After CE marking under the EU Medical Device Regulation, the right next market is almost never "all of them at once." The sequencing that works for most MedTech startups is: consolidate the EU launch first, then add one second market chosen by regulatory cost, time to approval, realistic market size, sales channel maturity, and strategic importance — not by the country that sounds most exciting. For the majority of startups, that second market is either the United States (via FDA 510(k) or De Novo) or a bundle of MDSAP jurisdictions (Canada, Australia, sometimes Japan and Brazil) reached through one extended QMS audit. CE marking itself has no extraterritorial effect. Every additional market is a new regulatory submission, a new budget line, and a new commercial plan.

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

TL;DR

  • CE marking under the Medical Device Regulation (MDR, Regulation (EU) 2017/745) gives market access to the European Economic Area only. It is not recognised by the US FDA, Health Canada, Australia's TGA, Japan's PMDA, Brazil's ANVISA, Switzerland's Swissmedic, or the UK's MHRA.
  • International expansion sequencing is a decision about which single next market to enter after the EU launch — not a plan to enter five markets simultaneously.
  • The five criteria that actually drive the sequencing decision are regulatory cost, time to market, realistic market size for your device, maturity of your sales channel in that country, and strategic importance to your funding and partnership plan.
  • For most Class IIa digital health and software devices, a typical sequencing is: EU first, then either US (510(k) or De Novo) or an MDSAP bundle covering Canada plus Australia — rarely both at once.
  • The most expensive sequencing mistake is assuming "CE marking = global access." It is one of the most costly assumptions a founder can write into a business plan.

A German company that learned the hard way

A German MedTech company Tibor worked with built its business plan on a simple assumption: once the device carried the CE mark, international expansion was a matter of choosing countries and hiring distributors. The US, Brazil, Japan, and Australia were all penciled in for year one after launch. Investor decks reflected the same assumption. Revenue forecasts compounded on it.

The reality arrived one country at a time. The United States does not recognise CE marking — a device placed on the US market requires an FDA submission (510(k), De Novo, or PMA) and clearance or approval by the FDA itself. Brazil does not recognise CE marking — ANVISA runs its own registration process, with its own Good Manufacturing Practices expectations that in practice are most efficiently satisfied through the MDSAP program. Japan runs its own regulatory framework under MHLW and PMDA. Australia runs its own framework under the Therapeutic Goods Administration, with the ARTG inclusion being the market access instrument and specific QMS evidence routes accepted by the TGA.

None of these authorities care that a Notified Body in Europe has issued a CE certificate. The CE mark is a statement about European conformity. It is not a passport.

The company revised its business plan. Delays were communicated to investors. The international timeline slipped by the better part of a year, and the additional budget for four parallel regulatory submissions was larger than the original EU certification budget in total. The device was fine. The plan was not.

This post exists so founders can avoid the same mistake without having to make it.

Sequencing criterion 1: regulatory cost

The first criterion is the total cost of getting through each target jurisdiction's regulatory gate. This is not just the submission fee. It is the full stack: classification analysis, gap work on the technical documentation for the new jurisdiction's requirements, translation of labelling and instructions for use, any additional testing or evidence the jurisdiction expects, local representation, submission fees themselves, QMS overlay work if the jurisdiction expects something beyond what you already run, and audit costs.

For a startup that already holds a CE certificate and operates a real QMS under EN ISO 13485:2016+A11:2021, the marginal cost of an additional jurisdiction varies by an order of magnitude across countries. A Switzerland market extension is usually cheaper than a full FDA 510(k) submission, because the Swiss framework is closely aligned with the MDR and the Swiss Authorised Representative requirement, while distinct, does not require a second technical file from scratch. An FDA 510(k) submission is more expensive than a Swiss extension but cheaper than a US De Novo submission. A PMA is in a different league entirely.

The honest sequencing move is to tabulate the real all-in regulatory cost for each candidate jurisdiction, not the headline submission fee. Startups that sequence by headline fees systematically underestimate their actual budgets.

