Legal Tech StartUp Focus Podcast

How Legal Tech Startup, Abstract, Turns Laws And Regulations Into Actionable Strategy

• Charles Uniman

Share episode

Charlie Uniman, your podcast host, and Pat Utz, CEO and co-founder of Abstract (https://www.abstract.us/), explore how Abstract turns legislative and regulatory noise into real-time, client-specific strategy, and why deterministic workflows beat agentic dreams for now. Pat shares candid lessons on funding, traction, hiring, and building a product that actually moves the needle.

• origins in AI research and mission to make government transparent
• proactive monitoring of bills, regulations, executive orders, and local agendas
• real-time enrichment with client context to flag risks and opportunities
• strategy layer and workflows that move from alert to outreach
• planned expansion into judicial opinions to shape case law awareness
• enterprise focus with long-term goal of a consumer version
• founder advice on VC vs bootstrapping, traction before funding, and hiring fit

If you're interested in legal tech startups and enjoyed this podcast, please consider joining the free Legal Tech Startup Focus community by going to www.legaltech startupfocus.com and signing up


SPEAKER_01:

Hi everyone, this is Charlie Uniman, your podcast host for the Legal Tech Startup Focus Podcast. I'm delighted to have as a guest on the pod today the CEO and co-founder, I think I have that right, uh, for a company called Abstract. We're going to learn all about what Abstract does and what its uh secret sauce is. That person is Pat Utz. Uh, Pat, welcome.

SPEAKER_00:

Hey, Charlie. Great to be on.

SPEAKER_01:

Great to have you. So um, full disclosure, Pat and I actually met uh at uh a conference in New York City a short while ago. So I had the pleasure of meeting him face to face and getting my own download about abstract. But um of course, what we want to do here is give Pat ample opportunity to tell listeners uh what it is that Abstract is all about and uh what its uh differentiating features are. So uh Pat, before you do that, give us a little bit of a background, as you did for me uh when we met about how you got into the space where you are with Abstract.

SPEAKER_00:

Yeah, no, absolutely, Charlie. So um the long story made short is that uh this story uh behind abstract started in research, actually, out of Loyola Marymount University. So we're sort of the typical co-founders that came from a technical background, but we're really sort of obsessed around, you know, trying to make the process of government legislation, regulation, and all these dense court documents more transparent. So uh this started off as an AI research project in 2018. Long story made short, after about two, three years of that, um, we realized that we could sell this as a service. Um, and so that essentially led to about uh three years working with lobbying firms. So we grew that business to about 200 firms as customers. And over the last two years, we raised um, you know, additional venture capital to go and now sell to large law firms and in-house legal um teams that need to understand you know what's going on with regulations and legislation. But really, the the genesis of the story was you know, just these passionate engineering kids that, you know, were a little naive, of course, uh and wanted to make government more transparent. But the mission is still the same today. It's make government transparent and accessible and use AI to distill all the noise that's going on across all these um levels of governing bodies.

SPEAKER_01:

Well, that's a pretty good uh place to get going. I I can't help but think in the world of AI, where um you do have some uh uh AI magic to uh to cast as a spell as part of your platform, and I've seen it work and it's really good. Um to have lobbyists, law firms, legal departments, uh all of which are uh desperate to figure out what the hell is going on with legislation, regulation, uh court cases that pertain to SAME. Um that's that's that's a good niche, and it's a good set of uh of uh of uh potential and actual customers. Um but tell us, Pat, um there are other players out there. Um I've uh even podcasted with someone who is adjacent to you. Um what is it about abstract and your research? Uh you did point out you started out as a researcher in this area, that uh helps uh abstract do the best it can and differentiate it from what others do in this uh legislation, regulation, court case space?

