To validate a MedTech idea before MDR, run eight steps in order: write the intended purpose sentence, identify the buyer, identify the user, quantify the patient benefit, interview twenty potential buyers, collect letters of intent from clinical partners, confirm the economics are ten to twenty times better for the buyer, and only then start regulatory planning. Skip any step and you will spend MDR money on a product the market does not want.

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

TL;DR

  • You cannot test the technology with patients before you certify it, but you can verify the market for it before you spend a euro on MDR. That is the whole game in MedTech validation.
  • The intended purpose sentence drives everything downstream. It defines whether you are a medical device at all, and under MDR Article 2(1) that one decision determines the entire regulatory path.
  • Buyer and user are almost never the same person in MedTech. The hospital buys. The clinician uses. The patient benefits. Validate all three.
  • The economic test is ten to twenty times better for the buyer, not the user. If the hospital does not see a clear operational or financial win, champions in the clinic will not save you.
  • Twenty buyer interviews and signed letters of intent from clinical partners come before any QMS, any technical file, any Notified Body conversation. Not after.

Why this matters for your startup

A few years ago I coached a founder in the education space who did something most founders refuse to do. He built a landing page. He had conversations. He sold the offer before it existed. People purchased, and then he built what they had just paid for. First quarter revenue was around EUR 50,000 on zero marketing spend. People were basically begging him to give them the onboarding. That founder did not have MDR to deal with, but what he understood is the only thing that matters at the start of any company: he answered the one question everyone had, and he charged money for the answer before he wrote a single line of code.

Now hold that story next to another one. A different founder spent EUR 1.8 million building a product that worked flawlessly in a lab and failed in the real world because users do not follow your process. They do whatever they think. The lab was clean, the protocol was elegant, and not a single user had been watched doing the actual work under actual conditions. Eighteen months of engineering, one and a half million euros burned, and a product nobody could use.

Here is the problem specific to MedTech. You cannot run the education-founder playbook directly. You cannot hand a prototype to a patient and learn from their experience, because the prototype is not certified and handing it over has legal consequences under MDR. But you can do everything else. You can verify the market. You can verify the buyer economics. You can verify the clinical workflow. You can line up your first twenty customers before you write a single SOP. That is pre-MDR validation, and if you skip it you will arrive at the Notified Body with a beautifully certified device that nobody wants.

What MDR says about the thing you are validating

The reason the intended purpose sentence is the first step of validation, not the tenth, is that MDR Article 2(1) ties the entire scope of the regulation to it. The regulation defines a medical device as:

"any instrument, apparatus, appliance, software, implant, reagent, material or other article intended by the manufacturer to be used, alone or in combination, for human beings for one or more of the following specific medical purposes..." — Regulation (EU) 2017/745, Article 2, point (1).

And Article 2 point (12) defines the 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" — Regulation (EU) 2017/745, Article 2, point (12).

Read those two definitions together and the conclusion is unavoidable. The sentence you write describing what your product does, and for whom, is the sentence that determines whether MDR applies to you at all, and if it does, which class you fall into. Validation therefore starts with that sentence, not with a prototype.

The eight steps, in order

Step 1 — Write your intended purpose sentence

Before anything else, write one sentence that names what the device does, for which patient population, in which clinical setting, by which user, and for what medical benefit. No buzzwords. No product names. No hedging. One sentence, plain language, under forty words.

If you cannot write that sentence, you do not have a MedTech idea. You have a vibe. A vibe is not a company. The sentence you write here will be used by a Notified Body to classify the device and by you to interview buyers. Both audiences will punish ambiguity. Write it, show it to three clinicians, rewrite it, show it to three founders, rewrite it again. This step costs nothing and saves months.

Step 2 — Identify the buyer

Who signs the purchase order? In MedTech that is almost never the person using the device. The buyer is a hospital procurement department, a GP clinic owner, a care home operator, an insurance payer, or in some cases a public tender authority. Each of these buys differently, on different cycles, with different approval chains.

Write the buyer down by title, not by institution. "Head of Procurement, University Hospital" is a title. "University Hospitals" is a market. You cannot validate a market. You can only validate a conversation with a specific person who has a budget line, a problem, and the authority to say yes.

Step 3 — Identify the user

The user is the person who touches the device in the clinical workflow. The scrub nurse. The ward physician. The sleep technologist. The lab assistant. The user is not the buyer, and the user's love for your product does not pay your bills. But the user's refusal to use your product will kill a purchase order faster than any procurement objection.

Map the user by role and by the specific task they currently do that your device will change. If you cannot name the task in one sentence, you do not understand the workflow yet. Go sit in the clinic for a day.

Step 4 — Identify the patient benefit and quantify it

The patient benefit is the reason the device exists under MDR Article 2(1). Write it down. Then quantify it. Not with adjectives. With numbers.

Time saved per procedure. Patients processed per shift. Daily hospital impact. Readmission rate changes. Length of stay. Cost per episode of care. Whatever the benefit is, find the number and confirm it with at least two clinicians who currently do the work. If the benefit is "better patient experience," you do not have a number yet. Keep digging until you do.

Step 5 — Interview twenty potential buyers before building anything

Twenty is the minimum, not the aspiration. Fewer than twenty and you are pattern-matching on noise. More than twenty is usually fine — keep going until the objections stop surprising you. These are not sales calls. They are discovery interviews. You are not pitching, you are listening.

I have watched too many founders skip this. One founder I coached went silent for eighteen months. I begged him to go out and talk to customers. For a year and a half he did nothing. No revenue, no LOIs, nothing to show investors. The company dissolved. I can name at least four similar stories. The silence is always the tell. Founders who cannot bear to hear a "no" from a real buyer will happily hear a "no" from the market eighteen months later, because by then the money is already gone.

