Hello, everyone and welcome to this Silicon Valley Innovation Center interview. I am your host, Rahim Rahemtulla. And let me just apologize for the slightly late running of the program today, so I apologize for any inconvenience that may have caused. But I’m very glad that you’re with us today because we’re talking about intellectual property and we’re doing that with Alec Sorensen, who is the Founder and CEO of Tradespace. And Tradespace, briefly, is a platform for intellectual property analytics, but I’m sure Alec will tell us more on that and many other issues. Alec, thank you so much for joining us today.
And if I could, I think just to start out, I want to ask you a little about the context of Tradespace and the context in which it came into being. My understanding is that, from what you’ve said in the past, a company’s value can, and can to a large extent, be determined by what IP it holds – in fact, up to 60% of a company’s value could be locked up in the IP – but the problem is that these companies don’t always necessarily know the value of the intellectual property that they have, that they’ve been generating through their years of research and development. And so, as I understand it, Tradespace is kind of an answer to that problem and it allows companies to see, potentially, what the value is of their intellectual property. Alec Sorensen: Yeah, exactly. So first of all, thank you for having me, it’s really a pleasure to be on and to get to talk about a topic that, as you mentioned, is increasingly important to really any company in the economy, regardless of what market they’re in. So, to answer your question, I think it’s important to first kind of understand where we’ve been and how we’ve gotten to this point. So if you look back 10, 20 years at what makes a company valuable, the typical answer is their plan, their factory, their equipment, maybe their sales channels, their people – all of these physical assets. And that’s how companies thought about what it meant to be a strong company. And fastforward to the last 10, 20 years, where we’ve entered this knowledge economy in which the lifeblood of companies are the technologies they build, assets that aren’t necessarily tangible and, as you said, now account for up to 60% of the value of a company and more so than any other one-asset class.And because this transition is happening so quickly, companies haven’t caught up. They don’t have any systems in the way that they have, let’s say, CRM systems to manage their sales or HR systems that manage their people. There’s nothing to help them understand intellectual property. And so what we do at Tradespace, as the largest global IP marketplace, is first to help companies understand what they have. It sounds like a simple proposition, but large companies go through a number of acquisitions over their history, they develop new technology, and simply keeping track of what you have is a really difficult activity. And so we make that incredibly easy and we use some some pretty interesting analytics to help companies really understand not just the assets they have, but what that means, what market areas they have exposure to and how strong these technologies are. And then – and I know we’ll talk about this in a little bit – we help them actually prioritize those technologies. So how do you look at 5000 different technologies that a company like General Electric has developed over 20 years and then make sense of it and figure out, “Well, what do we really want to think about? What’s important to us as a company? What technologies can we group together to build something that’s going to really have value?”
Rahim Rahemtulla: Yeah, I mean, 5000 is a really big number. And what I think is so remarkable is that these companies have all this intellectual property they themselves do not know what it is, they don’t necessarily have a good record of this.
Alec Sorensen: Yeah.
Rahim Rahemtulla: And so they can log into your platform, look at themselves, essentially, and the records that you’ll have that are actually better than what they have in-house.
Alec Sorensen: Right, right. Exactly. It really, again, speaks to this transition where you have an asset that’s become incredibly valuable and that the companies just don’t really have exposure to – it’s a major blind spot. And not just their own intellectual property, but what smart foreleading companies are realizing is that understanding what their competitors are doing, the technology investments that they’ve made as well, is now almost table stakes to being a competitive company in this economy.
Rahim Rahemtulla: Is this having some IP?
Alec Sorensen: Having IP and knowing what your competitors have, and knowing how what you have compares to that. Companies simply can’t afford to not be strategic about this asset.
Rahim Rahemtulla: Yeah, interesting. And the other thing that I know you’ve talked about is that, when you look at the IP activity, patent activity, I suppose, of a company, there’s actually more than one way to read that and actually you can look at this as being a sign of the health of the company. And I suppose this is kind of what you were talking about, that, in today’s economy, the healthiest, strongest companies are going to be one with a lot of patents, with a lot of activity in that sphere. Is that is that right?
Alec Sorensen: Right, exactly. So we actually recently took a look at a number of companies in a lot of different markets, and what we did was we looked at Forbes Most Innovative Companies and compared those to the Fortune 500. And what we found is that those companies on Forbes Most Innovative lists, not only do they have higher valuations and higher multiples, but they were almost twice as IP intensive, so they had twice as many patents for every dollar than any other company in the economy. And that really just hammers home the point that not only are they developing intellectual property, they’re being strategic in how they manage that and thinking about, “What areas do we develop IP and technology?” And then, “What do we do with that technology when we have it? We don’t let it just sit on a shelf but we actually develop an approach to get that technology in the market in the most effective way possible.”
