Chad Main: I’m Chad Main, and this is Technically Legal. A podcast about the intersection of technology and the practice of law. In this episode, I talk to Augie Rakow, and Max Cantor of Atrium LLP and Atrium LTS respectively.
We just heard from Augie. The Justin he’s referring to is Justin Kan, a silicon valley angel investor and serial entrepreneur. Augie founded Atrium with Justin, and Max was recently hired to head up Atrium’s artificial intelligence efforts. As you’ll hear, Atrium’s working on some pretty cool stuff and just might be on their way to building the law firm of the future.
Chad Main: Before we get to my talk with Augie and Max, some background on how Atrium is structured is important. Atrium is two business entities. Atrium LLP is a law firm, and Atrium LTS is a separate corporation.
Atrium is two companies for a couple reasons. One is for the sake of process efficiency, which is very important to the Atrium crew, but just as important to them and another reason that Atrium is two companies, is that because it is also designed to comply with the regulatory framework that governs law firms. Which, I might add, is not without detractors, and is a whole other topic for another podcast.
In short Atrium LLP provides legal services like any other law firm, but it utilizes the services of Atrium LTS for operational and technical support.
Before we get to the nitty gritty about what Atrium is up to, if you need any evidence that they’re coming at law from a different angle, you only need to talk to Augie and Max for about five minutes to figure that out. Before Augie helped launch Atrium, he was a patent lawyer and did legal work for startups for about 10 years. Most recently he was at Orrick Herrington, but most interestingly, before he began the practice of law, he went to divinity school.
Augie Goes to Divinity School
Augie Rakow: Yeah, so coming out of college, I ended up going to divinity school. Actually divinity school has a couple different tracks. There’s a professional track where you’re really going to become a preacher. That’s really what people think of with divinity school. You think of MLK having a Mdiv and stuff like that, that’s that track.
Also within divinity school there’s a religious studies cohort, I was in that group. I had always really been very interested in pretty rigorous, systematic, discourse about how we live and how society is set up and what’s the good life. Stuff like that.
That’s the connection between divinity and law firm for me. Those are really the two disciplines, where I think the core tools of the discipline consist of making sense of statements about how we should live.
Using fairly abstract concepts that capture human and social values about fairness, if they’re religious, things like mercy and love and forgiveness. If it’s legal stuff it’s things like is a patent obvious? Concepts of whether a patent is obvious or whether it’s a sin to do something, are these concepts that are totally made by humans, and both divinity and law, try to take these concepts and apply them systematically and rigorously, to solve day to day problems. If a couple splits up, who should get the house? To me those are really the commonality between them.
Chad Main: How do you make the jump from intellectual history to law?
Augie Rakow: To law, yeah. I was in a bookstore in Tokyo, I was just browsing through the bookstore, and I came across a law section, and I had had no interaction with law, I didn’t have any lawyers in my family, except a great-grandfather who I didn’t know.
We didn’t have it in undergraduate curriculums at the time, in the US. In Japan and most other countries, it’s a much more widely studied thing at the undergraduate level, and you have sections in bookstores about law and all that. I just picked up a book on constitutional law, written by a Japanese law professor, basically the Lawrence Tribe of Japan writing in Japanese about constitutional law issues in Japan for Japanese law students, and just totally fell in love with it.
Max Takes Flight
Chad Main: Not to be outdone by Augie’s cool story about going to divinity school, Max has a hobby, flies planes.
Max Cantor: So, personal hobbies, I’m also a private pilot. I got my pilot’s license in 2013. It’s always been one of my dreams. It’s tons of fun. That has probably been my greatest exposure to the law. When you start training, you get a three inch thick book on federal aviation regulations. When you say, “What do I have to read?” And the answer comes back, “All of it.”
Chad Main: What do you fly?
Max Cantor: Mostly single engine planes. Diamonds, Cessnas, Cirrus’.
Chad Main: I guess when Atrium blows up, they won’t have to look far to find a pilot to fly the company jet, but in the meantime, Max is responsible for Atrium LTS’ AI efforts.