SPEAKER_00:

Absolutely. So I always like to break it down uh sort of civics 101, right? Which is you have the three branches of government, the judicial branch, the executive branch, and the legislative branch. And that sort of concept exists at all three levels within the US, meaning local governments, so think cities, counties, special districts, um also at the state level, and then at the federal level. Um so if you look at our initial focus, we started off with uh mainly the executive and legislative branches of government. So within those legislative and um executive branches, you have the state legislatures, you have Congress, um, you also, you know, we argue that city meeting agendas are kind of like the legislatures at the city level, right? Um and then the executive side is mostly uh the regulatory agencies, right, is how we think about it. And these regulatory agencies also mainly exist at the federal and state level with some different varying flavors at the local. Um, so what we're doing with abstract, um, and what was really hard, you know, at least the the difficult part from an engineering perspective, is take all these updates on a daily basis because there's no there's no like API, there's no like center or rather uh you know core source of truth where you can get all these uh bills and regulations on an ongoing basis. You actually have to build your own scrapers and create these pipelines, right? It's not unified. Um, and there are a couple of different um open source projects out there, but again, there's no you know one sort of source that covers everything. And for reference, I always tell people, you know, there's about 140,000 plus governments in the US. If you think about all the cities, the counties, the regulatory agencies, et cetera. So again, our focus with abstract initially uh was looking at these executive and legislative branches. And so the type of data that we are abstracting, no pun intended, are these proposed bills and proposed regulations. So we very much have a focus on the proactive side of the equation, really helping clients figure out what's likely going to change the law or the regulatory statute, and then provide strategy on what to do about it. Now, the third side of that equation is what's going to come soon, which is uh the judicial side. And when we think about being proactive with the judicial side of things, we typically look at court opinions because these opinions, as they come out, are what's going to ultimately shape case law. So the court opinion piece is something that we're actually going to expand to soon. Um anyone who wants to talk to me about it, always happy to chat. But that is going to be a new development for us, and I'm really excited about it.

SPEAKER_01:

Well, you you had mentioned something, and uh, if I recall correctly from our uh face-to-face uh get together, uh you're not only making available the uh let's let's focus now on the legislative and and uh executive branches of these levels of government, not only making available the the uh the text, um you're doing so in in pretty much real time. Is that right, Pat?

SPEAKER_00:

Yeah, real time. And what's really key to what we do with Abstract um is we have context on each business, or in the case of a law firm, we have context on their clients. So what abstract is doing, it is it is real time scanning any changes across these uh government bodies. And then with context on clients, for example, it's able to determine hey, is this going to be a risk or an opportunity to the client? Here's what you should do, right, about it. You should maybe upsell some services, maybe you should, it's just like a hey, a warning of what's gonna come. Um, but that's exactly right. It's it's real time scraping these updates as they happen. And then we're using AI to enrich that with context. And so Abstract is acting like a team of analysts, right? It's like a team of people that are constantly on the lookout, figuring out what across these governing bodies are going to impact the clients or the business itself.

SPEAKER_01:

And and uh you had mentioned uh uh a short while ago during the pod now uh that there's a strategic analysis uh aspect to what uh abstract can offer. So it's not just the bare bones text and no more, given the the uh the context that you can obtain in dealing with law firms, legal departments, lobbyists too, I presume. Uh you you can also offer uh more than just uh uh words on paper mirroring the text. You can provide some sort of uh analysis.

SPEAKER_00:

Correct. So you know, as a concrete example, let's say a law firm's using abstract. Abstract detects uh a new executive order um from the presidential office, uh executive office, right, that says, hey, we're gonna implement a new tariff. So abstract is not only flagging that, we call that discovery, right? So it's discovering what's relevant to your client, it's then taking that tariff, reasoning through how that tariff may impact, let's say, your automotive um manufacturer client, and then providing strategy on what to do next, right? So it's taking a tariff saying, okay, if this tariff were to be enacted, this is how it's going to impact the supply chain. These are some talking points you should go and use to reach out to your automotive client. Um and then spoiler alert, uh very similar to Lagora and Harvey, uh, which are more focused on the case law and internal workflows. Um, we're building workflows on top of this, right?

SPEAKER_01:

Ah.

SPEAKER_00:

Right? So we're building workflows to then draft emails, communication points, etc. Um, and that's you know part of what really excites me about where we're going with the product.

