MedTech burn rate is determined by regulatory milestones, not feature releases. The runway that matters is the distance between where your cash sits today and the next certifiable milestone under the MDR. Technical file lock, Notified Body submission, audit close, CE certificate issuance. A burn rate plan that is paced to software-style quarterly releases will run out of cash between milestones. A burn rate plan that is paced to the regulatory critical path reaches CE marking without a gap round and keeps the cap table intact.

By Felix Lenhard and Tibor Zechmeister. Last updated 10 April 2026.

TL;DR

  • Burn rate in a MedTech startup is the monthly cash-out number, but the metric that actually matters is the cash distance to the next regulatory milestone under Regulation (EU) 2017/745.
  • Regulatory milestones. Technical file lock, Notified Body submission, audit close, certificate issuance. Are the valuation inflection points. Burn must be paced so that cash lands you on the far side of one of them, never in the gap between two.
  • The gap round is the bridge round you have to raise because your burn plan stopped one milestone short. It is the most expensive capital a MedTech startup ever takes.
  • Slowing burn is a regulatory strategy decision, not a cost-cutting exercise. You slow by subtracting work that does not trace to an MDR article. You never slow by cutting work that is load-bearing for the next milestone.
  • If you cannot fund the runway to the next milestone plus a contingency buffer, you do not have a burn rate problem. You have a round-size problem, and the fix is in the round, not in the spreadsheet.

Why regulatory milestones drive burn, not feature releases

A software startup measures burn against product velocity. Ship a feature, measure the lift, iterate. The burn rate conversation is a conversation about throughput. If the team is shipping, the burn is doing its job.

A MedTech startup measures burn against a completely different clock. The clock is set by the MDR, and it does not care how fast the team is shipping. Features that are not in the locked technical file are not yet on the device the Notified Body is assessing. Work that is not traceable to an Annex I general safety and performance requirement, an Article 10 manufacturer obligation, an Article 61 clinical evaluation, an Article 52 conformity assessment step, or an Article 83 post-market surveillance requirement is work that does not move the CE marking date. It may be useful work. It may be important work. It is not work that the burn rate is paying for.

Felix has a story he tells at the start of every coaching session with a first-time MedTech founder who is running the financial model on software-industry reflexes. A team arrived confident that they could pace their regulatory project the way they had paced their previous SaaS company. Quarterly releases, monthly burn reviews, feature prioritisation based on user feedback. They had raised a round that covered twelve months at the current burn. The plan was for the CE certificate to land at month nine, revenue to follow at month ten, and the round to carry the company to cash-flow positive. Month nine came and went with the technical file still in the first Notified Body review cycle. Month twelve came and went with a bridge round being raised on worse terms. The product worked. The burn plan was paced to the wrong clock.

Tibor sees the other side of the same pattern from the audit room. The startups that run out of cash mid-certification are almost never the ones whose devices fail the assessment. They are the ones whose burn plan assumed the assessment would finish on a date that was set by the founders rather than by the regulatory critical path. Burn paced to feature releases produces a runway that runs out in the middle of a review cycle. Burn paced to milestones produces a runway that lands the company on the safe side of an inflection point, where the next round can be raised from a position of strength.

The reframe is simple. Your burn rate is not a monthly number in isolation. It is a monthly number divided into the cash distance to the next regulatory milestone. If that ratio buys you the milestone plus a buffer, the burn is under control. If it does not, the burn is out of control even if the monthly number looks reasonable.

How to match spend to milestones

The milestones that matter for burn management are the ones a sophisticated investor will pay for. For a typical Class IIa software device under the MDR, the sequence is:

  1. Purpose and classification locked. Intended purpose written, classification defensible, scope of the device frozen. Cheap to reach, but load-bearing for everything downstream.
  2. QMS operational. A quality management system compliant with EN ISO 13485:2016+A11:2021 and the manufacturer obligations in Article 10, actually running rather than just documented.
  3. Technical file lock. The Annex II documentation package closed and internally reviewed. This is the point at which the scope of the Notified Body review is set.
  4. Clinical evaluation signed off. The clinical evaluation report under Article 61 and Annex XIV complete, with sufficient clinical evidence for the intended purpose.
  5. Testing complete. Every test required by the applicable harmonised standards finished and reports in hand.
  6. Notified Body submission. Application filed with the Notified Body under Article 52, review cycle started.
  7. Audit close. QMS audit and technical documentation assessment finished, findings closed, recommendation for certification in hand.
  8. CE certificate issued. The certificate itself, the point at which the device can be placed on the market.

Each of these is a valuation inflection point. Each of them, if reached with cash still in the bank, is a moment where a follow-on round can be raised on better terms than the previous round. The burn plan's job is to sequence spend so that the cash runs out after one of these milestones, never between them.

