Legal Tech StartUp Focus Podcast

From Copenhagen to Global IP: One Founder's Journey to Fix Patent (and other IP) Renewals

• Charles Uniman

Share episode

What if missing a payment by one cent could cost a company its billion-dollar patent? This eye-opening conversation with Mads Vibor Jørgensen, CEO and co-founder of PatentRenewal.com (https://www.patentrenewal.com), reveals the hidden complexities of global intellectual property management and how specialized technology is transforming this critical but overlooked field.

Most people assume patent renewals work like domain names – just enter your credit card information and you're done. The reality couldn't be more different. As Mads explains, companies seeking global protection must complete separate renewals annually in up to 240 jurisdictions, each with unique processes, laws, and payment requirements. One small error can result in losing valuable intellectual property rights worth millions or even billions.

Through his experience leading a global organization for young engineers and scientists, Mads discovered that the world's brightest minds were being bogged down by bureaucracy rather than focusing on solving important problems. This insight led him to create PatentRenewal.com, which combines specialized legal technology with sophisticated financial systems to navigate the labyrinthine world of global patent renewals.

The financial complexity is staggering – payments must be in exact local currencies with precisely formatted information. In Japan, renewals require physical letters using the Japanese imperial calendar system. These peculiarities make standard payment solutions inadequate and create opportunities for hidden fees, particularly in currency exchange where providers might charge 2-5% on transactions across a $150-200 billion market.

What's particularly fascinating is PatentRenewal.com's thoughtful technology approach. Rather than rushing to implement a wholly generative AI system, PatentRenewal.com (i) built a more hybridized system that features an expert system-first set of tools that optimize for reliability and (ii) combined those tools with a parallel generative AI system to perform follow-on auditing functions. This hybrid model offers valuable insights for anyone building mission-critical legal technology.

Whether you're a legal tech entrepreneur, IP professional, or innovator protecting your ideas, this episode offers valuable insights into building global solutions, market education challenges, and the critical importance of hiring exceptional talent. 

Subscribe now to the Legal Tech StartUp Focus Podcast (https://legaltechstartupfocuspodcast.buzzsprout.com) to hear more conversations with pioneers transforming the intersection of law and technology.

Speaker 1:

Welcome everyone. This is your faithful podcast host on the LegalTech Startup Focus podcast, charlie Uniman, honored and happy to have as a guest today Mads Viborg-Jorgensen I don't know if I got the pronunciation of his last name right, but, alas, mads. Mads is a helming and a founder at the patentrenewalcom app. We're going to hear all about what that app does and how Mads got to be a creator of that with his colleagues, and learn more about video tech startups from Matt Welcome.

Speaker 2:

Thanks, charlie, I appreciate the intro Name was all fine, no worries.

Speaker 1:

Also yeah.

Speaker 2:

I'm a mess. I'm CEO and the co-founder of of patent rentalcom. Uh, happy to to say a little bit about my story of founding the company and what we do excellent, excellent.

Speaker 1:

Mads is calling in from copenhagen. I am, uh, right outside new york city. So, as is often the case, we have an international cast um. But, mads, do tell us about where your story begins when it comes to being one of the originators of Patent Renewal.

Speaker 2:

Yeah, of course. So my story starts a few years before we founded Patent Renewal. So I, at the age of 18, became the president in the world's largest organization for young engineers and scientists, where I was based here at Copenhagen as well, and there I got super humbled, even though I like to think of myself as a smart guy on my intelligence, because my co-volunteers are still two today some of the smartest people I've met today Pretty sure I've met a Fusion Nobel laureate or two. That was a very learningful experience for me, because what I learned was that these people they just really, really want to work on solving the world's most interesting problems, but they actually hate working in leadership roles. They don't want to spend time on all of the fundraising aspects of science and problem solving. They don't want to spend time on bureaucracy. So in that leadership role I also learned that scaling a non-profit, only volunteer organization globally is not very scalable at all.

