Inside Canada’s Fastest Growing AI Company | Spellbook, Scott Stevenson

14:09Scott Stevenson:Yeah, yeah, yeah, yeah. I think like if you're outside it, it can, you might wonder like, yeah, is the hype real? Like it is probably one of the hottest verticals besides like AI for coding. It's, you know, maybe the hottest and most talked about vertical for AI right now. And I think there's like a reason why it has taken off so quickly. You know, one analogy I use is that large language models launching in like 2022 or with GPT-2 were kind of like the spreadsheet moment for lawyers. Like accountants back in the 80s, when spreadsheets were introduced, they started to be able to, you know, automate a lot of the work of like running a financial model.

14:56Scott Stevenson:Before spreadsheets, you know, it used to take like a basically a building full of people to run a complex financial model. After spreadsheets and databases, we were able using computers to be able to automate a lot of the basic, you know, rudimentary math and formulas of kind of financial models. And, And finance, maybe back in the 70s and 80s was run by an army of people, humans. And now today, it's maybe actually 95 % automated. If you think about the volume of transactions, how much bookkeeping is semi-automated, software like Stripe and all of the tools we have to automate finance, we've automated a lot over many decades.

15:43Scott Stevenson:that has not happened in law or has not really started to happen until large language models in 2022. So that's like most verticals have seen decades of software adoption and automation, whereas law basically up until 2022 still just ran 100 % on an army of humans. Like the biggest advancements we had were like the word processor and email, and that made things go a little bit faster, but still the core problem was software could not deal with unstructured text. So it couldn't read unstructured text and it couldn't write unstructured text. We had no, and that's all lawyers do is they deal with, you know, 60 page documents of unstructured text and we just had no way to ingest it to understand it.

16:27Turner Novak:So you would just kind of have to just read it and or you had 10 years of experience and just know these are the things I care about and I kind of know where to find them. And it might only take me 20 minutes or something or five minutes, but it's still. Yeah.

16:40Scott Stevenson:But you still have to read like every word. Like if you're reviewing a contract for a client, 60 page contract, like you have to read every word, basically, you know, there could be something in there that you're missing. So, you know, I think there's just been radical inefficiency in, in, in the practice of law. And that's, and now it's like the dam is breaking. There's all this like pent up demand for legal efficiency. just like in every other vertical that has been able to adopt software. And now it's finally able to be met because large language models are finally allowing us to actually help with the actual work that lawyers do.

17:15Turner Novak:So what was like the software stack of a lawyer, I don't know, five years ago, like pre-LLMs?

17:23Scott Stevenson:Word, Outlook. Do they have like a phone, like maybe Zoom or something? Zoom, maybe. I'd say five years ago, they're still doing a lot of calls by phone and like the phone bridges.

17:36Turner Novak:So basically, they were doing all these, they were producing documents basically and reviewing written, they're reviewing and writing written documents with basically word and email and then talking on the phone to communicate of what would be changed in the document essentially. Yeah.

17:54Scott Stevenson:I mean, that was essentially the lawyer stack.

17:56Turner Novak:Okay.

17:56Scott Stevenson:It's some like deeply specialized software for things like entity management. Like if you have a complex entity structure with, you know, parent and children orgs, you might have like a chart of that. Yeah, like a design to like a design the org structure or something. Yeah. But, you know, beyond that, especially for like commercial lawyers, which is where we're focused, you know, there really hasn't e-signing or DocuSign. I forgot about that. DocuSign. Oh, that's fair. Okay.

18:21Turner Novak:And I think, I don't know if you mentioned it. You might have talked about it earlier when we got lunch, but there's about$30 trillion in contracts that are signed per year.

18:31Scott Stevenson:Yeah, about$30 trillion moves through contracts every year.

18:35Turner Novak:Okay, so this is like economic value that is under a contractual agreement of some kind?

18:39Scott Stevenson:Yeah, that's right. Yeah, so it's there's just this massive money flow moving through these contracts. And you know what inspired us to start the company is like, if you think about how inefficiently it's happening, like, you know, these contracts are probably taking 10 times longer than they should to be drafted and reviewed. and then things are still being missed because, you know, I mean, just imagine just being a human, reviewing a hundred page contract and it's like 8 p.m., you have a deadline. It's an almost hilariously impossible task to actually review a hundred pages with a fine tooth comb on like a tight timeline.

19:21Scott Stevenson:So did they not do it or did they do like,

19:24Turner Novak:well, how did they do this back, you know, before AI existed?

19:28Scott Stevenson:I mean, they would try their best. Yeah, yeah. Lawyers would try their best and contract managers would try their best. I mean, one approach is standardization. Like before AI came around, I think the hope was things would standardize more and more. So you could kind of see, okay, here's the standard SaaS agreement. Like YC has like SaaS agreement that's pretty common that, you know, most startups use. And you can kind of do a diff between like, oh, what is different about this agreement compared to the YC standard agreement. And so that was one method, like shortcut you could use if there was a standard template.