SPEAKER_01:

Very good. You know, it's it's funny. Uh on LinkedIn earlier today, I saw a legal tech commentator uh post a poll um asking what the um the buzzword for legal tech for 2025 is going to be. Uh one was workflow, so you're right on top of that.

SPEAKER_00:

Yeah.

SPEAKER_01:

Uh the other one was, I think, uh boom or bust. I guess those are two words, funding, and I think the fourth was agentic. I I voted for um I voted for workflow. Uh so you and I are on the same page there. Because I think lawyers as witness the success that uh Lagore and Harrafi are having want to be able to uh you know adopt uh a solution that uh lays out for them the steps that uh that they have to take to uh to try to fit it into uh you know this particular subject matter that you're looking with fit it into their day-to-day. And if you can provide some workflow advice, that's that's going to be key. And and I was thinking about the executive orders, and you beat me to the question. You do cover that sort of stuff, I presume, at the federal level and and maybe even at some of the local levels as well.

SPEAKER_00:

Yes, that's correct. Um, by the way, fully agree with the workflow piece. Um agents will take over the world. Not yet.

SPEAKER_01:

Yeah, more work has to be done. Yeah.

SPEAKER_00:

Yes, we're we're uncovering like the rate of innovation there is astounding, but it's becoming very clear it's um it's gonna take a while. And by a while, I think at least another five years before we see more commercialized versions of these agentic workflows um that we can just trust to do things agentically um like a human would. And again, I always tell people like, you know, what's the difference between agent and sort of these workflows we talk about? The difference is like with the workflows, we have a very clear set of prompts and directions that we are fine-tuning for the client. And so it's really good for repeatable results. Yeah, an agent by definition is hey, go figure it out yourself, right? And and you can go take different turns on your own. We don't have to teach you about it. So we're still a while away where we can trust those agents uh to produce reliable outputs.

SPEAKER_01:

No, I quite agree with you, although I'm prepared to be surprised because uh and I don't disagree with the five-year horizon. Uh but I always, but I often discover that when I think it takes five years, lo and behold, some genius in a garage, or just as likely perhaps some genius working at a uh unicorn firm uh shows me uh that they they beat me by two or three years. But you know, uh workflows are deterministic and uh and lawyers I think are going to want to hang on to their hats when it comes to letting uh uh non-deterministic uh, at least for now, agents loose uh both internally and and and when it comes to rendering advice to to clients. So I uh again, I think we're we're on the uh same page. But I want to circle back to uh remarkable as it sounds, do I understand that you really are covering most, if not all, of those 140 different thousand um units of government for want of a better word?

SPEAKER_00:

Yeah.

SPEAKER_01:

Wow.

SPEAKER_00:

Yeah, yeah. So it's it's been um I mean it's been a core unlock, but I so I have been living here in New York for a bit now, um, New York City. And I think it's personally it's it's been really exciting to see abstract identify local agendas um that I you know sort of pertain to the issues I care about. Um with context about you know what I care about because you know, being the person behind the product, I I get to I have the privilege of I guess having my own workspace without having to pay for it. Um but you know, it's it really it's really incredible seeing the the platform sort of go out intelligently, pick up these agendas on housing issues, um sort of food issues, et cetera, and then just stealing what the city council is talking about, you know, on a daily basis without me having to go figure out what the heck is going on. And um some really key part here is that we leverage um web search capability as well. And so what that means is that if abstract picks up uh, let's say a city agenda about uh housing, affordable housing that the um council members are talking about. Abstract will also go out and do research across late all the latest news articles, et cetera, that are going on to further enrich that report. So it's able to go retrieve further context with deep research. And uh it's it's incredible. It's it's truly great.

SPEAKER_01:

Yeah, it is incredible. Um, are you uh and you may have told me this when we met face to face, I just don't recall you uh pitching uh mostly to the larger so-called big law law firms now. Um and and do you are are you thinking about if if that's the case, are you thinking about going a little further? Uh I won't say down market, I don't mean to be pejorative, but to small to medium-sized firms.