Sequencing criterion 2: time to market

The second criterion is calendar time from decision to market access. This is where FDA 510(k), MDSAP bundled audits, and jurisdictions with backed-up review queues diverge significantly from each other.

A well-prepared FDA 510(k) for a device with a clear predicate is typically measurable in months from submission, not years — though pre-submission work adds earlier months to the calendar. An MDSAP audit extension that covers Canada, Australia, and others in one coordinated visit can reduce total calendar time across multiple jurisdictions dramatically versus pursuing each one as a separate national inspection. Conversely, some jurisdictions have historically operated with long review backlogs that make them poor first next steps regardless of how attractive the market looks on paper.

The sequencing implication is that time-to-market matters most when your runway, your commercial plan, or your partnership commitments have a hard date. If your series A investors expect US revenue in quarter three of next year, the sequencing is not optional. If your runway is comfortable and your commercial plan is flexible, slower and cheaper jurisdictions become more attractive.

Sequencing criterion 3: realistic market size

Market size is where founders most often deceive themselves. The total addressable market for medical devices in the United States, Japan, or Germany is enormous in absolute terms. The addressable market for your specific device, in your specific indication, through the specific sales channel you can realistically build is almost always much smaller.

The right market size number for sequencing is not "the US MedTech market is X billion dollars." It is the number of specific hospitals, clinics, or end users who would plausibly buy your device in the first three years, multiplied by an honest price, minus the cost of reaching them. Run that calculation per jurisdiction. Then rank.

Some jurisdictions that look small on the macro map — Australia, Switzerland, the Nordics — have specific procurement dynamics, concentrated hospital networks, or insurance structures that make them unusually efficient first steps for certain device categories. Some jurisdictions that look huge are functionally closed without a local partner, a reimbursement pathway, and specific clinical evidence that took years to assemble.

Market size must be multiplied by accessibility. The product of the two is what drives sequencing — not either number alone.

Sequencing criterion 4: sales channel maturity

The fourth criterion is whether you can actually sell the device in the target jurisdiction. Regulatory approval is necessary and not sufficient. A device that is cleared by the FDA but has no US sales team, no US hospital contracts, no US clinical champions, and no US reimbursement story is a device that has spent submission budget without generating revenue.

A mature sales channel in a jurisdiction is: a named commercial partner or direct team, identified first customers, a reimbursement path that does not require starting from zero, a clinical champion or two who will use and advocate for the device, and some degree of local language and regulatory support for post-market activities. If you do not have at least half of these for a candidate jurisdiction, adding it to the sequence is a bet that you can build the channel in parallel with the regulatory work — which is sometimes right and sometimes catastrophic.

The sequencing rule that works: prefer jurisdictions where the commercial channel is already warm over jurisdictions where the market is theoretically bigger. A smaller market you can actually sell into generates revenue. A bigger market you cannot sell into generates burn.

Sequencing criterion 5: strategic importance

The fifth criterion is strategic — the way the jurisdiction supports the rest of the company's plan independent of direct revenue. Some jurisdictions matter for signalling and investor credibility: US clearance is often a prerequisite for Series B conversations with certain investor profiles, regardless of whether US revenue materialises quickly. Some jurisdictions matter for partnership conversations: Japanese market access is often discussed as a prerequisite for certain distribution partnerships across Asia. Some jurisdictions matter because your anchor customer or anchor clinical site happens to be there, and losing that relationship would be worse than the cost of the regulatory work.

Strategic importance is legitimate as a sequencing input when it is named and honest. It becomes dangerous when it is used to justify a decision the numbers would otherwise reject, with no one forcing the team to say out loud that the jurisdiction is being pursued for signalling rather than economics.