Max’s background provides some good insight into how Atrium is approaching its legal services. Max is not a lawyer, and before Atrium, had never worked in the legal services industry. He has a computer science background and he himself, is a tech company founder. Coincidentally, he, like Augie, spent some time in Japan.
Max Cantor: Went to school originally for political science and international relations, but just became fascinated with computer science after taking a breadth requirement.
I graduated with a Bachelor’s and Master’s in Computer Science and was offered a job in Tokyo. That’s where Augie and I, like ships in the night, crossed paths, not knowing each other. I was working in Tokyo for few years, notably at Jane Street Capital, running a small trading desk.
Responsibilities as a trader but spending six hours trading, seven hours trading and then another seven or eight hours coding or writing up models and all that and I loved it. After that, I moved to Singapore, where I was running another slightly larger book on another trading desk, and then started feeling the bug of technology more.
This was around 2010, we just had the financial crisis, and some really interesting stuff was happening, so I tried my hand at a startup in Singapore, not entirely successfully. Learned a great object lesson in founder market fit, and that going off and building a consumer loyalty product for a quick service retail, when you have no experience in quick service retail, consumer loyalty or marketing, is probably not the optimal strategy for success.
I moved back to San Francisco and reconnected with a good friend of mine from college, and we launched a startup called DocMunch, whose product was focused on extracting tabular and textual data from PDF’s.
The eventual tie-in to Atrium, years later, was this focus on data extraction and processing semi-structured or unstructured data. DocMunch was acquired by a company called Nitro, that makes an Adobe Acrobat competitor. It’s PDF processing software. I was running a machine learning group at Nitro for about a year and a half before I left to pursue other options.
Actually, Mike my Co-Founder, connected me with Bebe, who was one of the company-founders of Atrium. At the time I wasn’t quite looking for a job. I had some projects going I was really excited about. What sold me, on Atrium, was the combination of what I saw … It was a pretty innovative model. They combine both, really solid legal chops. Orrick was actually our representation for DocMunch, we love them, they’re a great Silcon Valley law firm. Knowing Augie left that to come to Atrium, was a pretty significant signal to me that they were on to something real.
On the other side, they also had Justin Kan, who is pretty much about as far away from a straight laced lawyer, that you can get. He does define himself as a power user of legal services. Much like someone who flies a lot, you can probably spot a lot of things wrong with the Aviation industry, I think he had a lot of insights into law.
The combination of that founding team, their early success, and the opportunities that present themselves, from a machine learning, natural language processing perspective, would you have access to these clients, to these documents, to the legal markup on those documents, was fascinating to me. That was something I just could not say no to.
Augie Meets Justin Kan and the Seeds for Atrium are Planted
Chad Main: As mentioned at the beginning of the podcast, the Justin Max is referring to is Justin Kan, a Silicon Valley serial entrepreneur who’s been involved in several startups, including a live video feed website called Twitch and another called JustinTV, which was a 24/7 live video feed of his life.
Amazon ultimately acquired Twitch. Justin ultimately became a part of Y Combinator, a premiere Silicon valley incubator. Around 2017 Augie and Justin connected. It was at this time, that they started to sow the seeds to what would ultimately become Atrium.
Augie Rakow: I represented Justin’s brother’s company, in a pretty big transaction. I didn’t meet him at that point, we actually met on Facebook, asking his followers how much they pay for their financings and a woman I had hired at Atrium, I’m sorry at Orrick, to help me to stay connected to the community and people like Justin, spotted that and sent me the link.
I didn’t know Justin at the time so I wasn’t connected with him on Facebook. She sent me the link and I responded to his post, saying. “Here’s how you manage the cost and financings. Here are the main two techniques, but that’s just how you do it today. Justin if you’re interested in a more radically new way of doing things, I’d love to talk.”
We met up and talked for four hours, and met up a couple days later and talked for another four hours. Our first three meetings were four and half hours each. We just had so much to share on this. Over the course of the next few months we got to know each other better. It was one of these decisions that sort of happened. The momentum just kind of took over.