Speaker 2:

So I decided to study computer science and in that opportunity, after work, volunteering 40 hours a week it was a bit boring decided to study computer science and in that opportunity, after work, volunteering 40 hours a week it was a bit boring, just to study. So I tried to get a job in San Francisco at Google and Facebook, but I actually ended up saying no to work in one of the largest banks in Europe, doing a kind of entrepreneurship journey where I got to get my roots in entrepreneurship. Where I got to get my roots in entrepreneurship but as I graduated I know I didn't want to work in a bank. I wanted to help these people, see if I could help them solve the most important problem faster, because I know they would be much better at it than myself. So my co-founder, jesper, who is a serial entrepreneur within innovation management, and I we went out and interviewed hundreds of companies and a lot of these people, both in mine and his private network, and to our surprise, they said that one of the absolute most important things for them, when they are truly solving a unique problem no one else has touched, is intellectual property and especially patents.

Speaker 2:

So that was a bit puzzling for us and as we started looking into the industry, we figured out this whole ecosystem for protecting and helping the innovators and inventors of the world hadn't really been innovated itself for ages. So there was plenty, plenty full of problems to solve with an IP and with our backgrounds within building software as a service technology and some of my background in the banking sector, it seemed very obvious to try and create a technology company that would solve some of these problems, and we had to start somewhere. Being 22 at the time when we founded the company both of us we needed to focus on one thing, and with my kind of fintech background from the banking sector, renewals seemed like a great place to start, because there was so much money to save for the inventors. There was some combination of legal tech and fintech, which were some of our backgrounds, and it was also just a pain area that a lot of these inventors communicated. So that's kind of how we got into all of it.

Speaker 1:

Good story. I guess Copenhagen is pretty well known for Nobel matters. Niels Bohr was from Copenhagen, wasn't he Exactly? I think he won the Nobel Prize, I'm not sure I know. Einstein did finally. But yeah, I don't know that you're really still involved with, or were ever involved with, quantum mechanics, but happy to have the connection.

Speaker 1:

I admire all the work that's being done in fintech and legal tech as much as I admire the work that's being done in quantum physics. How's that for a statement? But one thing that I have found interesting and Mads and I did have a chance to talk a little while ago in a separate call is the combination and we'll get to this in greater detail of fintech aspects, of legal tech aspects in patentrenewablecom's portfolio of services and offerings. So tell us a little bit about that. Tell us what Patent Renewable does I guess the name sort of gives it away. Tell us how what initially might seem to be a rather niche product is nevertheless one that is ripe in a large market, and what patent renewal does to distinguish itself from some of the other vendors in the space.

Speaker 2:

Yeah, I'm happy to. So yeah, we have the the just to comment on. I love quantum mechanics myself. I've always been a personal hobby and we, no sport, got a whole institute for for science named after him here in copenhagen, where I have some friends who are phds and postdocs. So always great inspiration, um.

Speaker 2:

But going back to to pattern renewal uh, what, what, what? I think we had an impression of not knowing much about IP. Founding the company was that renewals were pretty basic. I thought, you know, when I started looking into this, I could just build an entire tech stack in six months. Wow, have I learned since then?

Speaker 2:

The big learning is that I think a lot of people have the impression is that if you want to get a patent, you just apply for it and then maybe you know everything is kind of centralized. It's maybe like renewing a domain you just put in your credit card. Everything kind of happens automatically. You don't have to do much. The reality is that if you want global protection of your intellectual property or patents generally speaking, especially for patents, you need to renew it every single year separately in every single country. So if you just have a convention, you might need to do 200 renewals a year in 200 separate countries with 200 separate processes, laws, procedures and payment methods to do it. So it becomes severely complicated, which means that basically no other provider than our company today are doing it with technology in person, country by country, which makes this process very, very expensive and not very efficient and not very centered for the innovators of the world and saving their time and money.

Speaker 2:

So what we have built is this complicated legal tech software that understands, all around the world, in the 240 jurisdictions, you can get an intellectual property right how to do renewals.

Speaker 2:

Then, additionally, we have built a financial software where we cannot just use a you know restrictions you can get an intellectual property right how to do renewals, uh, then additionally, we have built a financial software where we cannot just use a you know standard, modern provider like riverdoodlewise. We've had to build our own fintech software to send money to the different 240 governments around the world, because if you just send one cent too much or too little, they'll just reject the renewal application and, worst case, people will lose their right. So we've really had to optimize for building a very efficient and safe product, but obviously also something that scales. So I think that was the biggest realization for us and why there's such a big opportunity here because no one else has dared diving into kind of building the product we have and the company has been going for eight years now and we still haven't seen anyone else try to build what we've done.

