Latest / The Operations Podcast with Fexingo: Process, People, and Profit in Modern Business / How a Medical Lab Cut Test Turnaround Time by 60 Percent
Transcript
- Lucas: A diagnostic lab in Cleveland — let's call it the MetroLab network — was running about 10,000 blood tests a day across three sites. And the average turnaround from sample intake to result reporting was hovering around 48 hours. That meant a doctor ordering a CBC at 9 a.m. on Monday might not see the result until Wednesday morning. Luna: And in a lot of clinical settings, that delay can mean keeping a patient in the hospital an extra day, or starting a treatment without full information. Lucas: Exactly. So MetroLab decided to treat their entire operation as a lean project. They brought in a team from a university medical school — the kind of people who study factory floor efficiency — and they mapped every single step of the process from the moment a phlebotomist draws blood to the moment the result hits the electronic health record. Luna: What was the biggest time sink they found? Lucas: Batching, by far. Samples would come in throughout the day, but the lab would accumulate them and run them in fixed batches every two hours. So a sample that arrived at one minute past the cutoff had to wait almost two hours just to enter the testing queue. And then, within each test type, they'd batch again — 50 samples at a time on certain analyzers. Luna: So the classic batch and queue problem that Toyota eliminated decades ago. Lucas: Right. The MetroLab team shifted to continuous flow wherever possible. For the high-volume tests — complete blood counts, basic metabolic panels — they reconfigured the analyzers to run as soon as a single sample was prepped. That alone cut the average wait from 110 minutes to under 8. Luna: Eight minutes, that's a different world. But I imagine piece-flow requires more frequent loading and unloading of the machines. Didn't that increase technician workload? Lucas: It did change the work pattern. But they compensated by cross-training every technician on at least three different analyzers. Previously, each tech specialized in one machine. So if the hematology analyzer had a backlog, the person running it was swamped while the chemistry analyzer operator was idle. After cross-training, they could shift people dynamically. Luna: That also reduces the risk of a single point of failure. If one tech calls in sick, the whole line doesn't stall. Lucas: Exactly. And they redesigned the physical layout of the lab. The old layout had sample accessioning in one corner, centrifuges in another, analyzers in a third, and result verification in a fourth. Technicians were walking an average of 2.1 miles per shift just moving between stations. Luna: That's almost the length of a golf course. So they brought everything closer? Lucas: They created U-shaped cells. In each cell, you have a centrifuge, an analyzer, and a verification terminal within arm's reach. For the most common test panels, they built dedicated cells. The technician can now process a sample without taking more than three steps. Walking distance dropped by 40 percent, and that freed up about an hour per tech per shift. Luna: That hour probably went back into handling more tests or doing quality checks. Lucas: Yes, and they also automated the manual transcription of results into the EHR. Previously, technicians had to type values into a separate system. That introduced keystroke errors. They integrated the analyzers' output directly with the lab information system, cutting out the intermediate step. Error rates on reported values fell by 80 percent. Luna: So faster turnaround plus higher accuracy. That's a rare combination — usually you trade one for the other. Lucas: The results after six months: average turnaround dropped from 48 hours to 18.7 hours. That's a 61 percent reduction. And they were able to handle 15 percent more tests without adding any new equipment or staff. The project paid for itself in eight months just from reduced patient days and better resource utilization. Luna: You know, it's interesting — if these kinds of operational conversations have helped you think about your own work, maybe in a small way, I just want to mention that this show stays ad-free because a handful of listeners chip in. It's at buy me a coffee dot com slash fexingo, and it's literally what funds making this many episodes. Lucas: Yeah, and it's a low-key thing — no pitches, no pressure. But for anyone who's gotten something out of these deep dives, that link helps keep the lights on. And the support we've already gotten has been genuinely meaningful. Luna: Absolutely. Now, one thing that surprised me about the Cleveland lab was that they didn't buy any new analyzers. The entire improvement came from rearranging people and processes. Lucas: That's the real lesson here. Too often, organizations think operational improvement requires a capital expenditure. MetroLab proved that tweaking workflow, layout, and staffing can deliver a step-change in performance. And they sustained it — after two years, turnaround is still under 20 hours. Luna: They also reported that technician job satisfaction went up. Less walking, less waiting, more variety in tasks. That's a nice side effect. Lucas: It's a virtuous cycle. When the process works well, the people feel better, and when they feel better, they find more improvements. The lab now holds monthly kaizen events where techs propose changes. Eleven of those suggestions have been implemented in the past year. Luna: So the culture shifted from 'we work in a lab' to 'we run a lab as a system.' Lucas: Exactly. And I think that's the lasting takeaway — operational excellence isn't a one-time fix. It's a mindset. MetroLab went from a batch and queue organization to a continuous-flow organization, and they're still finding ways to improve. Luna: For anyone listening who's in a high-volume service operation, what would be the first thing they should look at to diagnose their own batch and queue problems? Lucas: I'd say walk the full path of a single unit of work. Whether it's a blood sample, an insurance claim, or a customer service ticket, literally follow one from start to finish and measure the time it spends being worked on versus waiting. You'll almost always find that the waiting time is 90 percent of the total. Luna: That's what Toyota calls the 'value-added versus non-value-added' split. And the non-value-added is usually the gold mine. Lucas: Right. So for MetroLab, the value-added time — actual testing and verification — was about 45 minutes. The rest was waiting. By compressing that waiting, they compressed the whole process. Luna: A 45-minute process stretched over 48 hours — and they fixed it without buying a single new machine. That's pretty remarkable. Lucas: It's a great case study in how operations principles apply far beyond the factory floor. And it's a reminder that sometimes the biggest gains come from looking at the flow, not just the technology. Luna: Well said. And with that, we'll wrap up this episode of The Operations Podcast. Lucas: Thanks for listening.