---
title: When to Pivot in MedTech: Changing Direction Without Losing Regulatory Progress
description: Pivoting in MedTech is more expensive than in SaaS because the regulatory work is partly device-specific. Here is when to pivot and how to keep regulatory value.
authors: Tibor Zechmeister, Felix Lenhard
category: MedTech Startup Strategy & PMF
primary_keyword: when to pivot MedTech regulatory
canonical_url: https://zechmeister-solutions.com/en/blog/when-to-pivot-medtech-regulatory
source: zechmeister-solutions.com
license: All rights reserved. Content may be cited with attribution and a link to the canonical URL.
---

# When to Pivot in MedTech: Changing Direction Without Losing Regulatory Progress

*By Tibor Zechmeister (EU MDR Expert, Notified Body Lead Auditor) and Felix Lenhard.*

> **Pivoting in MedTech is structurally more expensive than pivoting in SaaS because a meaningful portion of the regulatory work — clinical evaluation, classification rationale, intended purpose, specific technical documentation, device-specific risk analysis — is tied to the exact device you were building. Pivot the intended purpose and that work partly evaporates. The honest rule is: pivot when the market evidence from real buyers, users, and clinical partners says the current direction will not produce a viable business, and pivot early enough that the intended-purpose work has not yet been locked. The regulatory work that transfers across pivots is the horizontal work — the QMS, the people, the supplier controls, the general risk-management process, the PMS infrastructure. The device-specific work does not transfer. Knowing which is which is the difference between a pivot that saves the company and a pivot that restarts the clock.**

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

---

## TL;DR

- Pivoting in MedTech is more expensive than in SaaS because part of the regulatory file is device-specific and moves with the intended purpose, not with the company.
- Under Article 2(12) of Regulation (EU) 2017/745, intended purpose is defined by the label, the instructions for use, the promotional and sales materials, and the clinical evaluation — change any of those and you have changed the regulatory input that drives classification under Article 51 and Annex VIII.
- Horizontal work transfers across pivots: the QMS, PRRC function, supplier management, training records, the general risk-management process, the PMS infrastructure, and the team itself.
- Device-specific work does not transfer: the classification rationale, the clinical evaluation specific to the old intended purpose, the GSPR checklist for the old device, labelling, IFU, and the technical file sections tied to the old design.
- Pivot early enough that the intended-purpose work has not been locked. A pivot in the exploratory phase costs weeks. A pivot after Notified Body engagement costs quarters and euros.
- The four most common MedTech pivots are the intended-purpose pivot, the patient-population pivot, the form-factor pivot, and the indication pivot. Each has a different regulatory cost profile.

---

## The Graz wellness pivot — when the pivot out of MDR is the right answer

A Graz-based startup had spent months building toward MDR certification. Notified Body shortlisted, draft technical documentation outline under way, a clinical evaluation plan taking shape. Every downstream deliverable was scoped to a specific assumed classification, and the runway clock was ticking against that scope.

Then the commercial signal shifted. Sales conversations were pointing toward a use case that did not need the medical framing. The underlying hardware and software were the same. What changed was the commercial intent — who the buyer was, what the buyer was paying for, what the company wanted to claim on the label and on the website.

We ran the Purpose Pass on the revised positioning. The clinical verbs — diagnosis, prevention, monitoring, treatment, alleviation — came out of the four sources Article 2(12) recognises: the label, the instructions for use, the promotional and sales materials, and the clinical evaluation. A revised intended purpose was drafted, one that honestly described the product as a wellness device rather than a medical device.

The classification did not just change. It disappeared. The MDR no longer applied to the revised product because it no longer met the Article 2(1) definition of a medical device. Market entry happened in weeks instead of quarters. The Notified Body engagement was avoided — not evaded, avoided, because the revised product genuinely was no longer a medical device under the Regulation.

That is the most dramatic version of a MedTech pivot. Most pivots are smaller and do not remove the MDR entirely. What every pivot has in common is the same mechanical question: what changed in the four Article 2(12) sources, and what regulatory work is still standing afterwards?