Chad Main: You and Justin keep talking and you just decide together, “Hey, I’m going to leave the practice of law, and we’re going to do this.” Is that what happened?
Augie Rakow: Yeah, there were two or three things that really … I came back from that meeting, the first weekend I met him, and I went back to the three biggest rainmakers in my group at Orrick, I said, “Hey I just met this guy, he’s a big deal. I think there’s a 50% chance he launches something that in the next two years will take 30% of our business. That’s my assessment.”
I got different responses. One said, “Oh yeah people have been trying to do this with law firms forever.” Kind of dismissive. “Investors aren’t interested in the kinds of law firms types of profits.” Another one said, “I’m jealous. That sounds cool.” Another one said, “Keep Justin close.”
Founder Segment – Everchron Founder Jeremiah Kelman
Chad Main: We’ll get back to my talk with the Atrium guys in just minute, but now it’s time in the podcast for the segment where I sit down with the Legal Tech founder and talk about their company. Today we’re talking to Jeremiah Kelman of Everchron. That’s a collaborative case management platform for litigators. Pretty cool product. Here’s what Jeremiah had to say about it.
Jeremiah Kelman: Everchron is essentially a collaborative litigation management analysis platform. A software that brings together all the key information on a legal case, and really meant to allow litigators and their team to collaborate effectively. Do analysis, and organize the most important documents in their case.
Chad Main: What’s the motivation? What’s the story behind Everchron? What inspired you to create the product?
Jeremiah Kelman: Sure. Both my company-founder and I were lawyers at firm, Irell & Manella in Los Angeles. We were doing a lot of large scale litigation and that sort of thing.
We were both there for about six or seven years before we left, and founded Everchron. What we were experiencing at the time was just a lot of frustration around working together as a team and being able to organize everything and understand the facts and documents in our cases. Really the way this is done, even today, often it’s just a little crazy and frustrating.
Using essentially Word documents and Excel to organize information. Then a whole host of different systems and tools where everything is all separated out. My company-founder and I, were basically just putting our heads together on what better ways could we bring together newer technologies and software to make the process of litigation a lot easier and more efficient and effective. That was the Genesis of Everchron really, to build a better collaborative way of working with litigation information.
Chad Main: Before the call today, I was checking out the website, it’s got some cool features. Tell us about a few of those.
Jeremiah Kelman: Really what Everchron does, is it gets you a centralized workspace to put together a lot of information today lives in a lit of different places.
For example, you might have all of your case filings and discovery responses and that sort of thing, organized on a file share, or in document management system, and on the other hand you might have a lot of discussion in your emails, you might have some stuff living in Word documents, you might have a whole host of documents in a large e-discovery review database.
Everchron brings in the best of these sources of information, puts it in a central space and then you can do things like, for example, very easily organize your information, your case documents, your key documents, into chronology of key facts and documents.
Our core focus is ease of use. We’ve realized that the idea is to get a whole host of different types of people in here. Could be a senior partner, could be a client, paralegal, so it’s very important to us to make it extremely easy and fast for this collaborating around this information. Very easy to search, very easy to create reports.
The other core functionality of Everchron is our master file, which is the filing system that would keep everything that would be exchanged between the parties, filed with the court. We have a unique filing system that shows relationships of documents to each other.
Finally, really in tune with our philosophy, which is, “A Little In, A Lot Out”, if you’re going to put work into the system, it better give you back a lot. We have automatically generated player profiles, or witness profiles. Just by working in the system, just by putting your documents in here, we automatically give you analysis, and pull together all that different information on a witness by witness basis.
Chad Main: Well cool. I appreciate your time today. It’s a cool product and where can people find you?
Jeremiah Kelman: Sure. You can find us online at Everchron.com or you can reach out to us by email at email@example.com and we’d be happy to help you get started in learning more about Everchron.
Atrium Attracts Investors Right Out of the Gate
Chad Main: Alright, lets get back to my talk with Augie Rakow and Max Cantor of Atrium. When we left off, Augie had just told us about meeting Justin Kan, and hatching the idea for Atrium. Now the next step was for them to raise money. When the Atrium team went out and raised money, they raised more than 10 million bucks, which is a nice chunk of change for any business, but it’s an especially nice sum for a legal tech business that includes a service based offering as part of its business plan.