Speaker 1:

Quite a boat, quite a defensive boat, very, very impressive. So I gather it's not as easy as renewing your domain name, which I do every year with just a click of a button. You have that multi-jurisdictional aspect to it and then there's that fintech aspect to it. Let's delve into the latter a little bit. We'll get to the more legal texture and stuff in a minute. Why can't people just is it merely I'll finish the thought why can't people just use Stripe or Revolut or something of that ilk? Is it merely a matter of being a penny off in your payment to the jurisdiction?

Speaker 2:

I think there's a lot of factors. One of them is the penny offer on. Generally, when you do global transactions, it's not super reliable like getting the exact amount on the bank account and they always need it in the local currency. You can never just send them dollars or euros or pounds. You need to send them in the in Chinese yen or Brazilian euro or whatever the patent office in the local country has as a currency. So they are super sensitive about that. The other thing is that you need to give them a very uh integrate message along with the payments, which needs to be formatted in a correct way. If you just again put one character incorrectly, if you use the wrong name of the company, you're renewing the patent for not even the wrong name of the company. You're renewing the patent for not even the wrong name of the company, but the wrong name that they have in their register. So if they made that typo back, then you have to use the typo name.

Speaker 2:

They have so many of these nitty-gritty like small processes and in many countries just making the payment is not enough.

Speaker 2:

There's no real systems for it, it's just something that happens kind of async. So in Japan, for example, we have to send a physical letter every time we have to make a renewal that fills out a standardized form where we have to use the Japanese year system, the year of the Kaiser, so we cannot write 2025. We have to write five years since the name of the current Kaiser has been established. There are so many of these things that we have to go through and then, other than just doing the work, which is the most difficult part, in many cases, becoming compliant to even do it is a very, very long process that's taking us years in many countries, and there are still a few countries where we haven't finalized the process because we need to make subsidiary companies, we need to have local employees and go various lengths to get there, and that's something we're fighting for in the last few places, but at this point, we're handling more than 90% of our renewals directly.

Speaker 1:

Yeah, I get the picture and as a former lawyer, and as many of our listeners are former lawyers, the devil is in the details and Lord knows you can't blow that patent Renaud. On the more legal tech side, could you have done this, could anyone have done this, without some form of artificial intelligence contributing to the process?

Speaker 2:

No, so I think one of our challenges is that a lot of this information about how to make renewals is not public, and there are so many edge cases in the renewal process that, even though if you would have perfectly read the processes and the law on the patent offices websites and registers and the government's websites which, by the way, most of the time is in the local language and not in English Even if you had done that to perfection, what you'll learn throughout the years of managing the renewals is that they have forgotten to specify a lot of edge cases. So and they're not good at making case law where they describe all of these you know edge cases when things go wrong. So, as we have built the software over the years, we've had to learn and take all of these edge cases when things go wrong. So, as we have built the software over the years, we've had to learn and take all of these things into account. So the way we've built our software, or the core software that handles renewals, is actually not with AI.

Speaker 2:

Even though I was teaching, co-teaching AI courses at my university and in the banking world I was working with AI as well we decided to not do that as the core product. Of course we've tried to use AI across the entire company to work more efficiently and we have kind of an AI renewal model that's faking renewals to kind of audit and make sure the core model is doing what it's supposed to do. But the core model is just, you know, good old super rigid software. Because in our scenario, if we just miss one out of 10,000, one out of 100,000 renewals, it could worst case mean that our customer would lose their patent, which is just unacceptable. That patent, just in one country, could be worth a billion dollars if it's the iPad or if it's a Sempic or if it's one of these major inventions.

Speaker 1:

No, I appreciate that, and Mads and I have talked, and Mads may be aware that, while I am a great fan of what can be done with generative AI, the latest and greatest flavor of various forms of artificial intelligence that have reached legal tech and the law, I do think that there is still plenty of room, and in fact, room to hybridize the use of AI with rules-based systems, which is one way of describing good, old-fashioned AI, hard-coded AI.