A typical sequencing for a Class IIa software-driven device

For a Class IIa software-driven medical device — a common profile for digital health startups — the sequencing that most often works is:

  1. EU first, under the MDR. CE certificate, EUDAMED registration, PRRC in place, a real QMS operating, post-market surveillance rhythm established. Consolidate for one to three quarters. The temptation to open a second market before the first one is stable is the single most common sequencing error we see.
  2. Second market: either the US or an MDSAP-anchored bundle. This is where the sequencing forks. - If the commercial plan, investor expectations, and runway point at the US, the second step is an FDA 510(k) or De Novo, depending on predicate availability. The expected effort is significant but bounded. See the MDR-versus-FDA comparison in post 049 and the dual-market strategy framing in post 050. - If the commercial plan points to multiple smaller markets where a single audit can cover the QMS layer, the second step is an MDSAP audit extension covering Canada (where MDSAP is effectively required for most device classes), Australia, and optionally Japan and Brazil as scope expansions. See post 635 for the full treatment of MDSAP as an efficiency lever.
  3. Third market and beyond: one at a time, driven by revenue signals from the first two markets. Expansion from three markets to five or seven is not a sprint. It is a rhythm — one jurisdiction per planning cycle, each one justified by a commercial case that reflects what the earlier markets actually delivered.

This is a template, not a formula. The right sequencing for a Class III implant, a high-complexity imaging device, or a niche diagnostic looks different. The frame — consolidate, then add one — is the same.

Common sequencing mistakes

The sequencing mistakes we see most often are variants of the same underlying error: treating international expansion as a map-colouring exercise rather than a capital allocation decision.

  • "CE equals global." The assumption that the CE mark provides access beyond the European Economic Area. It does not. Every other jurisdiction is its own submission, its own audit, its own budget line. This is the mistake in the German company story above.
  • Parallel launches in five markets at once. Trying to open EU, US, Canada, Japan, and Australia simultaneously. The regulatory work can sometimes be parallelised. The commercial work almost never can. One market at a time is usually faster end-to-end than five in parallel.
  • Sequencing by macro market size instead of addressable market. Choosing the US because it is the largest MedTech market in absolute terms, without validating that your specific device has a realistic commercial path there.
  • Ignoring reimbursement until after the regulatory approval. Approval without a reimbursement path is a device on a shelf. In markets with strong public payers or concentrated private payers, the reimbursement pathway determines the sales timeline more than the approval pathway does.
  • Choosing a second market because a single investor mentioned it once. Legitimate strategic input from investors should be weighed and named. An offhand comment in a pitch meeting is not a strategy input.
  • Neglecting Switzerland and the UK. These are frequently overlooked by EU-based startups precisely because they feel "close to the EU" — but both operate independent regulatory frameworks now and require specific steps. See post 624 for the UK picture and post 625 for Switzerland.
  • Not planning the MDSAP decision deliberately. Treating MDSAP as something to "decide later" when it is in fact one of the cleanest efficiency levers for multi-market QMS work — if the target markets match. See post 629 for the Canada-MDSAP link specifically.

The MDSAP bundle option

One sequencing pattern deserves its own section because it is often overlooked by EU-first startups: treating MDSAP as a single sequencing step that covers multiple jurisdictions.

The Medical Device Single Audit Program allows one QMS audit, performed by an MDSAP-recognised Auditing Organisation, to satisfy the QMS inspection needs of the US FDA (for routine QMS inspections, with exceptions), Health Canada (where MDSAP is effectively mandatory for most device classes under the Canadian framework), Australia's TGA (as one recognised QMS evidence route), Japan's PMDA, and Brazil's ANVISA. The audit assesses the manufacturer's QMS against EN ISO 13485:2016+A11:2021 plus the country-specific regulatory overlays for each declared jurisdiction.

For a startup whose commercial plan includes Canada and at least one other MDSAP market, bundling those jurisdictions behind a single audit extension is often the most efficient sequencing step. The MDSAP audit does not deliver premarket approval in any jurisdiction — you still need the FDA submission, the Canadian Medical Device Licence, the ARTG inclusion, and so on — but it collapses the QMS inspection layer into one visit instead of five separate national inspections.