SPEAKER_00:

Oh, completely. Um, this is just part of our go-to-market tactic, um, where we want to work with the bigger fish um initially and then use that to then tackle the rest of the market. I've been very vocal about this, where I want us to create a consumer version of this. I think the everyday person deserves access to this information, and it's why we started the company, right? Uh to begin with. So um long story made short, absolutely, that's the plan. Uh, but right now, you know, we're very much focused on servicing these larger sort of AMLA 200 firms and then uh in-house legal and government affairs teams across uh Fortune 500 companies.

SPEAKER_01:

Well, I I I commend you for that. Um uh looking ahead, I uh when I first got out of law school, um actually no, I guess I was a um uh first and ended my first year of law school, and I got hired by a local law firm in New Jersey where I lived. Uh of course, I won't say it's before the invention of the automobile, but certainly before the invention of the internet. And um I actually remember working on some uh local law stuff and uh being the uh uh junior guy in the totem pole there as a summer associate. You know, I had the unenviable task of trying to pull together all this stuff that was hardly available in print, let alone online. There was no online. Uh so I I uh remembering that, I I can imagine that small to medium-sized law firms, solo practitioners, are going to be wetting their licking their lips and wetting their chops to get their hands on this when you can make it available uh uh uh at a mutually beneficial pricing arrangement, because this is gold to to local law firms who just just and and of course the consumer aspect to it as well.

SPEAKER_00:

Right. No, I agree. Um and again, if some you know, sort of out outside of this like initial target that we have comes to us and says, hey, we're interested, we we we still have the conversation, right? Um uh we still take that in because it's really core to like how we develop our product roadmap. Um, but I cannot I can't agree more. Very good, very good.

SPEAKER_01:

Well, how long have you been uh uh doing this with abstract now?

SPEAKER_00:

Yeah, so it's been um technically five years. Um I guess our current product, um, we launched it this year. Um, so that's been under development for about a year and a half, two years. Um, but we've been operating for five years with the first half focused on that lobbying uh firm business.

SPEAKER_01:

Very good. So that means you've had uh some of the water behind your ears uh dry dry away or not wet behind the ears anymore. Oh which which leads me, and you know, and I we did have a chance to talk about your journey a little bit. Uh there are ups and downs. What and and this is the part of the podcast that I really enjoy uh talking about with founders, and that is asking them for business advice, good things that they have done that they're proud about, proud of, uh stumbles along the way that had they known what they know now, they might not have stumbled or stumbled as badly. What comes to mind as sort of the business advice, especially selling to the you know, the legal uh uh delivery uh provider market, what are some things you want to tell other founders that might help them if they're earlier in the journey or along with you where you are, um that uh that you'd like to impart to these founders?

SPEAKER_00:

Yeah, I mean, quite literally an infinite list here, but um some uh some some things that come to mind, right? So, first of all, if you're just entering this sort of idea or phase of like starting a company, um a tech company, um I would always ask the question of do you want it to be venture backed or bootstrapped? Um you know, you hear this advice a lot, but really What it means is, hey, am I building a business where I really think we're able to generate hundreds of millions of revenue? And so I'm going to go take venture capital to dilute myself to be able to grow two hundreds of millions of revenue. So then I can go sell for a larger sum of money. Or do I think this is a business where, hey, I can build something with a much smaller team, maybe just myself, right? Especially now with all the AI tooling out there. We have a lot more solo entrepreneurs, solo founding teams, um, that build product that delivers value to clients. They do not need any sort of outside funding. So they own 100% of the business. They're able to generate, you know, half a million bucks to let's say uh, you know, five, maybe even$10 million of revenue revenue, and then sell the business and keep all the proceeds for themselves, right? So that's a bootstrapped sort of route that um I always tell people like seriously consider. Now it's really impossible to uh predict that at first. And it's also like the tough thing is that it for most humans, right? If they're given the option of like, hey, here's a million bucks from venture capital today, go hire and build people, that is always a more uh what's the right word? It feels more comfortable, right? Uh it feels like you're de-risking it. Um, but just keep in mind that bootstrapping is such a you know uh valid way of doing things that um I would almost like suggest considering that heavier, especially now in the day of or in the in the age of AI tooling, right, where you can build a lot more. Um, and so that's one thing. The other quick thing is that um even if you're going to raise venture capital, ever since the uh Silicon Valley Bank uh explosion in 2023 happened, uh venture capital has been a lot more strict about putting in money until there's traction. So a lot of people think that they can, if they just have an idea, they can go raise money and um and go at it. But really, uh you you need to you need to build yourself or you need to find somebody, a co-founder, that can build product right with you. Um the name of the game is build something yourself, get some initial traction, whether it's paid users, users using the product um and giving you feedback. Uh, just any sort of traction before you go out to raise is absolutely necessary. Um, the only way that you raise money uh without traction in a product is if you're a second-time founder. But even then, second-time founders know better and they're gonna go build a prototype, they're gonna build something, they're gonna get some traction, and they're then they're gonna raise money. So I just reiterate that because I come across so many people that think they can just raise venture capital with an idea um or with loosely signed contracts. Um, building is really, really key to that.