Rahim Rahemtulla: Indeed. And then Tradespace can help them do this. And so maybe let’s talk a little bit about how it does that. As I understand, where Tradespace really adds a lot of value is that it rates patents relative to other patents in a given space. And so I think this is where it starts to get really interesting because that judgment that the platform is making is a qualitative one and so it’s potentially, you know, contentious, people might want to know how you make that judgment, where it comes from. And so perhaps you can tell us a little bit more about how this rating system works?
Alec Sorensen: Sure. So, again, if you look at our mission and what the market needs, it’s to find a better way for companies to share technology with each other to get more value out of that. And so we address two problems. And so you talked about the first, which is scoring. How do you actually pick which assets are valuable? And how do you understand how valuable they are? And then the second is doing that connectivity, so bringing companies together and figuring out which companies are after that technology. So I can talk a little bit about our approach to both.
So I came out of this space after seven years in consulting, working with companies that will go into a back room and kind of come out with a valuation judgment around a piece of IP. They will say, “This patent is worth a million dollars.” And there’s really not a whole lot of data behind that. And so, when we approach this at Tradespace, our goal is not to provide an absolute judgment about IP. One of the coolest things about intellectual property, frankly, is that it could be worth very different amounts to different companies, depending on how they would use something, and there could be 5, 10, 20 uses for a given piece of intellectual property. So what we endeavor to do, instead, is to show users how valuable and how strong a piece of technology is relative to every other patent for that technology. And we’re able to do that by looking at a number of data points in the patent themselves.
So we look at the network of citations. So what that means is that, every time somebody files a patent, they have to reference all the work that came before them, that built up to this patent. And so we’re actually able to understand where this particular piece of technology fits in his field of innovation.
Is this something that a lot of other technologies are referencing? Those technologies that they are referencing, are they strong, are the foundational? So we developed a really sophisticated sense for, again, how this compares to its peers. We look at other data points, things like, “How old is the patent? Is this still innovative technology? Does this give you broad coverage?”
We do a lot of other analyses that I won’t go into here, but I think the point is – and I think this will be a theme throughout – we do use some some advanced methodologies, we use things like machine learning and AI, but the foundation of all that is really an understanding of the market and the industry. So we spent months and months training our algorithms based on human inputs. So we worked with experts in a number of different technology fields to say, “Look, this is how we ranked these pieces of IP in a field. How would you do that? What would you change? And what are your considerations around that?” And by doing that, we’ve been able to develop a set of algorithms and a logic that ends up being incredibly comprehensive across a number of different market segments. And it’s always learning, too. As we go out and continue to serve customers, as new transactions occur on the platform, as we get new data, we’re continually learning and continually able to improve the way we help position pieces of intellectual property and contextualize them.
Rahim Rahemtulla: And so, if I understood correctly, you have this algorithm and you use machine learning to come up with a rating and you’re comparing that with human experts, with IP experts in the field, and they’re looking at that rating that Tradespace has come up with and they’re saying “Yes, I agree”, maybe, “No, I don’t.” And, if they don’t, then you go back to your system and tweak it until you find why it is that you’re not coming up with the human rating?
Alec Sorensen: Right, exactly. And, technically, we’ve created a very robust proprietary set of training data that these algorithms can use, and what that comes down to is one very unique way to understand the strength of a patent and then it’s marketability. So we’re able to understand market trends as well, to see, Is this IP in a market that’s expanding? Is it growing quickly? Is it shrinking? So we can tell companies, “Look, you have this technology, it’s strong and it’s growing. You need to prioritize this.” Or maybe you’re struggling, but the market is weakening as we speak and so it’s not necessarily worth the effort. So to very easily and quickly give somebody, an executive in a commercialization role, a simple intuitive answer to, “How do I get more value out of my IP?”
Rahim Rahemtulla: Sure, sure. And the information you have about the market, if it’s growing or shrinking, where does that come from?
Alec Sorensen: So it’s a combination of data. We look at new patenting behavior, and so this is something that I think we’ll probably talk about. The money that companies spend on developing technology and, ultimately, on patenting it ends up being highly correlated with market performance about 18 months down the road. So the technologies that a company are developing today are the same technologies they’ll be getting revenue off of in 18 months, 24 months. So we look at that, we look at financial data, we look at acquisitions, new company formation, a whole set of data that gives us a comprehensive look at what this market is doing from a number of different perspectives.