The practical rule is the milestone plus buffer rule. Identify the next milestone on the list. Estimate the cost to reach it, with the Felix doubling heuristic applied. Add a 20 to 30% contingency buffer for the things that always go differently than planned. That is the minimum cash position the company needs in the bank today to safely target that milestone. If the current cash is below that number, the burn has to slow, the milestone has to move closer, or the round has to be re-opened. There is no fourth option that does not involve a gap round later.

Stretching runway without cutting regulatory quality

Stretching runway in a MedTech startup is not the same as cutting costs in a software startup. Cutting in the wrong place makes the certification date later, not earlier, and later certification costs more runway than the cut saved.

The correct places to stretch are the ones where the Subtract to Ship framework finds work that is not load-bearing for any MDR obligation.

Template and documentation bloat. SOPs written for processes the company does not actually run. Risk analyses that enumerate hazards that cannot occur with the device. Technical file sections that exist because a template said they should, not because Annex II requires them. This is real money and real calendar time, and cutting it does not move the certification date by a single day.

Consulting hours on documents nobody will read. A consultant writing a 200-page document that the team will not maintain is a consultant producing a liability. When the next surveillance audit comes around, the document has to be current. If it is not, it produces a finding. Cutting these hours is a pure saving.

Parallel evidence chains. Two testing programmes where one would be sufficient. Two clinical data sources where one well-documented pathway covers the intended purpose. The Evidence Pass of the Subtract to Ship framework exists to find these.

Premature tooling. Elaborate project management infrastructure, custom dashboards, branded templates, and other work that exists to make the project look organised rather than to actually move a milestone closer.

The places you do not stretch are the ones where spending less produces findings later.

Clinical evidence work. Every euro here is load-bearing.

Testing lab fees for anything required by the applicable harmonised standards. The test has to happen. Running it later just moves the bill.

Notified Body fees. There is no lean version of the certification fee itself, and delaying the submission to save cash this month costs more cash next month in extended team burn.

PRRC qualification. The person has to be qualified. Downgrading this line produces findings at the first audit.

PMS build. The post-market surveillance system required by Article 83 has to exist at certification. Deferring it is deferring a finding.

The bridge round trap

The bridge round is the round a MedTech founder raises when the first round's runway stops one milestone short. It is the most expensive capital a startup takes. The terms are worse than the previous round because the position is worse. The valuation is flat at best and often down. The dilution on the next round compounds the damage. And the bridge round has to be raised at the exact moment when the founder has the least leverage. Mid-certification, between milestones, with no new proof point to show.

The bridge round trap is a burn rate failure that shows up as a fundraising failure. The spreadsheet said the runway covered twelve months. The burn stayed at the planned number. The certification took longer than the plan assumed, or the clinical evaluation needed another iteration, or a Notified Body finding triggered a re-test cycle. None of these are unusual events. They are the normal shape of a MedTech project, and a burn plan that does not budget for them is a burn plan that will need a bridge round.

The defence against the trap is the contingency buffer and the milestone alignment. The buffer absorbs the normal unknowns. The milestone alignment ensures that when the buffer runs out, the company is on the safe side of an inflection point where the next round can be raised properly. A burn plan that has both of these is a plan that does not need a bridge round. A burn plan that has neither is a plan that will.

When to slow and when to accelerate

There is a time to slow burn and a time to accelerate it. Getting the direction right matters more than the absolute number.

Slow when the work in front of the team is not on the regulatory critical path. Between milestone gates, when the next work item is internal preparation rather than external dependency, a lower burn rate preserves optionality. A smaller team can make steady progress on QMS build, risk management iteration, and clinical literature review without paying for capacity that is waiting for an external gate to open.

Accelerate when the critical path is blocked on your speed. Once the Notified Body submission window is open, or once the testing lab has a slot, or once the clinical evaluation review is waiting on a data analysis, the burn rate should rise. Not because you are spending faster for its own sake, but because calendar time is the expensive resource and throwing cash at it to compress the timeline is cheaper than extending the runway to cover a slower version of the same work.

The mistake is to do the opposite. To accelerate during the waiting phases because it feels productive, and then to slow during the critical-path phases because the cash is running low. This is the shape of burn that produces the worst outcomes. Cash gets spent on work that does not move the milestone, and then there is not enough left to finish the work that does.

Common mistakes founders make

Measuring burn as a monthly number without a milestone denominator. A burn of 80,000 a month is neither good nor bad. A burn that buys the next milestone plus a buffer is good. A burn that does not is bad.

Pacing spend to quarterly reviews instead of regulatory gates. The quarterly board review cadence is a reporting rhythm. It is not the cadence the regulation runs on.

Treating post-market surveillance as a post-launch cost. Article 83 requires the PMS system to exist at certification. Deferring it is not a saving; it is a finding waiting to happen.

Cutting load-bearing work to extend runway. Every euro cut from clinical evidence, testing, or risk management produces a later finding that costs more than the saving.