## Why pivoting in MedTech is more expensive than in SaaS

A SaaS pivot is cheap because the cost of the product is mostly horizontal. Codebase, infrastructure, team, brand, channel — most of it survives a change in who you are selling to or what you are solving. The founder rewrites the positioning, redirects the marketing, and ships a new feature set in a sprint.

A MedTech pivot is more expensive because a meaningful share of the cost is vertical. The regulatory file is built around a specific intended purpose and a specific device. Clinical evaluation under Annex XIV is bound to the clinical claims of that device. The classification rationale under Annex VIII is computed from that intended purpose. Labelling, instructions for use, the GSPR checklist in Annex I, the device-specific risk analysis — all of it is scoped to the product you were building, not to the company that was building it.

When the intended purpose changes, the vertical work moves with it. Some of it can be rewritten quickly. Some of it has to be redone from scratch. The horizontal work — the QMS, the PRRC, supplier controls, training, PMS infrastructure — stays with the company and transfers intact. The first job in any pivot is to separate those two stacks honestly so the founder knows what survives and what does not.

The second reason MedTech pivots are more expensive is timing. In SaaS the cost of a late pivot is a sprint. In MedTech the cost of a late pivot is whatever the Notified Body, the clinical investigators, and the testing laboratories have already been paid for work that no longer applies. Pivot before those invoices land and the cost is manageable. Pivot after and the cost scales with how deep the project has gone.

## Signs you need to pivot

The signal is never a single event. It is a pattern of evidence from the Phase 1 verification work that, read honestly, says the current direction will not produce a viable business. The following patterns show up often enough to name.

**Buyer conversations return the wrong numbers.** You have talked to ten or fifteen hospital buyers and the improvement metric they care about is not the metric your device moves, or the improvement is in the 1.5x range instead of the 10–20x range. You are not losing the buyer because your pitch is bad. You are losing because the device is not strong enough on the dimension the buyer counts.

**Clinicians who love the idea cannot get their hospital to listen.** The decision-making unit is layered, and champion enthusiasm that never converts into procurement interest is a signal that the value is real for the user and absent for the buyer. That gap does not close with more pitching. It closes with a pivot of some kind — buyer, indication, workflow, or form factor.

**The clinical evidence you can realistically produce does not support the claim you need.** You thought you could demonstrate the benefit with literature and equivalence. The honest answer is that you cannot, and a full clinical investigation is out of scope. The claim has to shrink, which means the intended purpose has to shrink, which is a pivot.

**The classification is heavier than the runway can support.** The Classification Pass produced a higher class than the business plan assumed, and the cost and timeline of that class do not fit the runway. Either the runway has to grow or the device has to move to a lower class — which usually requires pivoting the intended purpose.

**Reimbursement does not exist for the pathway.** The device works, the hospital likes it, the clinician signs off, and nobody will pay for it because there is no code, no bundle, and no precedent. This is a business-model pivot, and it almost always changes the buyer or the geography.

**The runway will not reach Notified Body readiness at the current burn.** A pivot to a lower-cost regulatory path is sometimes the only way to finish the certification at all. Pivoting here is not defeat. It is honest triage.

If two or more of these signals are present and you cannot explain them away with evidence, the pivot conversation is already on. The only question left is what kind of pivot and how early.

## Types of pivots and their regulatory impact

Not every pivot is the same. Four archetypes cover most of the real-world cases, and each has a different regulatory cost profile.

### The intended-purpose pivot

You change what the device claims to do, for whom, or in what clinical context. This is the heaviest pivot because Article 2(12) is the input to classification under Article 51 and Annex VIII, and changing it changes everything downstream. The label changes, the IFU changes, the website changes, the clinical evaluation changes. The classification may stay the same, move up, move down, or — in the rarest and most valuable case — move out of the MDR entirely. Most of the device-specific regulatory work has to be revised or redone. The horizontal work survives intact.