Rahim Rahemtulla: Sure. And I wonder if we could just pause on this idea of artificial intelligence and machine learning for a moment because it’s a very hot topic today. And I just want say, Do you think Tradespace is possible without machine learning? Or is it because we have these sophisticated systems now that you’re able to make it work? Because I mean, there’s so much data that you’re pulling down. You’re looking into patent offices all around the world, if I’m not mistaken.
Alec Sorensen: Absolutely. So there’s no doubt that what we’re doing today wouldn’t be possible without the infrastructure that exists. I heard a really interesting quote that, to start a company in the 90s – a software company – cost about $20 million with the pressure to develop the technology. And today, you can do that for about $50,000. So just take a moment to think about how far we’ve come. And so the ability for somebody to pull in that much data, to store it, to analyze it – you simply couldn’t, it would be cost-prohibitive to do that beforehand. But an important point as well is that you have to think about machine learning as being enabled by the application. And so when we look at the market – and we actually work with a lot of companies that are looking to commercialize machine learning and intellectual property – what we see is that machine learning is becoming increasingly commoditized, which is good for the market as a whole.
But without a strong understanding of the problem that you’re solving, of the real value you can add to a company, then machine learning isn’t necessarily valuable in and of itself. To give you example, we work with a lot of companies that have used other IP data solutions that will leverage intellectual property and leverage big data and analytics, but they’re not able to answer those core questions of, How do I actually commercialize the technology? How do I find new technology to bring in? And so, in order for these advanced techniques to be effective, they need to be paired with a deep understanding of the customer problem and an ability to execute on that as well.
Rahim Rahemtulla: It’s really interesting, the way your business is growing because of ML and how other businesses who want to commercialize ML are looking to you to help them do that.
Alec Sorensen: Right.
Rahim Rahemtulla: And so, from all sides, it’s making things happen. And so I just wonder, Alec, what do you think – if I can sort of digress slightly – what are your thoughts, having worked with ML a lot and seeing as well how it’s developing in the market, what are your thoughts about it generally in the economy, so to speak, and compared to using human experts who you also draw upon? Do you see a bright future for machine learning and artificial intelligence in our workplaces? Do you think it’s going to become more and more prevalent?
Alec Sorensen: I do. So I tend to think that it’s perhaps a little bit more nascent than we tend to believe. The point at which AI can completely replace human workers to solve core challenges, I think, is fairly far off, five to seven years. I think what the near-term in the mid-term future look like are people being made more effective by the use of these tools. So, to give you an example, when we come in to a university technology transfer office, for example, or to a technology scouting organization, our pitch is never that we’re going to eliminate that function. In fact, we don’t want to. We believe that there’s parts of that role in terms of structuring deals and working with the inventors and reaching out to potential investors that, at least for now, inherently needs to be done by somebody. And so what we want to do is we want to free up more of their time to do those critical value-added roles by eliminating some of the more kind of automatable commodity tasks.
And we also want to make ourselves more effective as we provide services to companies, as we help connect companies with pieces of intellectual property. And we’re able to be much more effective and one of our analysts can help five to ten companies because we’re using these powerful tools. So when I look at machine learning and AI, I think they’re incredibly powerful. But going back to my previous statement, you need to have that problem and you need to have an understanding of that problem. And for now, at least, I think that’s going to be achieved by working directly with people to make their jobs more effective.
Rahim Rahemtulla: Thank you, Alec. It’s interesting, do you ever find that the Tradespace algorithm comes up with ratings for intellectual property which you yourself could not have predicted or may not necessarily understand, as it were, at first glance? I mean, like a piece of technology which you thought maybe wasn’t that strong could come up with a high rating or vice versa. Does that ever happen?
Alec Sorensen: Absolutely. And I think, again, this is one of the key value adds of machine learning and AI: that when applied correctly they do a great job of removing human bias. And so, where I see this most at Tradespace is how we connect technology with companies. And, like I said before, any technology could have 10 or 20 different applications. And so often I’ll approach something, a problem, with my own expectations about what a technology could be used for it.
To give you an example, we worked with a company that provided data compression and they allowed you to edit that data in place and so we were responsible for helping them exit that market. And we had our own assumptions about which industries this would be good for going into this; we reached out to folks in Telecom, we thought about maybe some connected car applications. And it turns out where we saw a lot of strength was biometrics. Japanese biometric companies were investing significant amounts of money in collecting large pieces of data and they needed to be able to change that data after they exported and transmitted it. And so you have an example of a market that was prime for this technology that I would have never gotten to, even knowing how that market works.
Rahim Rahemtulla: Sure.
Alec Sorensen: But, by looking at the underlying technology, we got there quite quickly and we were able to do that. And we provided other markets as well that wouldn’t have necessarily been so obvious. So I think that’s where the lack of bias becomes critical: yes, in scoring and trying to cut through some of the hype around buzzwords, but also in making connections where the human brain might not necessarily go.
