Rarely has my “regulatory life” been as intense and as rich in insights and progress as recently. Three brief examples from just last week that also affect you: ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­    ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­  
Johner Institute

Medical Device Briefings

Issue 07/26

Why Your Developers Aren’t Developing

Hi there,

Rarely has my “regulatory life” been as intense and as rich in insights and progress as recently. Three brief examples from just last week that also affect you:

 

1. Automated TD reviews at Notified Bodies
Last week, a representative of a Notified Body attended my workshop “AI Power for the Development and Approval of Medical Devices,” where I demonstrated how alarmingly effective automated reviews of technical documentation (TD) have become.


He announced that his Notified Body would also pursue this path. In one of the upcoming podcast episodes, he will speak about it. So keep following the Medical Device Briefings.


But the consequences are already clear: full reviews of technical documentation will become the new standard. For AI, it makes little difference whether it reviews individual documents, individual files, or an entire TD.


2. My experiment: tailor-made software (n = 1) in just a few weeks
I was somewhat overwhelmed by the feedback on my planned experiment, which I announced here last week. There were many words of encouragement and well wishes. One comment, in particular, made me pause:

 

“I’m curious and secretly hoping for failure. Not out of malice, but out of fear for the future. 😀”


I understand - and share - that fear.


Consequence: If we would rather not become victims of AI, we must seize the unique opportunities it offers. In Johner’s AI Lab: The Expedition, I show my path.

 

If you would like to use it as a blueprint, you’ll need to register quickly - we have only a few places left.


3. Increasing development speed
Another question raised during the workshop was how to accelerate development. The obvious implication: with AI.


I have given this considerable thought.


My answer is neither comfortable nor related to AI - as you will see in today’s Executive Corner.

 

 

Regulatory Updates 

a. UK

The UK’s MHRA has launched a consultation on whether the CE marking should continue to be accepted indefinitely in the United Kingdom.

 

Common sense - or a political U-turn?

 

After Brexit, there was much talk about introducing distinct British regulations and moving away from EU-style bureaucracy. Yet no truly exemplary regulatory framework has emerged.

 

With the UKCA marking, the United Kingdom preserves its regulatory sovereignty - without reinventing the game for companies that wish to continue selling CE-marked products. This leaves room for innovative regulatory approaches - which we are all hoping for.

 

b. Team-NB on IVD reagents

In a new position paper, Team-NB specifies the evidence expectations for combined use of IVD reagents with other products.

 

The paper distinguishes between “closed systems” (clearly defined combinations with specified products) and “open systems.” In the latter case, manufacturers must define permissible combinations via critical product specifications or characteristics - without naming a specific product.

 

Overall, there is nothing groundbreaking, and it aligns with the Johner Institute’s understanding. We find the following particularly helpful:

  • The emphasis on a risk-based approach (ISO 14971)
  • Performance evidence for each combination or justified transferability
  • Clear IFU/labeling statements
  • The clarification that when RUO or General Laboratory Use products are part of the workflow, the responsibility for evidence lies with the IVD manufacturer (including a PMS focus on combination workflows)

My colleague Dr. Sophie Bartsch has updated our technical articles accordingly (see below).

 

c. AI Implementation Act

The German Cabinet has adopted the draft AI Implementation Act.

Its key content includes the designation of:

  • the supervisory authorities (including BaFin),
  • the component authorities for Annex I, Section A products (including medical devices and IVD): the same as before, i.e., ZLG,
  • the conformity assessment bodies: unchanged (good news!),
  • the composition of the “AI Market Surveillance Chamber” (including the Federal Network Agency, which is also assigned numerous additional responsibilities under the AI Act),
  • the powers of the market surveillance authorities.

According to the draft, the administrative burden for public authorities will increase by approximately € 49 million per year. No additional burden is expected for citizens or businesses, as this was already created by the AI Act itself.

  

 

Executive Corner: Why Your Developers Aren’t Developing

a. The wrong answer?

