Intended purpose is the manufacturer's legal statement of what a device is for, fixed in writing on the label, in the instructions for use, in promotional or sales materials, and in the clinical evaluation. Intended use is the colloquial term for what users actually do with the device in the real world. MDR Article 2(12) cares only about the first. The gap between the two is where founders accidentally turn a wellness product into a medical device, and where experienced teams deliberately narrow their regulatory path.

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

TL;DR

  • MDR Article 2(12) defines intended purpose as the use for which a device is intended according to the data supplied by the manufacturer on the label, in the instructions for use, or in promotional or sales materials or statements, and as specified by the manufacturer in the clinical evaluation.
  • "Intended use" is not an MDR term. It is an informal phrase that usually refers to how users in practice employ the device, which may or may not match the manufacturer's written intended purpose.
  • The MDR regulates based on the manufacturer's stated intended purpose, not on user behaviour. Your own words, on the four sources listed in Article 2(12), determine whether you have a medical device at all.
  • Change the intended purpose and you can change the classification, the conformity assessment route, the clinical evidence requirements, or the fact of whether the product is regulated as a medical device in the first place.
  • The four sources in Article 2(12) are an exhaustive list of places the manufacturer speaks. Anything you put there is evidence. Anything contradictory between them is a compliance problem.
  • Two real cases illustrate the leverage: a Graz wellness pivot where a small revision to the intended purpose moved a product from MDR territory to wellness, and an anti-intended-purpose strategy where documented evidence proved a product was not a medical device at all.

Why this matters. The Graz wellness pivot

A founder in Graz had been building an MDR strategy for months. The team had a Notified Body shortlist, a draft technical file outline, and a rough clinical evaluation plan. Everything was organised around the assumption that the product was a medical device headed for CE marking under MDR.

Tibor ran the Purpose Pass on their draft intended purpose. The claims on the website, in the IFU drafts, and in the investor deck were painting a picture that put the product squarely in medical device territory. Not because the technology demanded it, but because the language did. A small, deliberate revision of the intended purpose reframed the product as a wellness device. The hardware did not change. The software did not change. What changed was the set of sentences the manufacturer was willing to stand behind on the label, in the IFU, and in sales materials.

The product entered the market as a wellness device, under the applicable consumer product legislation rather than MDR, much faster than the original plan allowed. The Notified Body queue, the clinical evidence timeline, and the technical documentation workload for MDR were avoided. Not evaded, avoided, because the revised intended purpose genuinely no longer met the definition of a medical device under MDR Article 2(1).

That is the power of the distinction this post is about. The regulation keys off what the manufacturer says the device is for. Choose those words carelessly and you walk into MDR by accident. Choose them deliberately and you either take the MDR path because you actually need to, or you discover you do not.

The exact Article 2(12) definition

The MDR does not leave intended purpose to interpretation. It is defined explicitly in Article 2(12):

"'intended purpose' means the use for which a device is intended according to the data supplied by the manufacturer on the label, in the instructions for use, or in promotional or sales materials or statements, and as specified by the manufacturer in the clinical evaluation.". Regulation (EU) 2017/745, Article 2, point (12).

Read that sentence again, slowly. It contains four sources and one actor. The actor is the manufacturer. The sources are: the label, the instructions for use, promotional or sales materials or statements, and the clinical evaluation. Nothing else counts as intended purpose. A user's habit of repurposing the device does not count. A clinician's off-label use does not count. A competitor's marketing about a similar product does not count. The manufacturer, in these four places, speaks. And what the manufacturer says there is what the regulation calls intended purpose.

This is the hinge of the entire classification system. Article 2(1) defines a medical device by its intended purpose. Article 51 classifies devices by their intended purpose. Article 61 and Annex XIV require the clinical evaluation to demonstrate conformity for the intended purpose. The entire regulatory cathedral rests on this one sentence in Article 2(12).

What "intended use" means colloquially

"Intended use" is not a term of art in the MDR. It does not appear in Article 2 as a defined concept. In everyday speech. And in older FDA-influenced MedTech writing. "intended use" often refers to how a device is actually used, or expected to be used, by the people who pick it up. That is a description of user behaviour, not of manufacturer intent.

Three things follow from this. First, if someone on your team says "the intended use is..." they might be talking about user behaviour, or they might be using "intended use" as a loose synonym for intended purpose. You have to ask which. Mixing the two inside a single document is how regulatory files become incoherent.

