Speed to market matters just as much in medtech as any other industry. Funding rounds are priced on regulatory milestones, competitors announce Breakthrough Designations in press releases, and FDA itself has spent the last several years building programs whose entire purpose is to get promising devices to patients faster. If you're an early-stage founder or a lean QA/RA team, the gravitational pull is obvious: find the pathway, win the designation, compress the timeline.
But there's a problem. Teams chasing an expedited pathway start treating the pathway as the hard part and quality as something they'll sort out later. The designation becomes the goal, the quality system becomes a box for some future version of the company to check, probably right before the FDA 510(k) submission goes in.
That logic is backwards, and I want to explain why using a conversation I just had with Mike Drues on the podcast. Mike has spent more than 30 years in this industry, consults for manufacturers and for FDA itself, and the episode was nominally about knowing what you know and knowing what you don't. What it turned into was a tour of the programs early-stage teams hear about but rarely understand, and a fresh warning-letter case study that shows what "we'll sort quality out later" costs. My conclusion from that conversation is the thesis of this piece: an expedited pathway does not remove the need for a working quality foundation. It raises the cost of not having one, because expedited review means less runway to fix gaps later.
The lesser understood parts of the regulation
Mike opened with what he calls one of his favorite sections of FDA's quality framework: the preamble, where the agency explains not what to do but why. He highlighted three ideas from it. Manufacturers should use good judgment when developing their quality systems and apply the sections applicable to their specific products and operations. It's the responsibility of each manufacturer, not FDA, to establish requirements that result in safe and effective devices. And the responsibility for meeting those requirements may not be delegated, even though the actual work may be delegated.
Those three sentences are the entire premise of premarket quality, and they predate every expedited program by decades. FDA never promised to tell you what your quality system needs. It told you the opposite: figure it out, right-size it to your device, and own the result. No designation changes that.
What "later" actually costs: a warning letter from last month
Mike brought a case study that should sober anyone tempted to defer the fundamentals. In early April, FDA issued what he described as the first warning letter for GMP violations related to the use of AI in the manufacturing process. The company had used AI to generate product specifications, procedures, and production records, and skipped process validation. When FDA asked why, the company's response, straight from the published letter, was that they were not aware of the legal requirement, because the AI agent never told them.
So a manufacturer of products that go into patients outsourced its regulatory awareness to a chatbot, and the chatbot defined the whole compliance strategy. Mike's take was that AI is a fine tool for a rough draft, but that should be your starting point. It shouldn't be your stopping point. My take was that the AI didn't fail at all. The quality unit failed, because the regulation puts approval responsibility on the quality unit, and a quality unit that consists of unreviewed AI output is not a quality unit. Think about it: what if you hired an intern to write your SOPs and own your quality? Did that intern fail, or was the failure on the part of the organization that made the decision to outsource quality to an intern?
The same episode surfaced a second data point in the same spirit: FDA recently reported that more than 2,200 device manufacturers, nearly a third of those required, had failed to publish clinical trial results to clinicaltrials.gov. These are not exotic failures. They are basic obligations that companies missed because nobody in the building knew, or nobody owned the knowing. That is the unknown-unknowns problem in its purest form, and it is exactly the failure mode that speed-pressure makes worse.
The programs people like to name-drop
Here's what each expedited program actually is, and where quality gets skipped in each one. Mike walked through the first two on the episode; the next two we ran out of time for, so the summaries are mine, and they are the subject of the part two we already agreed to record.
Breakthrough Device Designation
A voluntary program for devices and device-led combination products that are reasonably expected to provide more effective treatment or diagnosis of a life-threatening or irreversibly debilitating disease or condition. Two things founders routinely get wrong, according to Mike. First, it's part of your marketing submission, not in place of it. He has customers ask whether a Breakthrough designation means they can skip the 510(k) or PMA, and the answer is no. If you are fuzzy on the difference between those submission types, start with our 510(k) vs. PMA glossary entry. Second, the emphasis is efficacy. You can have safety improvements inside a Breakthrough device, but efficacy is what the designation is built around.
What the designation buys you is interaction: faster, more frequent engagement with FDA, including Sprint discussions in place of a standard pre-submission. Mike mentioned he was preparing exactly that with a Breakthrough client the day before we recorded. Notice what that means for quality. More touchpoints with FDA earlier means your design controls, your risk documentation, and your data justifications get looked at sooner and more often, not less. A Sprint discussion built on a shaky quality foundation is a faster way to expose the shakiness.
Safer Technologies Program (STeP)
The sister program, also voluntary, for devices reasonably expected to significantly improve the safety of a currently available treatment or diagnostic. The two differences from Breakthrough, per Mike: the emphasis is safety rather than efficacy, and the condition does not need to be life-threatening or irreversibly debilitating. The guidance says a device cannot hold both designations, and Mike's regulatory-logic footnote here is a great point: the same physical device with different claims is, in the eyes of the law, two different devices. One version labeled around efficacy claims and one around safety claims can each pursue its own path. Claims drive everything, which means your labeling, your intended use, and the evidence behind them need to be disciplined from day one. That discipline is a quality-system function.
