Let me start by saying something that will annoy people on both sides of this: Brightree is not bad software.

I know that's not the opening you expect from an article with this title. But I've read enough breathless "why we fired our vendor" posts to know they're usually written by someone who never learned the system and wants to blame the tool. That's not this.

We were on Brightree for six years. It did a lot of things well. Payer connectivity was genuinely strong. When we hired billers, some of them already knew it, and that saved us real ramp time. There's a reason it's the incumbent.

We left anyway, and I want to explain why honestly, because "we left" and "it's bad" are different statements and the difference matters if you're making this decision.

Why we actually left

Three reasons, in order of weight.

Cost grew faster than we did. Every time we wanted a capability, it was another module and another line item. The number we'd modeled at signing and the number we were paying at year five were not close relatives. That's partly on me for not pushing harder up front — get the all-in number for everything you'll plausibly need, in writing, on day one.

Our field operation had outgrown the tooling. This was the real driver. Our delivery volume roughly doubled and our process for handling it was still substantially paper and phone calls. Drivers carrying route sheets. Signatures coming back a day or two later, sometimes three. Every one of those lags is a delayed claim, and at volume it stops being a nuisance and becomes a cash flow problem.

The pace of change didn't match ours. Not a criticism of anyone's engineering. Big platform, huge install base, enormous regulatory surface — that combination makes you careful, and careful is slow. We were a mid-size company trying to move fast, and we kept waiting.

None of that is "it's bad." All of it is "it stopped fitting us." If your situation is different — you're large, your ops are stable, the ecosystem matters to you — the same facts could point you the other way, and I'd respect that.

Here's what we evaluated.

1. NikoHealth (where we went)

The thing that won it was delivery, and I want to be specific because "better delivery module" is a phrase every vendor says.

What we needed: drivers with a real mobile app, not a mobile-ish web page. Signature and photo capture at the door. Proof of delivery attaching itself to the order the moment it's captured, not in an overnight batch. Dispatch is able to see where the trucks are and reassign when something changes, which it does daily. Failed deliveries handled as a real workflow instead of a phone call and a sticky note.

That's what hme delivery management should mean, and it's what we got. Our days-from-delivery-to-submission dropped in a way I could see on a graph, which is the kind of before-and-after you rarely get this cleanly.

The second factor was that it's one platform rather than a suite. Intake, inventory, delivery, billing — same system, same data. Our reconciliation work largely disappeared, and reconciliation was a genuine headcount cost we'd stopped noticing because it had always been there.

The third was implementation. I'd braced for a year. It was weeks. I want to be measured here — every migration has rough days and ours did too, mostly around historical data we'd let get messy. But it finished, the team stayed present through go-live, and nobody vanished at month four.

What I'd tell you to check: your payer mix, specifically your regional Medicaid plans, because that's where any platform's connectivity story gets thinnest. Ask them to demo your three worst denial scenarios end to end. And ask for a reference at your size in your product line — not any reference, that one.

The honest trade-off: you give up the sheer breadth of the Brightree third-party ecosystem. For us that was fine, because we used about three of those integrations and two were replaced by native capability. If you've built a serious stack of add-ons, count them honestly before you move.

2. Bonafide

Our runner-up, and a close one. Strong revenue cycle reputation, and the people we talked to who used it were positive about billing and collections specifically.

We chose differently because our biggest pain was operational and field-side rather than financial. If our AR had been the fire, this could easily have been the answer.

3. TIMS Software

Serious, mature, deep — particularly if you're pharmacy-adjacent or mixed-line. We're not, so a lot of its capability would have been paid-for complexity we'd never touch.

Also worth knowing: the interface has history, and new-hire ramp time is real. Factor it in.

4. HDMS (Universal Software Solutions)

Genuinely flexible, and consistently good things said about the team's willingness to work with you on custom requirements. That's not nothing — "can I get a human on the phone" is a real evaluation criterion in this industry.

Our hesitation: configurable systems need an internal owner, and we didn't have someone whose job could be that. Without that person, flexibility becomes drift.

5. CareTend (WellSky)

Strong for infusion and complex clinical-plus-equipment businesses. More system than straightforward HME needs, in my opinion.

If your business genuinely spans clinical and equipment, move it up your list.

6. Staying put and bolting on

We took this seriously for about six weeks: keep the platform, add a delivery and routing tool, add whatever else.

Two things killed it. Integration cost was higher than expected, and ongoing maintenance was worse — every integration is a thing that breaks when either side ships an update, and someone owns that forever. And we'd still have had multiple systems disagreeing about the same orders, which was half the problem.

When it's the right call: you're mid-contract and genuinely can't move this year. It's a bridge, and bridges are fine. Just don't move into the bridge.

What I'd do differently if I ran this evaluation again

Write down why you're leaving before you look at anything. Three specific problems, ranked. We didn't do this until demo four and it made the first three demos nearly useless — we were reacting to features instead of testing against needs.

Demo with your worst data. Not your normal data. Your worst.

Put a driver and a coordinator in the room. Our dispatcher caught a workflow gap in one demo that none of us in management would have seen, because she was the only one who knew what actually happens when a delivery fails at 4pm on a Friday.

Find an off-list reference. Vendor-supplied references are their happiest customers by construction. Ask in a state association or an industry group instead.

Ask the month-five question. "What happens if implementation slips?" The answer tells you how many times they've done this.

The honest summary

If you're large, stable, and deeply invested in an add-on ecosystem, staying may genuinely be right, and switching costs are real.

But if you've been saying "we'll switch next year" for three years in a row — that's not a software evaluation anymore. That's avoidance. And the migration doesn't get easier the longer you wait; you just accumulate more data to move and more workarounds to unlearn.

Write down your three reasons. Book the demos. Bring your dispatcher.