Second, user behaviour matters for other reasons. Usability engineering under IEC 62366-1, risk analysis for reasonably foreseeable misuse, and post-market surveillance signals. But it does not determine the intended purpose under Article 2(12). A user deciding to use a fitness tracker to diagnose an arrhythmia does not make the fitness tracker a medical device. The manufacturer's claims do.

Third, the gap between stated intended purpose and observed intended use is where trouble lives. If users are doing something with your device that you did not intend, you have two choices: narrow your intended purpose and actively discourage the behaviour, or broaden your intended purpose and accept the regulatory consequences. Letting the gap persist in silence is not a choice. It is a pending enforcement problem.

Why the MDR cares about manufacturer statements, not user behaviour

The MDR regulates economic operators, not end users. The whole structure of Chapter II. Manufacturer obligations, authorised representative obligations, importer obligations, distributor obligations. Is built around the idea that someone is making a claim about a product and placing it on the market on the basis of that claim. The regulation then checks whether the claim is supported and whether the product meets the requirements that flow from the claim.

If the regulation keyed off user behaviour, nothing would be stable. A manufacturer could never know whether their product was a medical device because some future user might repurpose it. A Notified Body could never finish a conformity assessment because the "intended" use would keep moving. Classification would be untethered from anything the manufacturer controls.

The Article 2(12) definition solves this by making the manufacturer's own written statements the anchor. You speak, in the four enumerated places, and your words define the scope. Inside that scope, the regulation then imposes obligations: GSPRs in Annex I, clinical evaluation under Article 61, post-market surveillance under Article 83, vigilance reporting under Article 87. Outside that scope, MDR does not apply to your product at all.

This is why the Article 7 prohibition on misleading claims is the companion piece to Article 2(12). Article 7 prohibits, in the labelling, the instructions for use, the making available, and the advertising of devices, the use of text, names, trade marks, pictures, and figurative or other signs that may mislead the user or the patient about the device's intended purpose, safety, and performance. The manufacturer's statements must be honest. And they must also be deliberate, because they define what regulatory obligations apply.

The four sources Article 2(12) recognises

The definition names exactly four sources. Treat this as an exhaustive list and audit each one before you consider an intended purpose to be settled.

The label. Everything printed on the device itself or on the packaging. Short by nature, so every word carries weight. Medical claims on a label turn a product into a medical device regardless of what the IFU says.

The instructions for use (IFU). The longest-form statement of intended purpose. This is where the full medical or non-medical framing is set out, including the intended patient population, the intended clinical condition (if any), the intended environment of use, and the intended user. If the IFU says the device is for the "monitoring, prediction, prognosis, treatment, or alleviation of disease" in the sense of Article 2(1), you have a medical device, full stop.

Promotional or sales materials or statements. This is the source that bites startups. Website copy, brochures, landing pages, investor decks, conference talks, social media posts, press releases, pitch videos, app store descriptions, CEO interviews. Each of these can supply data on intended purpose. You cannot write "wellness only" in the IFU and "clinically proven to reduce blood pressure" on the landing page. The regulation reads both. Borderline Manual v4 (September 2025) consistently treats marketing statements as first-class evidence of intended purpose, not as a soft signal.

The clinical evaluation. Under Article 61, the clinical evaluation must demonstrate conformity with the applicable GSPRs when the device is used as intended by the manufacturer. The intended purpose as specified in the clinical evaluation must match the intended purpose in the other three sources. If the clinical evaluation assesses the device for a broader use than the IFU describes, you have contradictory intended purposes. And a Notified Body will find it.

A coherent intended purpose is the same story, told four times, in language appropriate to each medium, without contradiction. An incoherent intended purpose is the fastest route to a non-conformity finding.

How changing the intended purpose changes the regulatory path

Intended purpose is not decorative. It drives every downstream decision.

Whether the MDR applies at all. Article 2(1) defines a medical device by intended purpose. Remove medical claims and some products fall out of the MDR entirely and into wellness, cosmetics, consumer electronics, or general product safety territory. Borderline Manual v4 gives dozens of worked examples of how the same underlying technology can be a medical device or not, depending on what the manufacturer claims.