### The patient-population pivot

You narrow or expand the patient population the device is intended for. Narrowing — from a broad population to a specific indication — is usually lighter than expanding, because a narrower population means a narrower clinical evaluation and often a lower class. Expanding — especially into paediatric, central nervous system, or critically ill populations — can escalate the class under Annex VIII and force a significant expansion of the clinical evidence. The existing QMS, supplier controls, and horizontal work all transfer.

### The form-factor pivot

You keep the intended purpose but change the physical form of the device — from a bench-top unit to a handheld, from a disposable to a reusable, from a tethered to a wireless variant. Form-factor pivots trigger a cascade of technical re-work: updated risk analysis for the new hazards, new usability engineering, potentially new electrical safety and EMC testing, updated biocompatibility assessment if materials change, and new manufacturing validation. Classification often stays the same because the intended purpose has not moved. The clinical evaluation may survive largely intact if the clinical claims are unchanged.

### The indication pivot

You add, remove, or narrow a clinical indication within a broadly similar intended purpose. Adding an indication is usually the most expensive — new clinical evidence for the new indication, updated clinical evaluation, updated labelling, and a classification re-check in case the new indication activates a different rule. Narrowing an indication is usually the cheapest and is often the quiet escape hatch when a broader claim is not supportable.

The pivot plan always starts by identifying which of these four archetypes is in play, because the regulatory cost profile is fundamentally different for each.

## What regulatory work transfers across pivots

Some of the work done before the pivot survives the pivot intact. This is the horizontal stack, and protecting it is the reason the pivot does not have to feel like starting over.

**The QMS.** The quality management system under EN ISO 13485:2016+A11:2021 is built around the organisation, not the product. The procedures, the records, the management review, the internal audit programme, the CAPA system, the supplier qualification, the training records — all of it stays with the company through a pivot. A QMS that was built honestly for the pre-pivot device can host the post-pivot device with procedure updates but not a rebuild.

**The PRRC function.** The Person Responsible for Regulatory Compliance under Article 15 is tied to the manufacturer, not the device. The appointment, the qualification evidence, the responsibility assignment — all of it transfers.

**Supplier controls and agreements.** The qualified supplier list, the quality agreements, the incoming inspection procedures — the suppliers may change with the pivot, but the control system that manages them does not.

**The general risk-management process.** The EN ISO 14971:2019+A11:2021 process — how you identify hazards, estimate and evaluate risks, control them, and verify the controls — is process, not content. The specific risk file for the old device does not transfer. The process that produced it does.

**The PMS infrastructure.** The post-market surveillance system, the data collection channels, the trend analysis process, the vigilance reporting procedure — all of it is infrastructure that transfers to the new device with content updates.

**The team and the lived experience.** This is the most underrated transferable asset. The founder who has lived through the first cycle of Purpose Pass, Classification Pass, Evidence Pass, and Operations Pass is a qualitatively different founder afterwards. The team knows how the Regulation actually reads, how a Notified Body conversation actually sounds, and how to tell a real obligation from a consultant-invented one. That knowledge does not appear on any balance sheet and it is the single most valuable thing a pivot preserves.

## What does not transfer

The device-specific work does not survive the pivot and pretending otherwise is the most common and most expensive mistake.

- The classification rationale for the old intended purpose.
- The clinical evaluation tied to the old clinical claims.
- The GSPR checklist and the Annex I conformity arguments for the old device.
- The device-specific risk analysis and the hazard list derived from the old use case.
- Labelling, the IFU, and the promotional and sales materials.
- Device-specific verification and validation test reports where the test conditions no longer match.
- Any Notified Body correspondence tied to the old device scope.

Some of this can be partially salvaged — a literature review done for the old clinical claim may still be useful for a related claim, and a test report for a component that did not change may still apply. The honest default, though, is to treat the device-specific stack as invalidated and to budget for redoing the parts the new device needs.