As I can tell you from experience, venture capitalists do not get that excited about service based businesses. If you don’t believe me, Google, “What should you do with your crappy little service business”, and check out the first article that comes up.
I asked Augie what was it that they were able to do to get technology investors excited about Atrium?
Augie Rakow: I think the thing that Justin did, at that phase, the kind of thing that he is uniquely talented to do … There’s a lot of stuff where he’s uniquely positioned. He’s well known, he has access to capital-you know access to all that kind of stuff.
What he did that was so unique, I think, was to articulate in extraordinary simple terms, two or three principles that have really broad ranging effects. The two or three principles he came up are really powerful. They’re kinds of things, the more you think about them, the more things start to make sense, and it really pointed a way forward.
I think other people probably would have not hit on those principles; that would be me, or would have made them … I don’t mean to throw myself under the bus here, or would have made them too complex, which also would have been me. They were extraordinarily deep, and they were extraordinarily simple. The value of them was just undeniable.
Chad Main: What are these principles?
Augie Rakow: The most fundamental one, was the insight that explains why so many legal tech startups fail.
It really is an explanation of why law firms don’t buy legal tech. The thing about the billable hour is part of it, I think, but I think the deeper reason is that there is no vehicle to capture …
In a law firm, there’s much less economic incentive to invest in the operations and it’s in addition to the billable hour. There’s an additional factor, and that is there’s no market for your law firm equity.
If you think of McDonald’s investing in developing a new hamburger, that might get more people in the door and you have greater revenue, but it also drives up the value of the McDonald’s stock.
That second part is completely missing in law firms because there’s no market for equity. It leads to all kind of incrementalism and conservatism. It also leads to short-term planning.
Another way to put it is that, I think there are very few people who are in a position to materially influence a law firm’s investment decisions, who are you enough and early enough in their practice to benefit from those investments 10 years out. For contrast, every time I see …
Think of Bill Gates, it’s been years since he was actively involved in Microsoft. He still personally benefits I’d imagine, still personally benefits, every time Microsoft decides to invest in its longer term growth. He personally benefits, right? It drives up the value of the stock, he can take financing against it. He’s beyond needing that, but to this day, he still personally benefits from long-term investments and his kids will continue to benefit, or his foundations will continue to benefit from it. Totally absent in a law firm.
Atrium is Two Companies
Chad Main: Another thing Atrium pointed out to investors when they went out to raise money, was that it’s sometimes hard for law firms to hire high-level non-lawyer talent to support the business. No small part of this difficulty is regulatory.
Specifically, regulations that prevent non-lawyers from having an ownership interest in law firms. By creating LTS as a separate non law firm entity, Atrium is better suited to go out and hire talented developers. Marketers. Business development people, that might otherwise go to bigger name tech companies, rather than a law firm.
Augie Rakow: One of these other principles that I mentioned, that were Justin’s insight into how to fix what’s broken here, is that traditional law firms don’t have very differentiated staffing.
It’s really only been in the last 1- to 20 years that they even started to have marketing teams, and things like that. Our experience is that that’s the specialized professional talent, non-legal professional talent, that law firms attract … Law firms don’t even compete with the Google’s and the Facebook’s and places like that, for that sort of talent.
It’s for obvious reasons that there’s a glass ceiling in most law firms, because of the regulatory rules. They can’t become owners of their law firm. They can take administrative or executive roles within the law firm, but they can’t become owners. This creates a sort of second class citizenship within the law firms. They tend not to be … There’s not the equity upside potential in a law firm. They tend not to attract a lot of the talent.
Chad Main: As noted earlier, Atrium LTS, the business side, provides operational support to Atrium LLP, the law firm side. However, outside of and beyond operational support, another reason Atrium LTS was founded, was to develop technology that the law firm could use, that his clients could use, and ultimately that other law firms could use.
