Financial modeling for a MedTech startup is not a SaaS model with a medical label. Revenue starts the quarter after CE mark, not the quarter after product-market fit, and the CE mark date depends on a notified body queue you do not control. Every assumption in the model has to bend around MDR Article 52 conformity assessment timelines.

By Tibor Zechmeister and Felix Lenhard.

TL;DR

  • MedTech revenue projections must anchor on the CE mark date, not the prototype date. Nothing ships before conformity assessment closes.
  • Notified body queues for Class IIa, IIb, and III devices under MDR Article 52 and Annex IX routinely add 6 to 18 months of idle calendar time to the critical path.
  • A credible model carries an explicit regulatory delay sensitivity: base case, plus a 6-month and a 12-month slip, each with the runway impact calculated.
  • Post-CE revenue ramps slowly in Europe because reimbursement is national and hospital procurement cycles are long. Plan for flat quarters after the certificate.
  • Runway buffer below nine months at the moment of notified body submission is not a lean startup. It is a bankruptcy countdown.
  • Investors who fund MedTech know these curves. A model that pretends MDR does not exist reads as naive and loses the room.

Why this matters

There is a spreadsheet pattern that kills MedTech startups. The founder opens a SaaS financial model template, renames the product, scales the assumptions, and presents it to investors. Revenue starts in month nine. Growth compounds monthly. Gross margin hits seventy percent by year two. The narrative feels clean.

None of it is possible. A MedTech startup with a Class IIa or higher device cannot sell in the European Union without a CE mark, and the CE mark cannot exist without a notified body certificate under MDR Article 52 and the conformity assessment routes in Annex IX through XI. The certificate has a calendar cost that does not respond to hustle. You can be the best founder in Europe and still wait six months for a stage 1 audit slot.

Financial models that hide this reality do not survive the first serious investor conversation. Models that confront it become a strategic tool. The model is where regulatory reality and business ambition meet, and the meeting has to be honest.

What MDR actually says

MDR is not a financial regulation, but three provisions anchor every cash flow assumption in the model.

MDR Article 52 — conformity assessment. For devices above Class I, the manufacturer cannot self-declare. A notified body must assess the quality management system and the technical documentation, and issue a certificate. The article sets out the routes. It does not set out how long they take, because that depends on the body, the device, and the queue.

MDR Annex IX — full quality assurance plus technical documentation assessment. This is the typical route for Class IIa, IIb, and III devices. It requires a QMS audit conforming to MDR Article 10(9), which for practical purposes means EN ISO 13485:2016+A11:2021 plus MDR-specific elements. It also requires technical documentation review sampled by class and risk. Each of those steps has a real duration.

MDR Article 120 and Regulation (EU) 2023/607. The extended transition provisions keep legacy MDD certificates valid under specific conditions until 2027 or 2028 depending on class. They do not help a new startup with a new device. New devices need an MDR certificate from day one, and the new device queue is the tight queue.

Plain language translation for the model: there is a mandatory regulatory step between finished development and first revenue, that step is done by a third party whose calendar you do not own, and the duration of that step is an assumption you must declare openly.

A worked example

A two-founder MedTech startup is building a Class IIa software-driven diagnostic device. They have raised EUR 1.8 million. Monthly burn is EUR 90,000. Runway looks like 20 months.

The founders build a model with first revenue in month 14. Month 14 assumes the notified body stage 2 audit closes in month 12 and the certificate issues in month 13. The model shows the company reaching EUR 40,000 MRR by month 20 and being break-even funded for a Series A conversation in month 18.

We rebuild the model with regulatory reality.

Step 1: Anchor on the certificate, not the prototype. Technical documentation complete in month 8. QMS mature enough for stage 1 audit in month 9. Stage 1 audit slot availability at a mid-tier notified body in month 11 to 13. Stage 2 audit six to eight weeks after stage 1 closes. Technical documentation review parallel, typically 12 to 20 weeks with one round of findings. Certificate issuance one to two months after the last finding closes. Realistic CE mark date: month 17 to 20 in the base case.

Step 2: Anchor revenue on the certificate, not the CE mark date alone. First paying hospital or clinic deal typically closes two to four months after CE mark because procurement requires the certificate number on file. First invoice ships one to two months after the purchase order. Realistic first revenue: month 20 to 24.

Step 3: Add the sensitivity cases. A six-month slip pushes first revenue to month 26 to 30. A twelve-month slip, which happens if the notified body issues major findings or if the QMS fails stage 2, pushes first revenue beyond month 32. At EUR 90,000 monthly burn, a twelve-month slip costs EUR 1.08 million. The original EUR 1.8 million raise does not cover the base case plus the twelve-month scenario.