Classification under Annex VIII. Article 51 routes every device through the classification rules in Annex VIII, and those rules read the intended purpose closely. A measurement function, a diagnostic claim, a therapeutic claim, a monitoring claim for vital physiological parameters, a software claim about informing clinical decisions. Each of these shifts the rule that applies and often the class.

Conformity assessment route. Higher classes require deeper Notified Body involvement under the annexes referenced in Article 52. A narrower intended purpose that keeps the device at a lower class can cut the conformity assessment effort substantially. Legitimately, when the narrower purpose genuinely reflects what the manufacturer wants to claim.

Clinical evidence burden. Article 61 and Annex XIV require clinical evidence sufficient to demonstrate conformity for the intended purpose. Broader intended purpose, broader evidence requirement. Narrower intended purpose, narrower evidence requirement. Every extra claim buys extra clinical work.

None of this is gaming. It is applying the regulation as written. The manufacturer decides what the product is for, within the limits of honesty imposed by Article 7. The regulation then imposes the corresponding obligations. Deliberate intended purpose drafting is how resource-constrained startups avoid paying for regulatory work that their actual business does not need.

The anti-intended-purpose strategy

There is a case that sits at the extreme end of this logic. A company came to Tibor convinced they were building a medical device, expecting to enter certification. Tibor looked at the product, the underlying technology, and crucially the way they were talking about it in public. The technology did not have to be a medical device. The language around it was making it one.

Tibor ran what he now calls the anti-intended-purpose strategy. The work was deliberate and documentary. He analysed the four sources from Article 2(12) one by one and identified every statement that was moving the product across the line into the medical device definition. He removed medical claims from the website. He rewrote the product page so that it described what the product did without describing it as diagnosing, preventing, monitoring, predicting, prognosing, treating, or alleviating disease. The operative verbs from Article 2(1). He separated the product page from the scientific literature page, because scientific context about the general field is not the same thing as a medical claim about the specific product, and the two had been tangled together in a way that created the appearance of a medical claim.

Then Tibor documented everything. Not just the changes. The reasoning for each change, the mapping back to Article 2(12) and Article 2(1), the evidence that the product as now described did not meet the medical device definition. If a competent authority had ever asked, the file would have answered.

The product stayed on the market. The company continued generating revenue. Years of certification work and hundreds of thousands of euros were avoided. Not because anyone cut corners, but because the manufacturer's words were finally consistent with a product that was genuinely not a medical device under the MDR definition.

The anti-intended-purpose strategy is not for every company. For some products, the medical claim is the business. Strip it out and there is nothing left to sell. But for companies where the medical framing was a default assumption rather than a deliberate choice, the strategy can save the company years.

Common ways founders accidentally create a medical device

Across the cases Tibor sees, the same few mistakes keep producing accidental medical devices.

  • Writing investor deck slides with phrases like "clinically proven to reduce X" before the clinical evaluation supports that claim. Investor materials are promotional materials under Article 2(12).
  • Publishing blog posts and scientific literature on the company website that frame the product as addressing a specific disease or condition, then linking to the product from those posts.
  • Using the product name plus a medical verb in social media copy. "MyProduct helps diagnose..." is a classification change in eleven characters.
  • Citing clinical studies in a way that implies the product does what the studies describe, even when the product is not the subject of the studies.
  • Letting the sales team free-form the pitch because "marketing is not regulatory." Sales statements are explicitly covered by Article 2(12).
  • Writing an IFU with a narrow intended purpose and a website with a broad one, and assuming auditors only read the IFU. They read both.
  • Translating the intended purpose into other languages loosely, so the German and English versions imply different scopes.

Each of these is a self-inflicted regulatory expansion. Each can be caught in a pre-launch review of the four Article 2(12) sources.

The Subtract to Ship angle

Intended purpose is the first and highest-leverage subtraction point in the entire regulatory project. Before you build a QMS, before you draft a technical file, before you engage a Notified Body, before you scope clinical evidence, you write the intended purpose. If the intended purpose is broader than it needs to be, every downstream activity inherits the unnecessary breadth.

The Purpose Pass in the Subtract to Ship framework is where this discipline lives. You take the draft intended purpose and ask: what is the narrowest honest statement of what this device is for? Not the narrowest possible statement. Honesty under Article 7 is non-negotiable. But the narrowest that still reflects the product and the business. Then you align all four sources to that statement. Label, IFU, promotional materials, clinical evaluation. One coherent story.