Max Cantor: Within the technology services, which I’ll assume includes the operation support that Augie mentioned, we have the technology group.
In technology, we have client facing products, which our clients can log on like a portal. They can see certain things. Internal facing products. Those would be tool that for example, our paralegals will use to load up a set of documents from a new client from prior counsel and markup, “This is a counter-party for this contract. This is the effective date for this contract, etc.”
We store that all on our custom document management system. Then we also have another layer behind all that, called the data platform. That’s really where things get interesting. With the data platform, what we’re doing is keeping track of every step in a workflow that our paralegals and attorneys undertake.
When our paralegal needs to markup a document and say this is the counter-party or this is the effective date, those actions are stored. Those actions are stored in a way that our algorithms can be trained against.
What we’re doing, while we’re building these front-end platforms for our paralegals, our attorneys, and our clients, on the backend, we’re training machine learning algorithms to make predictions based on what they’ve seen so far.
In some cases, our paralegals don’t need to manually type in counter-parties anymore. The system will automatically suggest counter-parties. They will suggest effective dates. That’s the approach that I’m trying to take. We’re not trying to replace anyone in the immediate future. We’re just trying to eliminate the manual repetitive tasks that dominate much of legal workflows.
Business People are Key to Atrium
Chad Main: Interestingly, a side effect of having an affiliation with a technology company that was founded by business people, run by business people, and staffed by business people, is that lawyers at Atrium LLP get to learn business concepts that aren’t always normally taught at conventional law firms. Concepts like how to manage subordinates.
Augie Rakow: One of the things I spend the most time thinking about, that I learned from LTS, or get from the LTS side, is really management coaching.Management advice.
I’ve learned a lot from Justin and Bebe in particular, about how to manage a report. It’s new thing for me. I’ve had teams. I’ve led many teams. I’ve sent out invoices to clients where there’s more than 50 … 100 timekeepers listed on the invoice. I’ve been responsible for the work of many people, but I have never been a manger to someone. In a sense of setting their goals, reviewing their goals, being responsible for their career progression, making sure that they … Never done that before.
Chad Main: Since opening it’s doors in June of 2017, Atrium’s already landed more than 100 clients and they employ a business development team.
Augie Rakow: We just celebrated getting our 100th client. Been in business since June 1. It’s eight months or so. 100 clients or so, actually about 120 to 125 now.
Gotten them through various different ways. A pretty large influx of companies have just reached out to the website when they seen the various press releases.
We also have a fantastic growth team here. We have three or four, five members on the growth team at the moment. Any how you divide it up, they’re all folks that have built businesses or have professional sales careers where they’ve done extraordinarily well, consistent with the ethical rules. The lawyers supervise them. There’s a lot of rules around that, that are very important to follow, but we have a professional sales team, and I think that’s really the difference.
We actually think of the services that we’re providing as products, and so we don’t just do … No law firm would say that they’ll just do anything that comes in the door, but even work that I would have done at my prior firm, we may pass on it. Not because it’s not good work, not good work to do, or not good clients, but we may pass on it because we’re just not providing that service quite yet.
By constraining ourselves, we can streamline our operations and we can be more strategic about how we do try to sell the services that we do sell. It’s that combination of more specifically defining the product. We call it productizing our services and then working with an extraordinarily talented sales team and marketing team to tell the community about those services and guide the sales funnel.
Atrium is Focused on the Startup Community
Chad Main: Like any good Silicon Valley startup, Atrium is laser focused on product market fit. This means, in the near-term, they’re limiting the legal services they offer just to those needed by startup companies. As they refine their processes and grow, they plan to organically move into other areas like M&A and litigation.
However, as Augie points out, that doesn’t mean that Atrium’s structure and MO can’t be applied to non-business related areas of the law.
Augie Rakow: The work we’re doing now, tracks the kind of work that startups get from the leading law firms in this space right now. The Wilson Sosinis, Cooley’s, The Gunderson’s, or a handful of others. Goodwin, Fenwick.
It’s basically that same model. That’s the same menu. There’s already product market fit there, we know that clients consume those services, we know that they’ll pay well for it, and we know there’s talent out there to provide those services. Let’s not screw with something that works.