## The intended-purpose pivot — the highest-leverage version

The intended-purpose pivot deserves its own treatment because it is the version with the most leverage and the most risk. Article 2(12) defines the four sources the Regulation reads: the label, the instructions for use, the promotional and sales materials, and the clinical evaluation. The pivot exercise is literally to rewrite the intended purpose paragraph and check every one of those four sources against it.

The Graz wellness pivot is the flagship example. A small revision to the positioning — honestly grounded in the commercial reality — removed the clinical verbs from the four sources and moved the product out of MDR scope entirely. That outcome is rare, and it only applies when the revised product genuinely is no longer a medical device under Article 2(1). It is not a trick and it does not work by accident. It works because the Purpose Pass was run deliberately, the revised positioning was honest, and the documentation was tight.

A less dramatic version of the same pivot happens inside the MDR. The revised intended purpose produces a different class under Annex VIII, or the same class via a different rule, or the same class and rule but with a narrower conformity assessment scope. Each of those outcomes changes the size of the regulatory project in ways that matter on a startup budget. The exercise is worth running every time there is serious evidence that the current direction is not going to produce a viable business.

## The patient-population pivot

Pivoting the patient population is often the cleanest way to resolve a classification that the runway cannot afford. Narrowing from a broad population to a specific one — for example, excluding paediatric use, or restricting to a specific disease stage — can reduce the clinical evidence burden and, in some cases, move the device down a class under Annex VIII. It can also reduce the size of the clinical investigation if one is still required.

The risk in this pivot is giving away the market along with the regulatory cost. A narrower population is a narrower business. Run the numbers on the narrowed population before committing — if the remaining market supports the business case, the pivot is a win; if it does not, you have traded one problem for a different one.

## The form-factor pivot

The form-factor pivot is the one founders underestimate most often. Changing a device from one form to another — bench-top to handheld, wired to wireless, disposable to reusable — looks like an engineering change but triggers a long list of regulatory consequences. New usability work under EN 62366-1. New risk analysis for the new hazards. New electrical safety and EMC testing under the IEC 60601-1 family if electrical properties change. New biocompatibility assessment if contacting materials change. Manufacturing validation for the new process.

The clinical evaluation often survives largely intact if the clinical claim has not moved. The QMS survives. The rest of the device-specific file needs a pass. Budget accordingly.

## The indication pivot

Adding or removing a clinical indication within a similar intended purpose is the pivot that most commonly triggers silent re-classification. A new indication can activate a different Annex VIII rule, or escalate within the same rule, without the founder noticing. The fix is the paper exercise: write the revised intended purpose with the new indication, re-run Annex VIII, compare the output to the old one, and note the delta. If the delta is non-zero, the downstream re-work is already scoped. If the delta is zero, the pivot is lighter than it feels.

## The combined HW/SW classification pivot — defending the lighter path

There is another kind of pivot that is not a change of direction but a change of defence. A combined hardware/software device we worked with had the software side clearly on Class IIa — no dispute — but the hardware side was contested: was it Class I, or Class I with measurement function? The distinction mattered because measurement function triggers additional conformity assessment requirements. A deep review of the actual measurement behaviour and the intended clinical use produced a defensible argument that the hardware was Class I without a measurement function. The Notified Body accepted the argument. This is not gaming the system. It is applying the classification rules correctly with the evidence that supports the lighter path.

The reason this belongs in a pivot post is that the move from the heavy interpretation to the defended lighter interpretation has almost the same effect as a pivot: it changes the regulatory scope, it changes the cost, and it changes the timeline. Sometimes the right "pivot" is not a change in direction but a change in how the current direction is classified and defended. Run the Classification Pass before you run a direction pivot, because the cheaper fix may be sitting in the rulebook.