A correctly narrowed intended purpose removes work from every later pass. The Classification Pass has a simpler input. The Evidence Pass has a smaller scope. The Operations Pass has a smaller QMS to run. The entire project gets cheaper and faster for the cost of one careful afternoon with the intended purpose text.

The opposite is also true. A broad, casual intended purpose silently inflates every other deliverable in the project. Founders who skip the Purpose Pass pay for that skip every month until certification, and sometimes for years afterwards.

Reality Check. Where do you stand?

  1. Can you read your current intended purpose statement out loud in thirty seconds, and does it match what your website says, what your IFU says, and what your clinical evaluation assesses?
  2. Have you audited the four Article 2(12) sources separately. Label, IFU, promotional and sales materials, clinical evaluation. And confirmed they tell the same story?
  3. Is there any sentence in your marketing, investor materials, or press coverage that uses the verbs from Article 2(1). Diagnosis, prevention, monitoring, prediction, prognosis, treatment, alleviation. In a way your IFU does not back up?
  4. If a Notified Body asked you to list every source that supplies data on your intended purpose, could you produce the list in an hour?
  5. Do you know, specifically, which words in your intended purpose are driving your Annex VIII classification rule, and what happens to the class if you change them?
  6. Have you ever deliberately decided NOT to make a medical claim that you could have made, because the regulatory cost outweighed the commercial value?
  7. Is there a gap between what users do with your product and what your intended purpose says. And if so, what are you doing about it?

Frequently Asked Questions

What is the difference between intended purpose and intended use under the MDR? Intended purpose is the defined term in MDR Article 2(12). The manufacturer's written statement of what the device is for, on the label, in the IFU, in promotional and sales materials, and in the clinical evaluation. Intended use is not an MDR-defined term; it is a colloquial phrase that often refers to how users actually employ the device. The MDR regulates based on intended purpose, not on user behaviour.

Where exactly is intended purpose defined in the MDR? In Article 2, point (12), of Regulation (EU) 2017/745. The exact text is "the use for which a device is intended according to the data supplied by the manufacturer on the label, in the instructions for use, or in promotional or sales materials or statements, and as specified by the manufacturer in the clinical evaluation."

Can marketing statements on my website really turn my product into a medical device? Yes. Promotional and sales materials or statements are one of the four sources Article 2(12) names explicitly. If your website makes a medical claim that meets the Article 2(1) definition of a medical device, the product is treated as a medical device, regardless of what the IFU says. This is a recurring pattern in the Borderline Manual v4 worked examples.

If users use my device for something I did not intend, am I responsible under MDR? The MDR regulates your intended purpose, not the user's behaviour. However, reasonably foreseeable misuse is part of risk management and usability engineering, and persistent off-label use patterns that you know about can create a compliance problem if you fail to act on them. Either by updating the intended purpose or by actively mitigating the misuse. Silence is not a defence.

Can I narrow my intended purpose to reduce the MDR burden? Yes, as long as the narrower intended purpose honestly reflects what you want to claim and sell. Narrowing intended purpose to bring a device into a lower class, or out of the MDR scope entirely, is a legitimate strategy when the narrower statement is genuine. Narrowing the IFU while still making broader claims in marketing is not a strategy. It is a non-conformity waiting to be found under Article 7.

Does the clinical evaluation really count as a source of intended purpose? Yes. Article 2(12) explicitly names the clinical evaluation as one of the four sources. Article 61 then requires the clinical evaluation to demonstrate conformity for the intended purpose of the device. If the clinical evaluation assesses a broader use than your IFU and marketing describe, you have a contradiction that a Notified Body will flag.

Sources

  1. Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on medical devices, Article 2(1) (definition of medical device), Article 2(12) (definition of intended purpose), Article 7 (claims), Article 51 (classification), Article 52 (conformity assessment procedures), Article 61 (clinical evaluation), Annex VIII (classification rules). Official Journal L 117, 5.5.2017.
  2. Manual on Borderline and Classification for Medical Devices under Regulation (EU) 2017/745 on medical devices and Regulation (EU) 2017/746 on in vitro diagnostic medical devices, Version 4, September 2025.

This post is part of the Device Classification & Conformity Assessment series in the Subtract to Ship: MDR blog. Authored by Felix Lenhard and Tibor Zechmeister. Intended purpose is the single most leveraged sentence a MedTech founder writes. Spend the afternoon it deserves.