SPEAKER_01:

Yeah, I uh and and it it it perhaps doesn't require saying, but uh being known for stating the obvious, I'll I'll uh live up to my reputation for that. You not only uh uh uh you know uh dilute yourself uh financially in terms of share ownership and the like, but you you you sure as hell dilute your uh your ability to control your destiny as well. Uh you're now answerable to the money people who can add value and are not necessarily bad folks, uh but you're you're on a rocket ship then, and you might not realize it, but you're not entirely in control of the rocket ship. And uh you're gonna have a lot of decisions that you once may have been able to make with on your own or with your co-founder, uh affecting your lifestyle, your work-life balance, uh your direction of the company. Um and and you know, just for completeness sake, to as a corollary to what uh Pat was saying, you know, uh uh you wanna, even if you are aiming to get on the rocket ship and aim for the tens, perhaps hundreds of millions of dollars that uh you know venture institutional venture capital insists upon as as a prerequisite to funding, um, you may want to delay that for as long as you can, because the longer you delay, the better terms you're gonna get. Um I guess most of our listeners know that, but for those who may uh need a reminder, um uh I'll offer that. Um and you've raised venture money, I I remember you're saying. Is that right?

SPEAKER_00:

Yeah, yeah. No, we've raised um, I mean, it's it's publicly out there. We've raised about nine million dollars over the course of um uh the five last five years in different sort of tranches and pivots. Um I've also experienced what it's like to restart a company and a capitalization table. Um so I always offer people um, you know, uh a second year if they're going through that phase of like, hey, we might have to do an extension round or we're running out of money. So we we went through that and um came out real pretty strong, quite strong actually, on the other end. Um, but definitely a more rare case, as I've I've come to realize. Um, most people end up just having to shut down their company. Right. Um, another random tidbit uh here is that a lot of people um actually end up in debt uh if they don't shut down soon enough. So people think that like, you know, I have all this runway after I raise money. Okay, we're just gonna operate until we run out of money. The reality is you still have expenses that you have to pay after you run out of money. So, you know, hopefully no one's in this position right now. But if you are, it's okay. It's normal, it's part of the process. Just remember you're gonna have to shut down probably two to three months earlier than you think if that's actually gonna happen. Um, so another piece of advice I'd offer to founders that are maybe experiencing that. Um, but yeah, I mean, so so many other I I've I've been blessed to have so many different uh mentors, coaches, um, tons of you know resources offered to me also in terms of reading, right? That it's building a company is an amalgamation of so many different um sort of areas of expertise, right? There's finances, there's people, there's product, engineering, customer success. And ultimately, like the job of the founder is to be a sponge, get good enough at each of these and understand them well enough until you then realize it's time to hire, right? Um, and when you do hire, um that's a whole nother kind of worms because the second you hire somebody, you are now saying, This is this person's part of my company, right? It's an extension of my company. And so this is where the idea of culture becomes really important. And, you know, I've been at fault for this. Uh, I know a lot of other founders have been at fault for this, where they sort of they found somebody that was like smart enough, like endearing enough, and they hired them, but the reality was they were not the right fit and they ended up burning a ton of money, or you know, sometimes it's the cause of the failure of the company, right? So hiring is so critical, and um, that's a whole nother podcast.