## Common pivot mistakes

- Treating the pivot as a pure business decision and not re-running the Purpose Pass. The Regulation will notice even if the founder does not.
- Updating the IFU for the new intended purpose and leaving the old claims on the website. Article 2(12) reads all four sources; a residual claim on the website keeps the old intended purpose alive in the Regulation's eyes.
- Assuming the clinical evaluation survives the pivot because the device looks similar. Clinical evaluation is tied to the claims, not to the hardware.
- Pivoting after the Notified Body has been engaged without notifying them. Significant changes have to be communicated through the QMS change control path.
- Discarding the horizontal work in a panic. The QMS, the PRRC, the supplier controls, the team knowledge — these survive. Do not rebuild them unless you have to.
- Pivoting too late. Every month of Phase 2 work on the wrong intended purpose compounds the cost of the eventual pivot. The cheapest pivot is the earliest one supported by evidence.

## The Subtract to Ship angle

A pivot is a second chance to run the four passes. The Purpose Pass rewrites the intended purpose. The Classification Pass re-runs Annex VIII against the revised paragraph. The Evidence Pass re-scopes the clinical evidence for the revised claim. The Operations Pass checks whether the existing QMS and PMS still match the revised device or need adjustment.

Subtraction in a pivot is ruthless. Cut the device-specific work that no longer applies. Keep the horizontal work that still does. Rebuild only what the new intended purpose actually requires under a specific MDR article, annex, or harmonised standard. What remains is the minimum regulatory work the pivoted product actually needs. Anything else is leftover work from the pre-pivot plan, and it is waste. Read [the Subtract to Ship framework for MDR](/blog/subtract-to-ship-framework-mdr) for the full methodology and [device classification for startup pivots](/blog/classification-startup-pivots) for the classification mechanics.

## Reality Check — Where do you stand on your pivot?

1. Have you written the revised intended purpose paragraph honestly, using the Article 2(12) structure — what the device is for, who uses it, on whom, for what condition, in what environment, for how long?
2. Have you checked the revised paragraph against all four Article 2(12) sources — label, IFU, promotional and sales materials, clinical evaluation — including the website and the investor deck?
3. Have you re-run Annex VIII against the revised paragraph and produced a new class with a specific rule citation?
4. Do you know which of the four pivot archetypes you are actually running — intended-purpose, patient-population, form-factor, or indication — and which cost profile applies?
5. Have you separated the horizontal work that transfers from the device-specific work that does not, on paper, before committing to the pivot?
6. Is the pivot supported by real buyer, user, and clinical-partner evidence, or by a new hypothesis you have not tested?
7. If the Notified Body is already engaged, have you planned the change notification or significant change assessment that the pivot requires?
8. Does the post-pivot business model check out before the post-pivot regulatory work starts, or are you hoping it will?

## Frequently Asked Questions

**Is pivoting in MedTech always more expensive than in SaaS?**
Yes, structurally, because a share of the regulatory file is bound to the specific intended purpose and device you were building. Clinical evaluation, classification rationale, the GSPR checklist, labelling, and device-specific risk analysis all move with the intended purpose. The horizontal work — QMS, PRRC, supplier controls, PMS infrastructure, and team experience — transfers intact, so the pivot is not a full restart, but it is always more expensive than a SaaS pivot where most of the stack is horizontal.

**When is the best time to pivot in a MedTech startup?**
Early enough that the intended-purpose work has not been locked and the Notified Body has not been paid for work that the pivot will invalidate. In practice, that usually means during the exploratory phase, before the serious technical documentation and clinical evaluation spend has started. A pivot in Phase 1 costs weeks. A pivot after Notified Body engagement costs quarters and significant euros.

**Which parts of the regulatory file survive a pivot?**
The horizontal work — the QMS, the PRRC function, supplier controls, training records, the general risk-management process, the PMS infrastructure, and the team knowledge. The device-specific work — the classification rationale, clinical evaluation, GSPR checklist, labelling, IFU, and device-specific risk analysis — usually does not survive and has to be revised or redone for the new intended purpose.

**Can I pivot out of the MDR entirely?**
Sometimes. If the revised product genuinely no longer meets the Article 2(1) definition of a medical device, the MDR no longer applies. This requires removing all clinical claims from the four Article 2(12) sources, documenting the reasoning, and keeping the revised positioning honest. It is a legitimate pivot, not a trick, and it only works when the revised product really is not a medical device. See [medical device vs. wellness product](/blog/medical-device-vs-wellness-product) and [borderline products under the MDR](/blog/borderline-products-mdr) for the mechanics.