Speaker 1:

And you know, we all know about the hallucination problem with generative AI. No matter how hard they work on it, it seems to be in a radical pull to some extent. And, moreover, generative AI is, to some extent, inherently probabilistic and you just can't rely on probabilities to do what Pat Renuto is trying to do renewal I was trying to do. So it's refreshing, in a way, to be able to talk to Mads get on my hobby horse about the need to somehow combine with the generative flavor of AI those rule-based systems that can ensure some sort of deterministic use of artificial intelligence that can be brittle, and I did hear you say, did I not? There's a flavor of AI that you're using to sort of supervise the code itself.

Speaker 2:

Yeah. So beyond the main model we've basically tried to build, build an actual AI model that could do renewals. The problem is that we can only ever get it to be 99.9% accurate. But you know we wanted this up. It's just too many for us. But it's a good kind of self-auditing system because if that system says our system is about to make a mistake, if that system says our system is about to make a mistake, then we can put our team of experts to kind of double-check that everything is okay. And then we have kind of a you know self-auditing two softwares that are self-auditing each other.

Speaker 2:

And I still think the probability that the models will become good enough to handle these renewals globally like we're doing is close to zero. But the model we're training to do renewals as a like pure generative AI model is becoming better and better and it means that we then preemptively might catch if it flags a hundred things. It might only be right in two out of a hundred times, but those two out of 100 times we can warn our system in something going wrong or have a learning without making a mistake. That just makes our product so much more solid and it means that our product is so much more advanced and more safe than using a local expert at doing the work, because every time, because it's not generative, uh, every time something goes wrong or we have a learning, the software learns from it and will never make a mistake again, whereas humans innately will get sick, get fired, go on vacation, unfortunately also pass away, uh. So so many things can happen that are unpredictable when you're relying on, on people, uh, whereas our software will just always give the perfect answer. And then we even have another software to audit it, and we have also the best experts at global renewals in the world to additionally audit the software.

Speaker 1:

Well, you've sort of flipped the switch. You know, I often think of the hard-coded symbolic AI as acting as a check on the generative. Hard-coded symbolic AI is acting as a check on the generative, but I haven't thought and you've just described a situation where the generative AI getting better and better, but never, you know, only asymptotically approaching zero errors not good enough can still serve as a check on some of the hard-coded stuff. Very interesting, that's a new thought for me when I have to think about that. But having them work together is, I contend, in some way the best of both worlds. That may be a bit too simplistic a way to put it, but that's my hobby verse. That's where I may be a bit of a contrarian. How long has Pat Renaud been in business? How long have you been doing this?

Speaker 2:

It's been eight years since we started working on the company and in three weeks time it's officially our eight year registration anniversary. So it's been some time going now and I feel like the more I work in the industry, the more passionate I get. And as a company, you can hear our vision. It's not just solve renewals, it's much broader than that. So I'm also excited for us as a company to you can hear our vision is not just solve renewals, it's much broader than that. So I'm also excited for us as a company to hopefully already at some point next year, starting to slowly expand our product horizon into many more intellectual property services.

Speaker 1:

Good, wonderful. Yeah, I wondered about expansion. I was going to ask that question we don't have to get too deeply into your plans that question, we don't have to get too deeply into your plans but it makes sense that there are all sorts of other sticky, very persnickety renewal-type issues and other IP-related issues from which humans can be relieved and some sort of coded maybe AI, backstop, generative AI backstop solution can be made available. So good for you. I look forward to hearing from you, matt, about some of the market expansion, what you're going to be doing, not just geographically, of course, but in terms of product offerings, if I were to ask you. We've already covered some of the things that I find myself being a contrary about and my hobby horse, any of those that occur to you that you want to discuss, where you may hold a view that's a little bit different from what is standard in the fintech or legal tech verticals, or something that you find is particularly attractive and somewhat neglected as an idea that you'd like to put forward.

Speaker 2:

I think I have a lot. One that comes very intuitively, because if you're working with even just a Danish company, they will not just have a patent in Denmark. They will have their patent not necessarily everywhere around the world, but in at least like 50 to 80 countries if they're like a decent sized global company. So we've kind of had to build our company globally from day one, and that means that already six, seven years in we had customers from over 40 different countries. We have more market presence in the UK and the US than anywhere else. But working so internationally has just given us so many learnings as a company that I think that is something untraditionally, especially legal tech companies who are so focused on figuring out and perfecting law in one country with their solution. I think there's so much to learn of working how do you say, globally from day one, both in terms of the service you're delivering, but also in terms of the customer basis. So I think that's a good starting point.