Rahim Rahemtulla: For sure. Of course, that stands to have huge value for companies because no one knows everything that’s going on in Japan or in South America or who knows in what markets or what companies are doing. So to be able to have all that data and actually draw insights out of it, of course, is going to be a huge bonus.
And so I wanted to ask you, just to move on, about once a company has their IP on the database, has the ratings for it and they think, “Okay, yes, let’s commercialize this, let’s get some revenue out of this after we’ve invested so much over the years.” And so you’ve talked a little bit about how you would help them find other markets, companies, where there might be applicable. And I’m wondering, what’s the next step in that process? I suppose, the transfer, if that’s the right way to put it, of getting that technology from your company to the other company. And I wonder how that sort of happens. Are there models for how you should license technology or how you should price it? That seems like a whole complex set of questions.
Alec Sorensen: It is, it is. And you’re absolutely right in that one of the major reasons why deals fail, once you’ve brought parties to the table, is in the inability to agree on a way to take that technology to market. And so one thing we do and where we add considerable value is to actually use data from the technology to help inform which approach to use to commercialize. So what we’ve seen is that there’s a number of different ways to take a technology to market. You could spin it out into a startup. You could work with another company to code-develop it, to mature it a little bit more. You can simply license it out to a company that is buying up as much IP as they can to avoid legal risk. Or you could license out to a company that was looking to grow that technology and actually invest there. And each one of those models has different implications for pricing, for access, for terms.
And what we see is that, if you look at the market for technology, that tells you a lot about how you want to commercialize it. If something is in a very fast growing market and maybe the technology is a little less mature, then spinning out a startup often ends up being the best way to de-risk that and the best way to pair the risk with the opportunity, whereas some companies may not want to license the technology that’s still in its infancy, even if the market is growing. However, if the market is a little bit flatter and the technology is still new but interesting, then working with an entrenched competitor to co-develop that could be the right option and you end up getting the best terms by structuring a deal like that.
If you flip that on its head and you look at more developed technologies, maybe in mature markets or growing markets, then licensing becomes more effective and more viable and by bringing technologies that are more mature – and our data is able to bear that out – we find that that licensing deals happen more quickly, the General Counsel’s Office raises fewer red flags. And so just taking a data-driven approach to how you choose to commercialize a piece of intellectual property ends up being a simple but really critical step. And so we’ll bring in our own support, people who have negotiated hundreds of these deals, to dot the I’s and cross the T’s. But, ultimately, just coming in at the right order of magnitude and the right general consensus on how to do something makes all the difference.
Rahim Rahemtulla: Fantastic. Well, Alec, thank you very much. It’s been wonderful to talk to you. And I think it sounds like it’s really interesting and, as you say, like an increasingly strategic area that companies need to think about, how they deal with their IP, how they manage it and so on. And just, I think, as we draw to a close here, perhaps if you’d like to share any final thoughts for executives, companies out there who are thinking about their IP. Maybe you’ve hinted at it already by taking a data-driven approach. Is that ultimately where you would say that the priorities should lie?
Alec Sorensen: I think even before taking a data-driven approach is the recognition across the enterprise that, regardless of how much IP you have as an organization, your ability to navigate this landscape and think critically about your IP and that of your peers will be the make-or-break factor in a company’s success. And their ability to navigate this going forward and effectively generate technology and bring technology into the enterprise will dictate their success in the knowledge-based economy.
Rahim Rahemtulla: Thank you. Well, that’s brilliant, Alec. I think it’s very sage advice for all the executives and the companies out there. And so, I think on that note we’ll wrap things up today. So Alec, just thanks very much for being with us today and sharing your insights. Alec Sorensen: Sure, it’s been my pleasure. I’m glad we were able to talk.
Rahim Rahemtulla: And we hope that you, our audience out there, have enjoyed it as well. We would just like to say that, if you want to find out more about everything we’ve been talking about today with Alec as well as the more general theme of innovation and digital transformation, which is so prevalent today, then I do encourage you to check out our website, siliconvalley.center and have a look at our Executive Immersion programs. They can really bring you right up to speed on what’s happening today in digital transformation and the intellectual property revolution which is going on. And I would also just add that, if you want to see the recording of my talk with Alec today, as well as our past expert talks, it’s also available at our website, at siliconvalley.center. And so, on that note, I think that’s where we’ll wrap things up for today. So thank you very much for joining us, once again. I’ve been Rahim Rahemtulla. My guest today is Alec Sorensen of Tradespace. And we’ll see you again very soon, I hope. Bye bye for now.