“How can we make our development departments more productive?” I hear this question often - most recently at last week’s hands-on workshop.

 

The participant expected the obvious answer: “Use AI, automate, digitize.”

 

That answer isn’t wrong. But it falls short.

 

Before we talk about AI, we should address the real issue: developers are not unproductive because they lack tools. They are unproductive because they don’t get to develop.

 

b. Diagnosis Part 1: What the Organization Does

Issue 1: Meeting Culture as a Time Destroyer

A typical developer’s calendar reveals the issue: daily stand-up, weekly, sprint review, sync meeting, department meeting, town hall, and information session. The problem isn’t any single meeting. It’s the sum of them.

 

Development requires a “creator schedule”: long, uninterrupted blocks for focused work. What developers typically have instead is a “manager schedule”: fragmented days squeezed between appointments. Real development does not happen in the 45 minutes between two meetings.

 

calendar-small

Issue 2: The Granularity Trap

The idea sounds reasonable: specialization, clear roles, and defined interfaces.

 

The side effect is often overlooked. Every additional person in the process increases the number of communication paths. More heads do not automatically mean more output. Often it means more meetings about the work - instead of actual work.

 

Issue 3: Bureaucracy No One Questions Anymore

Travel requests that require a degree in form-filling. Procurement processes with five approval levels. QM requirements whose origins have disappeared into the fog of corporate history.

 

Each individual requirement probably once had a reason. But does it still apply? Anyone who honestly reviews their QM system (as we do) often finds a surprising amount of ballast that no longer has any regulatory foundation.

 

Issue 4: Unchecked Complexity

Product variants, legacy versions, and special solutions for individual customers: all of it must be maintained, documented, and supported. This complexity ties up development capacity without creating value. The worthwhile question is: Which variants truly contribute to the business? And which ones still exist simply because no one dares to switch them off?

 

c. Diagnosis Part 2: What Developers Themselves Contribute

Now it gets even more uncomfortable - but necessary:

 

Issue 5: Acceptance of the Status Quo

Many developers have grown used to the situation. They accept the flood of meetings instead of questioning it. That’s understandable: meetings offer something seductive. Visible activity without measurable responsibility. You attended, you contributed, and you were busy.

 

Issue 6: Avoiding Cognitive Strain

Real development work is the opposite: quiet, invisible, and cognitively demanding. Solving complex problems means pushing your limits. That’s not for everyone.

 

In large organizations, it’s easy to hide. An eloquent contribution in a meeting often makes a stronger impression on a manager than an elegant technical solution that no one sees.

 

Anyone who simulates productivity instead of delivering it doesn’t need better tools. They need a different self-concept.

 

Interim Conclusion

reasons-developoment-problems-small

d. What Leaders Can Do in Concrete Terms

  • Radically review meetings. Every recurring meeting must have a defined work outcome. Can’t define one? Cancel it.
  • Protect time periods. Declare “deep work” blocks in the calendar as untouchable. At least four hours at a time, several times per week. And no, that’s not a luxury. It’s a prerequisite for engineering work.
  • Shrink teams. Fewer people per project means fewer interfaces, less coordination, and more focus. Small teams with clear accountability outperform large teams with diffuse responsibilities.
  • Declutter QM. Put every requirement to the test: Which regulatory requirement is behind it? None? Remove it. (We help with decluttering.)
  • Reduce complexity. Clean up the product portfolio; consolidate variants. Have the courage to simplify.
  • Correct incentives. Make the quiet high performers visible and value them - not just those who shine in meetings.

e. And Where Does AI Fit In?

Yes, AI helps. But not as a miracle cure - rather as an amplifier for people who are already allowed to work productively.

 

AI automates routine tasks: documentation, research, and preparatory work.

More importantly, it enables individuals to take on tasks that previously required several specialists. The result: smaller teams, fewer interfaces, less coordination - and therefore more real development time.

 

But the prerequisite remains: the organization must protect that periods. Otherwise, you simply automate distractions more efficiently.