**Does changing the patient population count as a pivot?**
Yes, and it is one of the four archetypes. Narrowing the population is often the cheapest way to reduce a heavy classification and a heavy clinical evidence burden. Expanding the population — especially into paediatric, central nervous system, or critically ill populations — can escalate the class under Annex VIII and expand the clinical evidence significantly.

**Do I have to tell the Notified Body about a pivot?**
If the Notified Body is already engaged, yes. Significant changes to intended purpose, classification, or the device itself have to be communicated through the QMS change control path. Pivoting in silence is not a lean move — it is the fastest way to lose the certificate when the change is discovered at the next surveillance audit.

**Is a classification defence the same as a pivot?**
No, but it has similar effects. Defending a lighter classification — for example, arguing a combined hardware/software device into Class I without a measurement function when the evidence supports it — changes the regulatory scope, cost, and timeline without changing the product. Run the classification defence exercise before you run a direction pivot, because the cheaper fix may already be in the rulebook.

## Related reading

- [Product-Market Fit for MedTech Startups](/blog/product-market-fit-medtech-startups) — the hub post for PMF in the MedTech context and the starting point for the pivot conversation.
- [The MedTech Startup Paradox](/blog/medtech-startup-paradox) — the structural reason pivots in MedTech are bounded by regulation.
- [Medical Device vs. Wellness Product](/blog/medical-device-vs-wellness-product) — the line that determines whether a pivot can leave MDR entirely.
- [Borderline Products Under the MDR](/blog/borderline-products-mdr) — the borderline mechanics that a pivot often has to respect.
- [MDR Device Classification Explained](/blog/mdr-device-classification-explained) — the Annex VIII walkthrough you will re-run on the revised intended purpose.
- [Device Classification for Startup Pivots](/blog/classification-startup-pivots) — the classification mechanics of a pivot in detail.
- [The Minimum Viable Regulatory Strategy for MDR](/blog/minimum-viable-regulatory-strategy-ce-mark-limited-resources) — the lean regulatory path the post-pivot device should inherit.
- [How to Validate a MedTech Idea Before MDR](/blog/validate-medtech-idea-before-mdr) — the validation work that prevents the late pivot in the first place.
- [When to Start MDR Regulatory Work](/blog/when-to-start-mdr-regulatory-work) — the timing decision whose delay creates most of the pivot pain.
- [Preserving Regulatory Work Through a Pivot](/blog/preserving-regulatory-work-through-pivot) — the practical companion to this post on protecting the horizontal stack.
- [The Subtract to Ship Framework for MDR](/blog/subtract-to-ship-framework-mdr) — the methodology that runs on both the pre-pivot and post-pivot device.

## Sources

1. Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on medical devices, consolidated text. Articles cited: Article 2(12) intended purpose; Article 51 classification; Annex VIII classification rules. Official Journal L 117, 5.5.2017.
2. Regulation (EU) 2023/607 of the European Parliament and of the Council of 15 March 2023 amending Regulations (EU) 2017/745 and (EU) 2017/746 as regards the transitional provisions for certain medical devices and in vitro diagnostic medical devices. Official Journal L 80, 20.3.2023.

---

*This post is part of the MedTech Startup Strategy & PMF series in the Subtract to Ship: MDR blog. Authored by Felix Lenhard and Tibor Zechmeister. A pivot is a second chance to run the Purpose Pass. Take it on paper, honestly, before you take it to engineering — and protect the horizontal stack while the device-specific work gets rebuilt.*

---

*This post is part of the [MedTech Startup Strategy & PMF](https://zechmeister-solutions.com/en/blog/category/startup-strategy) cluster in the [Subtract to Ship: MDR Blog](https://zechmeister-solutions.com/en/blog). For EU MDR certification consulting, see [zechmeister-solutions.com](https://zechmeister-solutions.com).*