Speaker 2:

But I think, just to underline it, the conversations we've had already now just only relying on generative AI if you want to deliver a very sensitive or mission critical service in legal sec, I think could get you off to a false hope, at least at current state. You know my hope and dream is that generative AI is going to kind of become so good five, 10 years from now, or even sooner, that you can rely on it. But I just think right now that's just too optimistic and you might end up losing customers and in most spaces in legal and especially with something like essential as renewals of IP trust and safety is just the number one fate. So I think if your product line is just only a generative ai solution, uh, you might be gambling with that trust yeah, I, I agree.

Speaker 1:

Um, people shouldn't be surprised hearing that. For me, even what I just said about my view of how ai ought to be hybridized a bit. The way I put it is you can vibe code to some extent and you can vibe a brainstorm to some extent, and generative AI can do a great job vibing with you. But there's only a very small extent, I think, where you can vibe lawyer, you know, maybe brainstorming, maybe coming up with various takes on arguments that could be made or negotiation strategies. But beyond that and I hope to be proven wrong someday, but given the way the current architecture of gender works, I don't think that's going to get us to it where we get beyond the vibes and we get down to a level of precision that could leave a lawyer of some of that worry over the stuff that breaks prettily but nevertheless has to be hard-coded in. I agree with, I think even I was a corporate lawyer. I know litigators pretty well. There's some vibing to be done there. But you need to be and I hope the industry is coming around to this you have to think of marrying the two general approaches good old-fashioned AI, symbolic AI, ontology, graph-based AI, which can by itself speed up the work stupendously with generative eye that can overlay language fluency and human mimicry. I think we'll get there when AI becomes conscious.

Speaker 1:

Other frightening things may happen, I don't know, but I think we're a ways off from that. One of the things I like to do when I have the leader of a legal tech, an in-tech company, on is have them talk a bit about their learnings when it comes to running the business. You've been in business for quite a number of years. Vulnerably speaking, you're an OGogy and old guard person in legal tech. If you could ask yourself what you would have done differently at the outset, or if you could think about things that you could have improved upon earlier in the journey, or things that you didn't know about until you came across the edge case and had to handle it, what sort of business-related advice, whatever area of business, would you like to impart to the audience, many of whom are legal tech startup leaders?

Speaker 2:

Yeah, I think one thing a bit related to kind of maybe a contrary on you too is related to why we're doing so much fintech, like it might be that like the revenue opportunity for like the entire global market for a company like mine on intellectual property renewals is only around 10, 15 million US dollars a year. 15 million US dollars a year. But when doing that, you have to send a lot of fees to the patent offices in order to pay for the renewal in administrative fees and various things. So the transaction volume is actually closer to 150 to 200 billion a year. That's a lot of money. So, other than obviously us being able to save some money for the customers, there's obviously a big opportunity on optimizing things so that our solution is so efficient that we can be more profitable as a company.

Speaker 2:

And what I've seen when I go and work with some of the largest intellectual property law firms in the world some of the largest patent owners in the world is that most people just have a very, very vague sense of lists. Is that most people just have a very, very vague sense of lists. People still have the mentality that if I'm getting charged just $5 per renewal per country, then I'm getting a super cheap setup. But then what they're not noticing is that their provider might be charging them 2% to 5% in a foreign exchange rate fee. But if you're taking 2% to 5% off of $200 billion, versus $2 to $5 per renewal per country, obviously that's a lot more expensive. So our industry is really suffering from a lot of hidden fees and that's why transparency is so important to me and for the company. But I think overall in other industries, when large transaction volumes are involved as a part of the service, there's a huge opportunity here and the lack of education in the market has been jaw-dropping for me. So I think our main marketing efforts since we founded the company and sales efforts has not really been necessarily about telling people about our product. It's actually just been educating them about how the market actually works, what's happening behind the scenes, because most people's impression is that they're using a local intellectual property law firm with maybe 40, 50 people and they're doing all the work. But, as you heard earlier, those guys are outsourcing it and maybe even outsourcing it to someone who's outsourcing it to these local experts in the many hundreds of countries around the world who are then doing it manually. So that, and then also the financial aspect on top. I can only imagine that could be an even bigger problem in other industry than ours, and that's just such a big opportunity to build great and valuable software. So I think that joined kind of into the previous topic.

