I started in medical billing in a windowless room with a fax machine that jammed if you looked at it wrong, and I finished as the person who signed off on a seven-figure AR number every month. Eleven years. Three different systems. One extremely specific set of opinions.

I don’t work in a billing office anymore — I consult now, which mostly means I get invited into other people’s billing offices to explain why their money is stuck. And I’ve noticed something: the conversation almost always starts in the wrong place.

People ask me “what’s the best billing software.” What they should ask is “why is my money slow.” Those sound like the same question. They’re really not, and the gap between them is where most of this industry’s revenue goes to die.

So here’s my list. It’s organized around what actually fixes the money problem, not around which vendor has the prettiest dashboard.

One note before we start: I’m writing from a DME and HME background. If you’re a physician practice, some of this transfers and some doesn’t — the equipment side has documentation requirements that a primary care office never deals with, and that difference matters more than people expect.

1. NikoHealth

Here’s the thing I learned around year six that reorganized my whole understanding of this job: my department was not a billing department. It was a quality control department for other people’s paperwork.

I did the math once, because I was annoyed. I pulled a quarter of denials and traced every single one back to its root cause. Somewhere north of 70% of them had nothing to do with anything my team did. Wrong or missing documentation at intake. An authorization that quietly expired between the order and the delivery. A proof of delivery that never came back from the truck. A CMN that was never collected because nobody at the front asked for it.

My billers weren’t billing. They were detectives, reconstructing what should have happened days earlier.

That experience is why NikoHealth is first on my list, and it’s not really a statement about their claim engine. It’s a statement about architecture. When intake, documentation, delivery, and billing live in one system, the system can stop a bad order at the front door instead of letting billing find it three weeks later. The order can’t be submitted incomplete because the software knows what this payer needs for this product. The POD attaches itself, because delivery is in the same platform, not in a separate app that syncs overnight.

That’s the difference between medical billing software and a revenue cycle that actually works. One processes claims. The other prevents them from being wrong.

On the pure RCM side, everything you’d expect is there — eligibility verification, claim scrubbing, ERA posting, denial workqueues, aging by payer. Their rcm software overview walks through it. But I want to be honest with you: most platforms have a competent claim engine in 2026. That’s table stakes. The differentiator is upstream, and it always has been.

What I’d ask in a demo, as a billing person: “Show me the order that can’t be submitted.” Make them demonstrate the system blocking an incomplete order in real time. If they can only show you reports about problems after the fact, that’s a reporting tool, not a prevention tool.

Where I’d push back: Ask hard about your specific payer mix, especially regional Medicaid plans, which are where every platform’s connectivity gets thin. And ask what the transition period looks like for AR — running two systems while your old AR winds down is the genuinely painful part of any billing migration, and a vendor with a real plan for that is telling you they’ve done it before.

2. Bonafide

If you’re going to shop this category seriously, Bonafide belongs on your list. Their reputation in revenue cycle specifically is strong, and the billing people I know who use it generally like the workflow, which is not something I can say about every platform I’ve touched.

If your operations are basically fine and your problem is purely financial (slow AR, high denial volume, collections chaos) this is a serious option.

What I’d ask: Show me the front end. I’d want to see what intake looks like, because that’s where my denials come from.

3. Brightree

The one everybody knows. Deep payer connectivity, huge install base, a real third-party ecosystem.

The underrated advantage from a department-manager perspective is hiring. I could post a billing job and get people who already knew the system, and in a market where good billers are hard to find, that’s worth real money in reduced ramp time.

What I’d ask: Full cost with every module you actually need, and how fast support moves when a payer issue is costing you money this week rather than next quarter.

4. CareTend (WellSky)

If you’re in infusion or you’ve got a genuinely complex clinical-plus-equipment business, the clinical and billing integration here earns its keep.

If you’re straightforward HME, I think it’s more system than you need. Complexity you don’t use still costs you — in license fees, in training time, in the number of clicks between your biller and a submitted claim.

5. TIMS Software

Mature and capable, especially in pharmacy-adjacent operations. It has handled complicated billing scenarios for a very long time because it’s had a very long time to encounter them.

The honest trade-off is the interface and the learning curve. Budget more training time for new hires than you think you need.

6. Clearinghouse-only setups

Some smaller shops run a basic practice management system plus a clearinghouse and call it a day. I want to name this because it’s more common than people admit.

It works, sort of, at low volume. What you give up is visibility — you’ll know claims went out and some came back, and you’ll have a much harder time answering “why.” The moment you want denial analytics by reason and payer, you’ve outgrown it.

The five numbers I lived by

Whatever you buy, these are what I’d put on a whiteboard:

• Clean claim rate. First-pass acceptance. This is the master number. Everything else is downstream of it.
• Days in AR, segmented by payer. Never look at the aggregate alone. The aggregate is an average of one payer who pays in 12 days and one who’s been sitting on you for 90, and the average tells you to do nothing.
• Denials by reason, sorted by dollars. Sorted by dollars, not count. I watched a team spend a whole quarter chasing a high-volume, low-value denial reason while a handful of five-figure claims aged out. Count-based reporting caused that.
• Touches per claim. Almost nobody tracks this and it’s the truest measure of whether your software is helping. A claim a human opens once is a healthy claim. A claim opened four times is a process failure you’re paying salary to absorb.
• Days from delivery to submission. If this is creeping, your documentation flow is broken, not your billers. Go look at where PODs are sitting.

What I’d tell my year-one self

You are going to get extremely good at fixing denials. That skill will make you valuable, and it will also trap you, because a department that is excellent at cleaning up messes gives everyone upstream permission to keep making them.

The fix was never a better worklist. It was a system where the mess couldn’t be created.

If you’re shopping right now, skip the feature matrix. Take your top five denial reasons from last quarter into every single demo and ask one question: “Where does this get stopped before it’s a denial?”

The vendor with the most specific answer is your vendor. The one who says “our team would catch that in review” is describing your labor cost, not their product.