 

f. The Litmus Test

One question to take with you: How many hours per week do your developers actually spend developing?

 

If the answer is below twenty, you don’t have a technology problem. You have an organizational issue.

 

The solution does not begin with adding new tools, new processes, or new meetings. It begins with subtraction. 

Practical Tips

a. Use SVG as an Output Format

Do you need a precise drawing? Graphics or diagrams where you can control the details? Then let AI help you - but don’t generate images.

 

Tip 1: Don’t have Mermaid or PlantUML diagrams generated. Instead, use SVG.

 

Advantages

  • Infinitely scalable and therefore sharp at any zoom level
  • Small file size
  • It can be displayed directly in the browser
  • Convertible into other formats such as PNG or JPG
  • Pixel-perfect control

b. Use Output as Input

If you’re not completely satisfied with the result from Tip 1 - for example, because text extends beyond a border - proceed as follows:

 

Tip 2: Provide the AI with the result (e.g., as a PNG) and describe the issue. The AI can “read” images and correct errors. There is no need to manually fiddle around in SVG files. 

Best regards,

Christian

Founder Johner Institute

Upcoming seminars

Practical Workshop “AI Power for the development and approval of medical devices”

March 10, 2026

Learn in just one day how to use large language models (LLMs) specifically in developing medical devices – from the initial idea to post-market surveillance. 

This practical workshop with product expert Mario Klessascheck and regulatory and AI expert Prof. Dr. Christian Johner will teach you specific methods that will enable you to drastically reduce the time needed to develop and document your medical devices while remaining compliant with the law.

 

Sign up here

 

 

Computerized Systems Validation (CSV)

March 11, 2026

In this seminar, you will learn how to validate your computerized systems (e.g., ERP, compilers, or self-developed EOL testers) in a risk-based, fast, and regulatory-compliant manner and how to create the necessary documents so that you can confidently face future audits.

 

Sign up here

 

 

IEC 60601-1 Safety of Medical Electrical Equipment

March 23-24, 2026

In this seminar, you will develop a comprehensive understanding of the basic concepts of the IEC 60601 family of standards. You will learn how to successfully implement the relevant requirements in practice using various exercises and how to create standard-compliant documentation.

 

Sign up here

 

New and updated articles

Technical Article “Class IIa Medical Devices”

Class IIa medical devices are among the most common products on the European market. Our latest article explains which products fall into this class, which classification rules are relevant, and what regulatory consequences result from this classification.

 

Read the technical article

 

Updates

We update technical articles every week. Particularly relevant this week:

  • General laboratory equipment: What manufacturers and laboratories need to know to avoid trouble and unnecessary effort
  • Laboratory products “For Research Use Only” (RUO)—often a dangerous claim

Support from the Johner Institute

The Johner Institute's Product & Service Portfolio

 

The Johner Institute supports the medical device ecosystem (manufacturers, authorities, notified bodies, operators) in all regulatory challenges.

  • Strategic Consulting
    e.g., regulatory and clinical strategies for global device approval, quality systems, technical documentation, artificial intelligence
  • Business Process Outsourcing
    e.g., regulatory monitoring, post-market surveillance, clinical evaluation
  • Software (Enterprise SaaS)
    e.g., for regulatory monitoring, post-market surveillance
  • Product Testing
    e.g., clinical investigations, usability tests, pentesting, biocompatibility tests
  • Training
    e.g., seminars and e-learning
  • Taking on Regulatory Roles
    e.g., QM representative, legal manufacturer, in-country representation

Prof. Dr. Christian Johner

 

Johner Institut GmbH

Reichenaustr. 39a

78467 Konstanz

 

 

+49 (0) 7531 94500 20 
info@johner-institute.com
www.johner-institute.com

Privacy Policy   |   Subscription Settings

Johner Institut GmbH, Reichenaustr. 39a, Konstanz, Baden-Württemberg 78467, Deutschland

Manage preferences