The MDSAP bundle does not cover the European Union. The MDR conformity assessment under Annex IX remains a separate activity performed by an EU Notified Body, and a CE certificate does not flow from an MDSAP audit. For the mechanics of that separation, see post 635.

When MDSAP bundling is the right second-step sequencing choice, it is usually because Canada is in the plan. Canada is the forcing function. Everything else — Australia, Japan, Brazil — is scope added to an audit that was going to happen anyway.

The US-first versus EU-first decision

One sequencing question deserves direct treatment: for startups that have not yet entered any market, should the first regulated market be the EU under the MDR or the US under the FDA?

There is no universal answer. The decision depends on where the founding team sits, where the anchor clinical evidence exists, where the first customers are, and where the first investors expect the company to land. A startup based in Graz with a clinical partner in Vienna and a first customer in Munich should usually start with the EU. A startup based in Boston with a clinical partner at a US academic centre and a US patient population should usually start with the US. A startup sitting in the middle has to decide honestly which market is closer to its actual operating reality.

What we see go wrong is the reverse decision — an EU-based team that starts with the FDA because the US market feels bigger, without a US clinical partner, without a US commercial team, and without a plan for US post-market activities. The regulatory submission succeeds, the commercial launch fails, and the runway is gone. The mirror error — a US-based team that starts with the MDR because Europe is "easier" — is equally dangerous and based on an outdated picture of what MDR conformity assessment actually costs in 2026.

The dual-market planning question has its own post. See post 050 for the full treatment and post 601 for EU founders orienting themselves to the FDA framework for the first time.

The Subtract to Ship angle on expansion sequencing

The Subtract to Ship methodology maps directly onto expansion sequencing, because sequencing is subtraction. A startup that tries to enter five markets simultaneously is not running a multi-market strategy — it is running five competing half-strategies that share the same overhead. A startup that sequences one market at a time is running a multi-market strategy that respects how capital, calendar time, and management attention actually behave.

Subtraction in this context means cutting the jurisdictions that are in the plan for signalling rather than economics, cutting the parallel submissions that compete for the same regulatory affairs bandwidth, cutting the markets where the commercial channel is fantasy rather than fact, and cutting the MDSAP scope for jurisdictions that are not in the three-year plan even if the audit overlay is technically cheap to add. Every cut frees capital and attention for the markets that remain.

The MDR is the single point of truth for the EU step. The FDA framework is the single point of truth for the US step. MDSAP is the efficiency mechanism for multi-market QMS work where it fits. None of these replace the others. All of them have to be respected on their own terms. Subtract to Ship is not about doing less regulatory work — it is about doing only the regulatory work that traces to a market you are actually entering. See post 065 for the methodology in full.

Reality Check — Where do you stand on international expansion?

  1. Can you state, in one paragraph, which specific markets are in your three-year plan, in what order, and why each one is in the plan at all?
  2. For each market in the plan, do you have a named commercial channel — a distributor, a direct team, a clinical partner — or is the channel still a hypothesis?
  3. Have you calculated the full all-in regulatory cost of each target jurisdiction (not just submission fees), and does the sum fit within the capital you actually have or can realistically raise?
  4. Have you separated the markets in your plan from the markets in your deck? If investors were reading the deck alone, would they believe in a more ambitious expansion than the plan actually supports?
  5. If Canada is in the plan, have you sequenced MDSAP accordingly? If Canada is not in the plan, have you stopped carrying MDSAP as a default assumption?
  6. If the US is in the plan, do you have a clear answer to "FDA 510(k), De Novo, or PMA?" for your specific device, and does the sequencing reflect that answer honestly?
  7. Has your team explicitly rejected the "CE equals global access" assumption in writing, or is it still silently embedded in a spreadsheet somewhere?

If more than three of these produced a "not yet," your international expansion plan is not a sequencing yet. It is a wishlist.

