. The busier you get, the more the system breaks. The more the system breaks, the more time your team spends fixing errors instead of preventing them. And at some point, a hospital administrator decides they’ve had enough.
The Three Things You Actually Need
I’m going to resist the temptation to tell you that you need to buy software. Maybe you do, eventually. But the companies that avoid the angry email — the ones whose hospital partners describe them as “easy to work with” — share three operational characteristics that have nothing to do with technology.
1. A Single Source of Truth for Every Case
Somewhere in your organization, there should be exactly one record for every surgical case that includes: what was used, which PO it maps to, what was invoiced, and what the current status is. Not a spreadsheet on one person’s desktop. Not a text thread. Not “check with Sarah, she knows.”
One record. One place. Updated by whoever touches the case, visible to everyone who needs it.
What this looks like in practice: When the hospital calls and asks “what happened with Dr. Chen’s case from January 14th?” — your team can answer in 30 seconds. They don’t have to dig through emails. They don’t have to call the rep. They pull up the case record and walk the hospital through it: here’s what was used, here’s the PO, here’s the invoice, here’s the payment status.
What this looks like when it’s missing: Three people spend 45 minutes reconstructing what happened. The rep says they texted the details. The coordinator says they never got the text. The biller says they invoiced it but used the wrong PO number. The hospital has now been on hold for an hour, and their confidence in your organization drops another notch.
Most teams think they have a single source of truth because they have a shared spreadsheet. They don’t. A shared spreadsheet is a single file. A single source of truth is a system where the data is entered once, at the point of origin, and flows downstream without re-entry. The distinction matters because re-entry is where errors live.
2. A Defined Process for PO-to-Invoice That Doesn’t Depend on Tribal Knowledge
Can you write down, right now, the exact steps that happen between “rep reports usage” and “invoice goes out”? Not the theoretical steps. The actual steps. Who does what, in what order, with what tools, and what happens when a step fails?
If the answer involves the phrase “well, it depends on who’s doing it” or “Sarah usually handles that” — you don’t have a process. You have a collection of habits held together by institutional memory, and it works exactly until Sarah goes on vacation.
What a defined process looks like:
Step 1 is usage capture — rep reports case details in a structured format (not a free-text message) within 24 hours of the case. Step 2 is PO matching — the case is linked to the correct purchase order using a documented lookup method, not someone’s memory. Step 3 is pricing validation — contract pricing is confirmed against the master price list before the invoice is generated, not after the hospital rejects it. Step 4 is invoice generation — the invoice goes out within a defined SLA, with all required fields populated. Step 5 is confirmation — someone verifies the invoice was received and entered into the hospital’s AP system.
Each step has an owner, a deadline, and an escalation path. When step 2 fails because the PO can’t be found, there’s a defined process for that, too — not a Slack message that says “hey does anyone know the PO for Memorial?“
3. Visibility for Your Hospital Partners
This is the one most companies skip, and it’s the one that prevents the angry email.
Your hospital partners are not unreasonable. They’re not sending you threatening emails because they enjoy it. They’re sending them because they have no way to know what’s happening on your side. From their perspective, devices were implanted, weeks have passed, and they have no invoices, no status updates, and no way to reconcile their own records.
Give them a way to see what’s coming.
This doesn’t have to be a portal (though it can be). It can be a weekly reconciliation report sent by email. It can be a shared document that shows pending cases and their billing status. It can be a standing call where you walk through the outstanding items together.
The format doesn’t matter. What matters is that your hospital partner never has to wonder whether you’ve forgotten about them. The angry email is always — always — preceded by silence. They called your office and didn’t get a callback. They emailed your coordinator and got a vague response. They asked for a reconciliation and received a spreadsheet with last month’s data.
Proactive visibility eliminates the conditions under which the angry email gets written.
The Operational Audit: Run This Before Friday
If you’re an ops leader reading this and wondering how bad your PO mess actually is, here’s a diagnostic you can run this week. No software required. Just honest answers.