The honest model tells the founders two things the optimistic model hid. First, the current runway covers the base case with almost no buffer. Second, the Series A conversation has to happen before the stage 2 audit, not after, because after means a broken company if anything slips. That is a strategic insight the optimistic model prevented.

The Subtract to Ship playbook

The goal of the model is not to impress. The goal is to survive the meeting with regulatory reality and come out with a plan.

Build the model backwards from the certificate. Start with the CE mark date as an input, not an output. Ask what has to be true for that date to hold, and then move every cost and every revenue assumption into alignment. The model is a planning tool, not a forecast.

Declare the notified body queue explicitly. A single line item labelled "NB queue wait" with a duration in months. Do not bury it inside a generic "regulatory" block. Investors who know the space look for this line. Its absence is a red flag. Its presence, honestly estimated, is a credibility signal.

Run three cases, not five. Base case, six-month slip, twelve-month slip. Any more than that becomes theatre. The point is to show runway impact per scenario, not to fill pages. For each case, show the month the company runs out of cash and the month a bridge or Series A has to close.

Split burn into regulatory and non-regulatory. Regulatory burn — QMS consultancy, notified body fees under Annex IX, testing lab costs, clinical evaluation work — is largely inflexible once the process is in motion. Non-regulatory burn — growth hires, marketing, commercial pilots — can be paused. The split tells you which levers exist if the timeline slips.

Cap post-CE revenue ramp in year one. European MedTech revenue does not compound the way SaaS does. Reimbursement is national, hospital procurement cycles are nine to eighteen months, and every new country adds work. A model showing EUR 1 million ARR twelve months after CE mark in a single European market is almost always wrong. Flat quarters after the certificate are normal.

Carry a nine-month runway buffer at the moment of notified body submission. If you submit the technical documentation with six months of runway, you have handed your company to chance. The submission is the point of maximum calendar uncertainty. Nine months of runway beyond that date is the minimum a sensible plan carries. Below that, you are not managing burn — you are praying.

Make the model a shared artifact with your regulatory lead. The PRRC, the QMS lead, or the external regulatory consultant must see the model and sign off on the durations. A model the regulatory person has not seen is not a financial model. It is a wish list.

Reality Check

  1. Is your CE mark date an input in your model, or is it derived from a bottom-up engineering schedule that does not include notified body queue time?
  2. Do you have an explicit line item for notified body queue wait, and does its duration match what you have heard from actual notified bodies in the last three months?
  3. Have you run a twelve-month regulatory slip scenario, and does your current cash cover it?
  4. Does your post-CE revenue ramp assume European hospital procurement cycles of nine to eighteen months, or does it assume SaaS-style monthly compounding?
  5. Does your regulatory lead agree with every duration in the regulatory section of the model?
  6. At the moment you submit technical documentation to the notified body, will you have at least nine months of runway left under the base case?
  7. If your next fundraise were delayed by six months, which quarter does your company run out of cash in the base case, and what do you do in that quarter?

Frequently Asked Questions

How long does a notified body queue actually take in 2026? Wait times vary by notified body and by device class, and the situation has been improving slowly since the worst bottleneck years. For new Class IIa and IIb devices, assume six to twelve months from initial contact to stage 1 audit in your base case, and verify the number with the bodies you are actually talking to.

Should I model revenue in local currency by country or as a single European number? Local currency by country, because reimbursement, pricing, and sales cycles are national. A single European number hides the sequencing problem, which is where most European MedTech go-to-market plans fail.

What if my investors want hockey stick growth in the model? Show them the base case with honest MDR timing, and show them a second version with the aggressive growth assumptions and the additional funding required to hold the regulatory timeline while pursuing that growth. The comparison does the work.

Does MDR Article 120 extended transition help my cash model? Only if you have an existing MDD certificate that qualifies. For a new device on a new company, the extended transition under Regulation (EU) 2023/607 does not change your timeline.

Is it reasonable to assume the notified body queue shortens? No. Plan for current wait times and treat any improvement as upside, not as a base case assumption. Models that assume the regulator gets faster have never aged well.

What runway should I have when I submit to the notified body? Nine months minimum under the base case, and enough that a six-month slip still leaves you with three months and an active fundraise. Below that, you are accepting a real risk of bankruptcy during the certification process.

Sources

  1. Regulation (EU) 2017/745 on medical devices, consolidated text. Article 52 (Conformity assessment), Article 120 (Transitional provisions), Annex IX (Conformity assessment based on a quality management system and on assessment of technical documentation).
  2. Regulation (EU) 2023/607 amending Regulations (EU) 2017/745 and (EU) 2017/746 as regards transitional provisions for certain medical devices.
  3. EN ISO 13485:2016+A11:2021 — Medical devices — Quality management systems — Requirements for regulatory purposes.