Frequently Asked Questions

Does CE marking under the MDR give me any market access outside the European Economic Area? No. The CE mark is a statement of conformity with the Medical Device Regulation (Regulation (EU) 2017/745) and gives market access to the European Economic Area. It is not recognised by the US FDA, Health Canada, the Australian TGA, Japan's PMDA, Brazil's ANVISA, Switzerland's Swissmedic, the UK's MHRA, or any other non-EEA authority. Every additional jurisdiction is its own regulatory submission.

Which country should I enter after the EU launch? There is no universal answer, but for most Class IIa software-driven devices the realistic choice is between the United States (via FDA 510(k) or De Novo) and an MDSAP-anchored bundle covering Canada and one or more of Australia, Japan, and Brazil. The decision should be driven by commercial channel maturity, investor expectations, runway, and whether Canada is in the plan at all. Entering the US and MDSAP markets simultaneously is rarely the right first expansion step.

How long after CE marking should I wait before entering a second market? Long enough to stabilise the EU launch. In practice, that usually means running at least one full post-market surveillance cycle, closing any Notified Body findings, establishing the vigilance rhythm, and having the commercial channel producing revenue rather than just promises. Opening a second market while the first is still on fire is the most common sequencing mistake we see, and it almost always costs more than the delay would have.

Is MDSAP the right default for any startup expanding beyond the EU? No. MDSAP is the right efficiency lever when Canada is in the plan and at least one other MDSAP jurisdiction is in the plan. It is not free, it adds scope to the QMS audit, and it provides no value for jurisdictions outside its five participating authorities. If your expansion plan is EU plus the US only, MDSAP is usually not the first efficiency move to reach for.

What is the single most expensive sequencing mistake? Writing "CE marking = global access" into a business plan and building investor expectations on top of that assumption. Every major non-EU market has its own regulatory authority, its own requirements, its own timelines, and its own costs. Discovering this one jurisdiction at a time, after commitments have been made, is the pattern that destroys international timelines.

How do I decide whether to start with the EU or the US for my very first regulated market? Start where the team, the clinical partners, the first customers, and the first investors actually are. An EU-based team with EU clinical partners and EU customers should usually start with the MDR. A US-based team with US clinical partners and US customers should usually start with the FDA. Teams that start in the geography opposite their actual operating base almost always pay more in delays and failed commercial launches than they save in regulatory fees.

Sources

  1. Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on medical devices (MDR), Articles 1 and 2 (scope and definitions) and Annex IX (conformity assessment based on a quality management system and on assessment of technical documentation). Official Journal L 117, 5.5.2017. Cited here to establish that the MDR governs EU market access only.
  2. EN ISO 13485:2016 + A11:2021 — Medical devices — Quality management systems — Requirements for regulatory purposes. The QMS standard that forms the spine of multi-market QMS work and the core of the MDSAP audit model.
  3. US Food and Drug Administration, Health Canada, Therapeutic Goods Administration (Australia), Ministry of Health, Labour and Welfare / Pharmaceuticals and Medical Devices Agency (Japan), Agência Nacional de Vigilância Sanitária (Brazil), Swissmedic (Switzerland), and MHRA (United Kingdom) — the non-EU regulatory authorities discussed in this post. Each maintains its own framework referenced here at the general framing level; for current submission parameters in any specific jurisdiction, consult the authority's current guidance or a recognised regulatory partner for that market.
  4. Medical Device Single Audit Program (MDSAP) — program documentation maintained by the International Medical Device Regulators Forum (IMDRF) and the five participating regulatory authorities. Referenced here at the general framing level; for current program parameters, consult a recognised Auditing Organisation.

This post is part of the FDA & International Market Access series in the Subtract to Ship: MDR blog. Authored by Tibor Zechmeister and Felix Lenhard. International expansion is where MedTech business plans most often meet physics: the number of markets the company can honestly enter in a given year is smaller than the number of markets the deck describes. Sequence the ones that are real. Leave the rest in the appendix until they become real. That is the whole discipline.