Keeping a full team through the waiting phases. A team sized for peak throughput is burning cash during the phases where peak throughput is not possible. Variable external capacity is often the right answer for the waiting phases.

Raising round size based on the founder's optimistic timeline rather than the honest milestone plus buffer calculation. This is the root cause of most gap rounds. If the round is too small to reach the next milestone, the burn plan cannot fix it. Only a bigger round can.

The Subtract to Ship angle

The Subtract to Ship position on burn rate is the same position it takes on regulatory spending. The goal is not to burn less. The goal is to burn correctly. Every euro of burn must trace to work that moves a regulatory milestone closer. If it does, fund it at the pace the milestone demands. If it does not, cut it entirely.

The framework's four passes each reduce burn in a specific way. The Purpose Pass prevents burn on work aimed at an intended purpose that will not survive review. The Classification Pass prevents burn on conformity assessment work for a higher class than the device actually is. The Evidence Pass prevents burn on parallel clinical data chains when a single well-documented pathway suffices. The Operations Pass prevents burn on QMS and documentation bloat that does not match real operations.

A startup that runs the four passes honestly typically finds that the honest burn rate is lower than the original plan, the honest runway is longer than the original plan, and the next milestone is reachable without a bridge round. That is the outcome the framework is built to produce. It is not a cost-cutting exercise. It is a regulatory strategy exercise with a cash-flow consequence.

Reality Check. Where do you stand?

  1. What is your current monthly burn rate, and what is the cash distance from today to the next regulatory milestone on your critical path?
  2. Does your runway cover the next milestone plus a 20 to 30% contingency buffer, or does it end somewhere between two milestones?
  3. If you listed every line item in your monthly burn, could you trace each one to a specific MDR article, annex, or harmonised standard it is helping you satisfy?
  4. Have you identified the work in front of the team right now that is on the critical path versus the work that is waiting on an external gate?
  5. Is your team sized for the current phase, or is it sized for the peak phase and burning cash during a waiting phase?
  6. Is the PMS system build in your burn plan for the months before certification, or have you mentally deferred it to after launch?
  7. If your next milestone slipped by two months, would your cash still carry you past it, or would that slip trigger a bridge round?
  8. Is the round you last raised large enough to reach the next inflection point, or did it close at a number that was always going to need a bridge?

Frequently Asked Questions

What is a healthy monthly burn rate for a Class IIa MedTech startup? There is no single healthy number. The healthy burn is the one that buys the next regulatory milestone plus a 20 to 30% contingency buffer within the current cash position. For a typical Class IIa software startup with 18 to 24 months to CE marking, monthly burn rates of 60,000 to 150,000 are common depending on team size and geography, but the number is only healthy if the runway lands the company on the safe side of an inflection point.

How do I know whether to slow burn or raise more? Work out the honest cost of reaching the next milestone, doubled from a naive estimate, with a 20 to 30% contingency on top. Compare that to the cash in the bank. If the cash covers it, the burn is fine. If it does not, slowing the burn will usually not close the gap. The round was probably too small in the first place, and the fix is to re-open the round or change the round size in the next raise rather than to cut load-bearing work.

What is a gap round and why is it so expensive? A gap round is a bridge round raised between two proper milestones because the previous round's runway stopped short of the next inflection point. It is expensive because the company is mid-certification, has no new proof point to show, and the terms reflect that weak position. Flat valuations are the best case, down rounds are common, and the dilution damage compounds on every subsequent round.

Can I cut regulatory spending to extend runway? Only if you cut work that does not trace to an MDR obligation. Template bloat, unused SOPs, consulting hours on documents nobody will maintain, and parallel evidence chains can all be cut without touching any regulatory requirement. Load-bearing work. Clinical evidence, testing, Notified Body fees, PRRC, PMS, risk management. Cannot be cut without producing findings that cost more than the saving.

Should post-market surveillance be in my burn plan before certification? Yes. MDR Article 83 requires the PMS system to exist at certification and to be operational from day one post-launch. That means the build happens in the months before the certificate is issued, and the build cost belongs in the pre-certification burn plan. Not deferred to a vague "after launch" line.

Sources

  1. Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on medical devices. Article 10 (manufacturer obligations), Article 52 (conformity assessment procedures), Article 61 (clinical evaluation), Article 83 (PMS system). Official Journal L 117, 5.5.2017.
  2. EN ISO 13485:2016 + A11:2021. Medical devices. Quality management systems. Requirements for regulatory purposes.
  3. EN ISO 14971:2019 + A11:2021. Medical devices. Application of risk management to medical devices.

This post is part of the Funding, Business Models and Reimbursement series in the Subtract to Ship: MDR blog. Authored by Felix Lenhard and Tibor Zechmeister. Burn rate is not a monthly number in isolation. It is a cash distance to the next regulatory milestone, and the plan that survives the Notified Body is the plan that was paced to that clock from the beginning.