Total Product Life Cycle Advisory Program (TAP)
This one is mine, since we hit time before Mike could get to it. TAP is FDA's program for giving enrolled devices, drawn from the Breakthrough pool, earlier and more strategic engagement across the whole product life cycle, including input from stakeholders like clinicians and payors, with the explicit goal of shortening the distance from concept to commercialization. The quality trap is the same as Breakthrough's but amplified: TAP front-loads FDA engagement into the phase where early-stage teams have historically had the least documentation rigor. If your design and development records are being improvised, the program designed to accelerate you becomes a recurring appointment to demonstrate that.
Early Feasibility Study (EFS) pathway
Also mine. The EFS pathway lets you run a small early clinical study, under an investigational device exemption, before your device design is finalized, with flexibility to iterate on the design during the study. It is a genuinely valuable way to learn from real clinical use early. It's also the pathway where deferring quality is most tempting and most dangerous, because "the design isn't final" gets misread as "design controls don't apply yet." They do. An EFS is a clinical study on human beings, and the documentation, risk management, and change control around every design iteration are precisely the evidence FDA and your future self will need. For more on how that pathway works in practice, see our guide to the Investigational Device Exemption process. Skipping the documentation doesn't save time. It creates a gap you will reconstruct later, from memory, under deadline.
And one the pillar map doesn't cover yet: the TEMPO pilot. Mike spent real time on the Technology Enabled Meaningful Patient Outcomes program, a joint FDA and CMS pilot for certain digital health devices aimed at improving patient outcomes in specific condition areas. His shorthand for it was elegant: premarket real-world evidence. Qualify, and you can ask FDA to exercise enforcement discretion, lowering some requirements that would normally apply. But listen to the second half of what he said. Enforcement discretion holds, in his words, unless and until companies start doing stupid things, and when they do, FDA starts regulating again. His parallel was laboratory-developed tests, where decades of discretion tightened after Theranos. A pathway with reduced oversight is not a pathway with reduced responsibility. It is one where the QMSR preamble's "good judgment" burden falls entirely on you, with nobody checking the homework until something goes wrong publicly.
The pattern across all five
Line them up and the pattern is hard to miss. Every one of these programs trades time for scrutiny concentration. Breakthrough and TAP concentrate FDA's attention earlier. STeP concentrates everything into your claims discipline. EFS concentrates design-control obligations into the phase teams are least prepared for. TEMPO removes the external check and makes your internal one the only one.
None of them removes a single quality obligation. The QMSR preamble was clear about that before any of these programs existed: the responsibility is the manufacturer's and may not be delegated, not to a consultant, not to a designation, and, as one company found out in April, not to an AI.
This is where I'd like to mention what a working quality foundation looks like at this stage. What the best teams use is a right-sized quality management system you actually operate: design controls that capture decisions as you make them, risk management that lives alongside development instead of after it, and document control that means your Sprint discussion or EFS submission is an export, not an archaeology project. That is the right-sized QMS idea we've written about before, and it's the difference between an expedited pathway that compounds your speed and one that compounds your gaps.
BONUS RESOURCE: How to right-size your QMS so it accelerates you instead of slowing you down
Know what you don't know, then fix the problems
Mike ended the episode with Socrates: true knowledge is knowing what you know, knowing what you don't know, and knowing the difference between the two. For an early-stage medtech team, the honest version of that exercise looks like this. Do you know which of these programs your device could actually qualify for, and what claims discipline that requires? Do you know what FDA will expect to see at a Sprint discussion or in an EFS submission, and does your documentation exist somewhere other than someone's laptop? Do you know which quality obligations apply to you right now, or are you waiting for something, a consultant, a designation, an AI, to tell you?
If any of those answers is "we'll figure it out when we get there," the expedited pathway you are chasing will get you there faster than you're ready for. The teams that actually convert these programs into speed are the ones whose quality foundation was built before the clock started, sized to their stage, and operated for real. Everything about your FDA 510(k) submission and the FDA clearance at the end of it gets easier when that is true, and every one of these programs gets riskier when it is not.
Keep reading
- Exploring Breakthrough Device Designation: what the designation actually buys you, and where founders misjudge it.
- Is FDA's Safer Technologies Program applicable to your device?: how STeP differs from Breakthrough and who qualifies.
- Understanding the Investigational Device Exemption (IDE) process: the mechanics behind the EFS pathway.
- A right-sized QMS: what does your pre-market team really need?: the four must-haves for an early-stage quality foundation.
- FDA cleared vs. approved vs. granted: the terminology every submission strategy depends on.
- 510(k) vs. PMA: what's the difference?: the two most common submission types, compared.
Greenlight Guru is purpose-built to keep design controls, risk management, and document control operating together instead of scattered across spreadsheets, so your quality foundation is ready whichever pathway you pursue. See a demo of Greenlight Guru and bring your current pathway strategy. We'll show you what a right-sized quality system looks like for your stage.
BONUS RESOURCE: Change Impact Analysis Checklist
