We got a call from a mid-sized hospital network last year. Their radiology director was frustrated—new MRI installed, top-of-the-line, 3T, the works. Yet their patient throughput was barely higher than the 1.5T they'd replaced. They assumed a hardware issue. Maybe a bad install. Maybe our team messed something up.
I spent a week with their team. I don't have hard data on industry-wide waitlist causes, but based on reviewing about 200+ imaging-related contracts annually for the past four years, I can tell you what I keep seeing: it's almost never the scanner. The machine works. The problem is everything around it.
The Surface Problem: Slow Patient Throughput
The question we got was straightforward: "Why is our new MRI slower than the old one?" They had metrics. Average scan time was within spec. But the time between patients? That was the killer. 30 minutes between exams when the protocol only needed 20 minutes of scan time. Over an 8-hour day, that's four lost slots. Four patients. Every day.
Their immediate assumption: the scanner software. Or onboarding. Or some setting we'd missed. It wasn't an unreasonable guess. When a new machine underperforms, the vendor is the first place you look. I get that. I've rejected equipment deliveries for smaller deviations—once flagged a coil connector with a 0.3mm tolerance variance that the vendor insisted was 'within industry standard.' Spoiler: it was out of our spec, and they replaced it.
But this wasn't a hardware or software spec issue. The MRI data was clean. The platform logs showed the system was ready for the next patient within 5 minutes of the previous one finishing. The problem was the 25 minutes between 'ready' and 'next patient prepped.'
The surprise wasn't the hardware. It was how many people and processes touched that gap.
Deep Root Cause 1: The Workflow Friction (That No One Owns)
The MRI itself is a beautiful piece of engineering. But a scanner doesn't schedule patients. It doesn't prep them. It doesn't move them from the waiting room. It doesn't handle the contrast injection, the breath-hold instructions, the anxiety management, the post-scan recovery.
What I found was a classic gap: the radiology team owned the scan. The nursing team owned patient prep. The front desk owned scheduling. No one owned the handoff between them. The 'readiness' signal from the scanner went to the technologist, then the technologist had to physically walk to the waiting area to find the next patient. Who was ready? No one. Their paperwork wasn't complete. Or they hadn't been changed into a gown. Or the referring doctor's order was missing a detail.
That's not a scanner problem. That's a workflow problem. And it's way more common than you'd think.
I still kick myself for not flagging this earlier in the engagement. If I'd spent more time on the floor before the install, I'd have seen that the new scanner's speed just exposed an existing bottleneck. The old scanner was slow enough that the staff could keep up. The new one just made the gap visible.
Deep Root Cause 2: The Disconnected Tech Ecosystem
Here's something I've noticed across multiple hospital network audits: the imaging equipment is often the most advanced piece of technology in the room. Everything around it—the patient transfer devices, the scheduling system, the PACS, the EMR—runs on a different pace of innovation.
In this particular case, they had a brand new Philips MRI. The patient transfer device (the table that moves the patient from the stretcher to the scanner) was from a different vendor. Not incompatible, but not optimized. The contrast injection system was a third-party unit that required a separate control interface. The technologist was flipping between three screens to manage one scan.
The efficiency loss was incremental—maybe 2-3 minutes per patient—but multiplied across 15-20 patients a day, that's 30-60 minutes of lost capacity. Simple.
I wish I had tracked that specific metric more carefully from the start. What I can say anecdotally is that when we mapped the full patient journey, the scanner itself was idle less than 15% of the total 'patient appointment' time. The other 85% was prep, transfer, recovery, and documentation.
The solution wasn't a faster MRI. It was designing the room and the workflow around the MRI to eliminate friction points.
Deep Root Cause 3: The Training Gap (That Isn't A Training Gap)
The hospital network had trained their technologists on the new system. Standard onboarding. Two days of training from our clinical education team. Pass rates were fine.
But training happened at the console. In a lab setting. With simulated patients. The technologist learned how to run the scan perfectly. What they didn't learn was how to run the scan while also managing a real patient, handling a contrast reaction, and responding to a nurse's question about the next case. In other words, they learned the system in isolation, not in context.
I saw this because I sat in on a post-install follow-up (something I started doing after a $22,000 redo on a different project because we'd missed a simple communication step). The technologists were technically competent. They struggled with the situational competence—the ability to adapt the new tool to real-world chaos.
Never expected that the gap between training and real-world performance would be so large. Turns out it's not about knowing what to do. It's about knowing what to do when something unexpected happens. That's the part no training manual covers well.
The Cost of Not Solving This
Back to that hospital network. The cost of the slow throughput wasn't just patient frustration (though that was real). It was financial. Four lost slots a day, at an average reimbursement of about $1,500 per MRI (depending on insurance mix), means $6,000 a day in lost revenue. Over a year of operation (let's say 250 days), that's $1.5 million in unrealized revenue.
The new MRI cost about $1.8 million installed. So the throughput gap alone was eating almost the entire ROI of the machine in year one. They bought a Ferrari but parked it in traffic.
I don't have hard data on how many hospitals face this exact situation, but based on our audits, I'd guess it's a significant minority—maybe 20-30% of new high-end MRI installations underperform for non-hardware reasons. That's a lot of lost value.
The Actionable Part (Kept Short)
So what actually fixed it? Three things:
- Workflow mapping. We sat down with the radiology director and mapped the entire patient journey—not just the scan. Found six handoff points where time was lost. Fixed four of them with simple process changes (no new tech required). The other two required a software integration that we sourced through a partner.
- Better partnerships. The patient transfer device and contrast injector? They were from different vendors. We worked with those vendors—and brought in a third-party integration specialist—to unify the control interfaces. The technologist now manages all three from one screen. That alone saved 2-3 minutes per patient.
- Contextual training. We redesigned the training to include simulated interruptions and real-world distractions. The difference was way bigger than I expected—pass rates on a practical assessment improved by 34% after the change.
The result? Three months after the changes, the same MRI was scanning 18 patients a day instead of 14. The director stopped calling us. Then they ordered a second system for their sister hospital.
The lesson I keep coming back to is this: the best imaging equipment in the world underperforms if the ecosystem around it isn't designed to support it. And that ecosystem isn't just hardware—it's workflow, training, integration, and partnership. Period.
The next time you hear about an MRI waitlist problem, ask whether the scanner is really the bottleneck. It probably isn't. The answer is usually more boring—and more fixable—than you think.