Step 6 — Get letters of intent from real clinical partners

After twenty discovery interviews you should know which three to five institutions are genuinely interested. Go back to those and ask for a letter of intent. A letter of intent is not a purchase order. It is a signed document that says: if this device meets the specifications we discussed and is CE marked, we intend to purchase it under these conditions.

Letters of intent do three things at once. They force the buyer to make their interest concrete. They give you clinical partners who will participate in your clinical investigation and become your first real customers afterward. And they make you investable — no LOIs, no investors, no matter how brilliant the technology.

Step 7 — Check the economics — ten to twenty times better for the buyer

The rule is simple and nobody likes it. Your product must be ten to twenty times better for the buyer — the hospital, the payer, the clinic operator — not just the user or the patient. Ten times is the floor for adoption. Twenty times is where switching costs stop mattering.

Why so high? Because the buyer has an existing solution that works, a procurement process that punishes risk, and a budget already committed somewhere else. A two times improvement disappears in the noise. A ten times improvement gets a meeting. A twenty times improvement closes a deal. Do the math on time saved, patients processed, cost per outcome, and operational burden. If the math does not get to ten, you do not have a validated idea yet. You have a hypothesis, and you need to either change the offer, change the buyer, or change the device.

Step 8 — Only then start regulatory planning

If and only if the first seven steps pass, start regulatory planning. That means reading the intended purpose sentence back to an expert, classifying the device under Annex VIII, mapping the Notified Body options, sizing the clinical evidence strategy, and building the QMS against a real product, not an imaginary one.

Notice what this sequence prevents. It prevents you from paying for QMS setup before you know what device you are building. It prevents you from burning a clinical investigation on a device that cannot be sold. It prevents you from spending twelve months on technical documentation for a product with no buyer. Pre-validation is the cheapest insurance you will ever buy in MedTech.

The Subtract to Ship angle

Subtract to Ship is about removing work that does not move the needle. Pre-MDR validation is the purest form of that discipline. Every hour you spend validating is an hour that might save you a year of certification work on the wrong device. Every euro you spend on twenty buyer interviews is a euro that protects the hundreds of thousands of euros in regulatory spend downstream.

The instinct for founders with strong technology is to start building and figure out the market later. In non-regulated industries that instinct is expensive. In MedTech it is fatal. Subtract everything that is not validation until validation is done. Then subtract everything in your regulatory plan that does not trace back to a specific MDR article for the specific device you just validated. That is the whole method, applied to the first ninety days of a MedTech company.

Reality Check — where do you stand?

Answer these honestly. Not as a founder pitching. As a founder deciding whether to commit another year of your life.

  1. Can you write your intended purpose sentence in under forty words, without buzzwords, right now?
  2. Do you know the name and title of the specific person who would sign the first purchase order, at a specific institution?
  3. Can you describe the current clinical workflow your device changes, in one paragraph, without looking it up?
  4. Do you have a number for the patient benefit, confirmed by two clinicians?
  5. Have you completed twenty buyer interviews with written notes?
  6. Do you have at least three signed letters of intent?
  7. Does the math get you to ten times better for the buyer, with numbers you can defend?
  8. Have you resisted the urge to start QMS work before all of the above are done?

Zero or one yes means you are not ready to spend money on MDR. Two to four means you are partway through validation and should finish before anything else. Five to seven means you are close and should close the remaining gaps this month. Eight means start regulatory planning today, with an expert, and do it properly.

Frequently Asked Questions

How do I validate a MedTech idea without a working prototype? You validate the market, not the technology. Use the intended purpose sentence, buyer interviews, clinical workflow observation, and letters of intent. The prototype question — does the technology work — comes later in a controlled feasibility phase. The market question — does anyone want this — has to be answered first, and it can be answered with conversations and documents alone.

How many buyer interviews are enough before I start MDR work? Twenty is the minimum. Keep going until the objections stop surprising you. Fewer than twenty and you are drawing conclusions from too little data. The interviews are cheap compared to an MDR project that has to restart because the intended purpose was wrong.

Do letters of intent have legal weight in a MedTech sale? A letter of intent is not a binding purchase order, and you should not treat it as one. Its value is different. It confirms that the buyer has internalised the proposition, it commits them to a conditional intent, it gives you a clinical partner for your evidence generation, and it signals investability to funders. A well-written LOI is worth more than a hundred "this looks interesting" emails.

Why ten to twenty times better for the buyer, not the user? Because the buyer is the one who signs, and the buyer lives inside a procurement process that penalises small improvements. The user's enthusiasm matters, but user love has killed many MedTech companies whose buyer economics did not work. We have seen clinicians begging to have a device in their department while the procurement committee refuses to budget it. Get the buyer math right first.

When is it safe to start spending money on QMS and technical documentation? When steps one through seven are done. Not before. Starting QMS setup before you know the device definition is almost always wasted work, because the intended purpose and classification drive the scope of the QMS, and both of those come out of validation. Build the QMS against a validated product, not an imaginary one.

Sources

  1. Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on medical devices, consolidated text. Article 2, points (1) and (12). https://eur-lex.europa.eu/eli/reg/2017/745/oj
  2. Regulation (EU) 2023/607 amending Regulations (EU) 2017/745 and (EU) 2017/746 as regards the transitional provisions for certain medical devices and in vitro diagnostic medical devices. https://eur-lex.europa.eu/eli/reg/2023/607/oj
  3. Felix Lenhard coaching case files, 2021–2026, anonymised for publication.

This post is part of the MedTech Startup Strategy and Product-Market Fit cluster in the Subtract to Ship: MDR blog. Authored by Felix Lenhard and Tibor Zechmeister. If your validation sequence is underway and you want a regulatory sparring partner for the step eight conversation, Zechmeister Strategic Solutions works with founders at exactly that point in the journey.