Over time, we expect to expand. We’ll most likely follow similar growth patterns to the other law firms that have done well in this space. You represent a lot of venture act startups, pretty soon you start selling them to Google. Start selling them Salesforce. Soon it becomes an M&A practice, then you may turn that M&A practice into a buy side M&A practice and gradually expand out from there. We will mostly be a business law firm. I think these principles apply to everybody in our industry.
Chad Main: Not surprisingly, Atrium’s pricing structure’s a little bit different than a conventional law firm. Instead of maximizing billable hours, much of the way Atrium builds clients is subscription based.
Augie Rakow: This market segment of clients really want to use legal, and they want it to be fast, transparent, and price predictable. The fast speaks for itself.
Transparent means they want to know the status of things. They want it to be price predictable. Not necessarily cheap. A lot of times, as long as it’s predictable, they’re willing to pay well for it. Especially if they feel like they’re getting something really good for it.
The subscription really speaks to that price predictability. There’s going to be some variants. A client may need a lawyer’s time for four hours one month, then two hours another month, and ten hours the next month. If you are striving to be unit profitable at the task by task level, then you really do need to track hour by hour, or six minutes by six minutes.
If you only need to be profitable at a client by client level, or at a quarter by quarter level, that gives you a lot more freedom to absorb the variants. We really put that on ourselves to accurately price in aggregate, much like the insurance company might. What’s going to be the likely cost of this? That’s how we absorb the variants on our side.
The clients we’re taking on now … The subscription price varies a little bit. We’re rolling out new pricing in the next couple weeks. Mostly, clients are paying 11500 to 3000 a month.
The Future of Atrium’s Tech
Chad Main: So I closed out this great conversation with Augie and Max with a question to Max about what the future holds to the tech development at Atrium LTS.
Max Cantor: In the short-term future, our goal is to just capture more of the workflows that the LLP attorneys and paralegals are performing. The idea being, that they have to spend less time in Word, less time in Outlook, more time on our systems. That gives them higher value and it’s more holistic.
We can also report back, or give them the ability to generate better metrics. Augie can turn around and say, “This is where our bottlenecks are.” When we’re on a subscription business, in a way we’re arbitrizing that difference and it’s absolutely vital for us to know what we’re spending time on, so that we can then fix that. On the data platform side, our goal is just improving predictions, eliminating more and more tasks.
One of the distinctions that I like to think about, for machine learning, is what I call auto-complete versus Siri. Auto-complete type machine learning tools are tools that have incremental value. If I’m using Google auto-complete, and it’s wrong 75% of the time, I just keep typing. I still get 25% of the value out of that.
If I’m using Siri, and it’s wrong 75% of the time, I get no value out of it, because now I just wasted my time talking to my phone, like someone who might have some issues. Now I gotta go do whatever I was going to do myself. My interest here, has always been to build the auto-complete. To build the auto-complete tools for attorneys.
Not zero touch automation that has to be perfect, but something that helps our attorneys and paralegals 75% of the time, and that is useful 75% of the time. If our engineering team can make tech that makes the attorneys as happy as … Having Augie cover their backs makes them happy, that’s my goal.
Our goal, in a sense, is attorney retention. Make it so much more pleasant to do legal work here, because we have amazing legal tools, that if you go work at another law firm, you lose. If we have that product loyalty to our internal product then that’s a huge win for us.
Chad Main: That’s a wrap for this episode. If you want to learn more about Augie and Max, or find links and more information about anything we talked about, you can check out their page on our website, TLPodcast.com.
If you want to subscribe, you can find us on most major podcasting platforms such as iTunes, Stitcher, and GooglePlay.
If you want to get a hold of me, my email address is firstname.lastname@example.org. That’s email@example.com.
Hope you’ll tune in next time, where I talk to Angie Hickey of Chicago based Levenfeld Pearlstein. She’s the CEO of that firm and she talks about what it’s really like to run a law firm like a business. Until next time, this has been Technically Legal.