Usage reporting lag. Pull your last 20 consignment cases. For each one, measure the gap between the surgery date and the date the usage was entered into your billing system. If the average is more than 3 business days, you have a lag problem. If any are over 2 weeks, you have a crisis.
Invoice accuracy rate. Pull your last 50 invoices. How many were rejected, credited, or rebilled? If the answer is more than 5%, your PO-to-invoice matching process has gaps. Above 10%, and you’re actively damaging hospital relationships.
Case reconstruction time. Pick a random case from 60 days ago. Time how long it takes your team to tell you: what was used, what PO it was billed against, whether the invoice was sent, and whether it was paid. If it takes more than 2 minutes, your data is fragmented. If it takes more than 10 minutes, you don’t have a system — you have an archaeology project.
Reconciliation currency. Ask your top 5 hospital accounts: “Is there anything outstanding between us that we haven’t addressed?” If the answer surprises you, your reconciliation process isn’t running often enough. If you’re afraid to ask the question, that tells you everything.
Single point of failure test. Identify the one person on your team who handles the most billing volume. Now imagine they give two weeks’ notice tomorrow. How much of what they do is documented? How much lives only in their head? If losing one person would create a billing backlog, you’ve built a system around a human instead of a process.
Score yourself honestly. Most teams discover they’re worse off than they thought — not because they’re bad at their jobs, but because the system was never designed. It evolved. And evolved systems accumulate debt faster than anyone notices.
What “Good” Looks Like
Good isn’t perfect. Good isn’t zero errors. Good is intentional.
The medical device companies that maintain strong hospital relationships and clean billing operations share a few traits that have nothing to do with which ERP they run or how many people they have on their ops team.
They treat billing operations as infrastructure, not overhead. The difference is subtle but important. Overhead is something you minimize. Infrastructure is something you invest in. Companies that treat billing as infrastructure assign senior people to own it, measure it, and improve it. Companies that treat it as overhead assign the cheapest person available and hope nothing breaks.
They have written playbooks, not oral tradition. When a new coordinator starts, they don’t shadow Sarah for three weeks. They read a documented process, follow the steps, and escalate when something doesn’t match the playbook. The playbook gets updated when the process changes — not six months later when someone remembers.
They reconcile proactively. Instead of waiting for the hospital to call with discrepancies, they run reconciliation reports monthly and share them first. This completely changes the dynamic. You’re no longer the company that can’t keep track of its own billing. You’re the company that catches problems before the customer does.
They measure what matters. Days from surgery to invoice. Invoice rejection rate. Average time to resolve a billing discrepancy. These aren’t vanity metrics — they’re early warning signals. When the numbers drift, you fix the process before it becomes a relationship problem.
None of this requires a technology investment. It requires a decision — a decision that your commercial operations infrastructure is worth the same attention you give to product development, sales strategy, and regulatory compliance.
The Decision
The work between surgery and collected revenue matters. Everything in this guide comes down to whether you decide to treat it that way — and then build the operational discipline to back up that decision.
The companies that make this decision early build a moat. Their hospitals prefer working with them because the billing is clean, the communication is proactive, and problems get resolved before they become relationship damage. Their reps sell more because they’re not spending half their week chasing down PO numbers. Their finance team recognizes revenue faster because the data flows instead of sitting in someone’s text messages.
The companies that don’t make this decision eventually get the email. And by then, you’re not building operational discipline. You’re scrambling to save a $2M account with a spreadsheet and a prayer.
A little intentionality goes a long way. But it has to be intentional. The PO mess never fixes itself — it just gets quieter until the next email.
Sources
[1] Huntington Bank. “The True Cost of Healthcare Revenue Leakage.”
[2] MD Clarity. “Revenue Leakage in Healthcare: Sources, Impact, and Fixes.”
[3] Salesforce. (2024). State of Sales Report, 5th Edition.
[4] ImplantBase. “Field Inventory Management: Challenges and Solutions.”
[5] Billtrust. “How to Reduce Days in AR in Medical Billing.”