Speaker 2:

But the highlight I always mentioned to to new entrepreneurs which is always a little bit of a dilemma too is that I've always, uh, if I reflect back on the past year or two years in the company, uh, thought that I have been, uh, not critical enough when, when making new hires get on the app best possible people in my team is, uh, the most important thing I can do as a founder and ceo. Uh, the problem is that I become smarter and develop myself every single year too. So a person who I might have thought would have been, you know, the absolute best a year ago, uh, might not be. Uh, how do you say my absolute best a year ago might not be. How do you say my absolute best today?

Speaker 2:

And as we grow as a company and become larger and increase our customer volume and grow faster, we also start unlocking a type of hire that we couldn't have a year ago. Of course, you don't always have to finances to attract and hire the most talented people around the world. Of course, you don't always have to finances to attract and hire the most talented people around the world, but, iterating on that, I think that that has been my toughest learning and we fortunately have an amazing team in the company today, but we still have many, many future hires that could bring so much more to our company that I'm looking to attract and I feel like that's an area where I'm continuously developing myself and that, especially in my early years as a founder, highly underestimated.

Speaker 1:

Yeah, wonderful advice, and I think that if you're a legal tech startup that can find ourselves to our beloved vertical here and you're looking someday to raise money, if you're not interested in hiring the best, either out of the gate, if you can afford to be picky, or later on as your revenue is increasing, you could be a little bit more careful than you're hiring, because you have a bigger hiring spend in your budget. If you're not paying enough attention to it, not only will your business suffer, of course, but your investors or potential investors are going to be paying a lot of attention to the team you build, and you might as well get into the mindset early on about how important that hiring is. And while you may try to do everything at the outset as a sole founder or a very small team and you may have to do that as you grow you're going to have to hire. Of course, you're going to spend money to do that. You're going to attract investors because of how you've done that.

Speaker 1:

But to Matt's point, get in the mindset, make it something as important as even product development, because if you don't, I think there's a penalty to be paid, and one that may only be paid by your losing grant to your competitors, which is sort of a silent killer until it isn't silent anymore. Great advice. Thank you so much for highlighting that. We try to keep the pod down to half an hour 35 minutes. So before we wrap up, and thank you very much, it's been a wonderful discussion, illuminating in so many ways. Mads, how do people find Pat Renaud? How do they find you if they want to get in touch and learn more?

Speaker 2:

Well, the previous thing about your company name being your website is that it's very easy to find us once you know. That thing about your company name being your website is that it's very fine, as once you know that one. One philosophy I've actually had since very early on in the company is that we try to create as little sales material as possible but always publish everything on our website. So we have so much learning material on our website. We've created so many blog posts explaining a lot of the things I've covered here today. We even have published like guides on how to renew a patent yourself, in case you just have a few patents and it doesn't make sense using a law firm or even our company. So patentworldcom is the place to find us, and there's even a little section about me where you can also find our LinkedIn. And then my odd Danish name is also the only one with it in the world for myself, so finding me on LinkedIn should also be straightforward, as I'm the only one with the name.

Speaker 1:

Very good. Yeah, you know, I think maybe a compatriot of yours, mads Mikkelsen, is that the actor. Yeah, that's where I came across Mads, first, so it wasn't so strange to me. Thank you so much, mads, for, as I said, an illuminated conversation. As I say to everyone with whom I do a podcast and who is living right next door, I look forward to meeting you in person someday. Hopefully I'll get to Europe at some conference and we'll shake hands. But if, as a standard not-going invite that I make to anyone and everyone within my podcast, when you get to the Big Apple I'm only 30 minutes from that lovely city let me know and we'll get together in my neck of the woods.

Speaker 2:

Likewise Tony Great to catch up, as always.

Speaker 1:

Thank you for listening to the LegalTech Startup Focus podcast. If you're interested in LegalTech startups and enjoyed this podcast, please consider joining the free LegalTech Startup Focus community by going to wwwlegaltechstartupfocuscom and signing up. Again, thanks.