SPEAKER_01:

That's a whole nother podcast, exactly right. And and a shameless plug for the legal tech startup focused community. Uh I posted today a uh link to uh an article by a former VC who was now a coach for VCs. And the the subject of the article was uh, hey, you your your startup has gone bust and you've shut it down. This uh uh former VC by the name of Charlie O'Donnell, located in Brooklyn, uh near where uh Pat and I are located, you said it's important to do a post-mortem with your investors, however awkward that might be. Uh, you know, a formal postmortem, because you never know when someone might be reaching out to those investors to get a beat on you for another venture that you have. And it's always good to not leave loose ends dangling and things unaddressed. Yep. So read the article, subscribe uh uh uh or become a member. It's free to the Legal Tech Startup Focus community at legal techstartupfocus.com. See what I posted on LinkedIn about this. Um one of the best pieces of advice I've come across recently uh uh from the standpoint of startup management and and startup life. Um Pat, I can't thank you enough. Uh I'm always very jealous of my listeners' time, so we do try to keep it to about a half an hour. So this has been uh power-packed. Uh I I think uh remembering my days as a summer associate at a very small law firm in small town in New Jersey, I think you're doing God's work in getting these um uh uh in getting this info and analysis out to people who have to keep oppressed of what's going on in government. And you have me um you have me fascinated, but I'll I'll I'll let you take your time developing it and announcing it when you can go public with it uh about what's coming with court cases and the like. That sounds exciting. If uh if people want to reach out to you um in abstract, uh how do they how do they find you in the company?

SPEAKER_00:

Yeah, yeah, I mean, three main ways is um our website, it's abstract.us. Um you can also find me um on LinkedIn, it's just pat oots, uh P-A-T space ut. And uh send me a message there. I would say if you're listening to this podcast, uh you know, I I'm more than comfortable sharing my email as well. So my email is pat at symbolabstract.us. So pat at abstract.us.

SPEAKER_01:

Very good. Well, reach out uh to Pat. He's uh he's a great guy to talk to. I enjoyed meeting him face to face. I hope you enjoyed our discussion. And uh, you know, I'd say uh uh when you're in New York, if you were outside of New York, we'd have to get together. But you're in New York, so uh when I get into the city, exactly uh we'll we'll meet up. You know, we didn't have a chance for a drink last time. We'll make sure, we'll make sure that we do that.

SPEAKER_00:

Oh yeah, and I always I always tell people, look, they're all meeting in midtown, etc. I there's some great stuff going on in Dumbo. So I I always selfishly tell people, hey, you should check out Dumbo if you haven't been there. Uh there's some nice views. So I'm I'm always I'm always a big proponent of that.

SPEAKER_01:

Yeah, and for people living on the other side of the world, Dumbo is a spot near the Brooklyn Bridge in Brooklyn. Brooklyn is um uh the homeland for many people, including my uh one of my kids. Great place to live. And uh yeah, I'm glad you put a plug in for Brooklyn and Dumbo. Nice.

SPEAKER_00:

I had to. I had to.

SPEAKER_01:

Very, very nice. Pat, it's a pleasure. We will see one another in uh in Dumbo. And uh uh I can't wait to get this one published. It was uh it was great to have you on board.

SPEAKER_00:

Thanks, Charlie. Thanks for me.

SPEAKER_01:

Thank you. Thank you for listening to the Legal Tech Startup Focus Podcast. If you're interested in legal tech startups and enjoyed this podcast, please consider joining the free Legal Tech Startup Focus community by going to www.legaltech startupfocus.com and signing up. Again, thanks.