20:00Scott Stevenson:And like with venture capital financing transactions, you know, there's a standard set of templates. You can easily see what's changed. So that was one approach that would make these reviews easier, especially with complex transactions. But I think most of law has been like surprisingly resistant to standardization because the deals people want to make are all, you know, kind of unique and bespoke. Yeah, yeah. That actually connects to like where we started with Spellbook. So Spellbook was the second product we launched in, or one of the kind of the last product we launched in 2022. But we initially thought we were going to drive standardization with templates.

20:37And that was kind of where we had started.

20:42Turner Novak:Yeah, I definitely wanted to ask you about that. But I think maybe what we're still talking about, just general kind of like AI, non-Spellbook specific stuff. like, I don't know, when I think of a couple months ago, made a contract, I used to go to like ChatGPT and just say like, make me a contract, like make no mistakes, et cetera. Like, couldn't lawyers kind of do that? Like, is there like some sort of thing where sort of the generic AI products trip up on legal stuff?

21:08Scott Stevenson:Yeah, good question. So, I mean, the first thing I would say is lawyers don't actually draft anything from scratch for the most part, like especially contracts, because they want to start with a trusted precedent that they understand inside out because if they go to ChatGPT and ChatGPT outputs a whole contract, then they have to review every single word of that and make sure they understand it completely. That's very difficult. So ChatGPT is not great at modifying existing work or building on kind of your existing library. So we have a few features in Spellbook. Like one, yeah, you can start with the precedents that you're familiar with and we'll kind of modify those.

21:46Scott Stevenson:You can start with a sales agreement and be like, you know, make this GDPR compliant and it will kind of surgically make those edits for you. We also have a feature called library where you can kind of have your whole history of, you know, all the deals you've ever worked on and, you know, use that to kind of influence the output of the Spelbook as well. So these are, you know, some of the, one of the things I would say is like working off of your existing corpus of docs as a lawyer is really, really important and ChatGPT doesn't do that super well. But two, I think like, you know, lawyers want things built into their existing workflow.

22:22Scott Stevenson:I think like the chat interface is great, but it's still like the terminal UI of AI. Like, I don't think chat is the be all and end all. and we just have a lot of unique user experiences that would just never fit inside the shape of ChatGPT. For instance, one thing you can do in Spellbook is compare it to the market. So if you're, say, signing a commercial lease in Manhattan, you can say, compare this to the average commercial lease in Manhattan and tell me what's not normal. And then you can actually dig into the data, into the charts and actually look at the data that we've collected in real time from millions of contracts and explore that through this visual interface that has nothing to do with chat.

23:13Scott Stevenson:It's very, very distanced from that. So I think there's a huge number of experiences that people want that don't fit in a chat box.

23:21Turner Novak:Yeah, and so this is all within the Word interface?

23:25Scott Stevenson:A lot of this is in the Word interface. Some of it is in our Spellbook Associate product as well. Yeah.

23:30Turner Novak:Okay. And the there's like market data. It's basically you take everything that's run on Spellbook of every customer and it's like anonymized and you can see what like dates of comps or something like that, or it's like some kind of like database of data where you can compare it to the contract. Yeah. So it's like an opt in model.

23:51Scott Stevenson:And most of our customers have opted in. And the way the way it works is, yeah, we take anonymous aggregate statistics. So we only capture things like, what's the average price per square foot in a commercial lease in Manhattan? What's the average late payment interest rate for SaaS agreements? So the only thing we end up capturing is these very high level statistical pieces of data, and that's what gets exposed. So it allows it to be really privacy friendly. And yet it's an alternative to the approach of like fine tuning, like this idea of fine tuning was really hyped for a while. And you know, every, I think every founder wanted to like sell VCs on fine tuning because it sounds very complex and defensible and like, you're going to have this great moat, but it actually works pretty terribly for like a whole bunch of reasons.

24:47Scott Stevenson:And I can talk about that if you want.

24:49Turner Novak:Yeah, I think it'd be interesting just because it's because that was kind of like the meme or like the meta was, if you are an AI company, you must build your own model because there's no defensibility and you, you know, you probably need to buy a bunch of GPUs and you need to train them all. And, you know, it's like, if you're just a chat GPT rapper, like it was like, it was like this derogatory, like slur basically to like call someone a chat GPT rapper. Yeah.

25:14Scott Stevenson:So I think, I think that like was very like wrong. Like I think, I think this is an idea where like there, there are narratives that founders learn investors are hungry for and then they pitched them because it's very legible and easy to understand for an investor. Like, you know, this idea of training models, like, oh, it's going to be like OpenAI. You know, OpenAI trained their model and it was expensive and cash was like a moat, basically became a moat. But that did not really pan out in really many other areas, like for a bunch of reasons. Like you saw Bloomberg made Bloomberg GPT. That was one of the early ones.

25:49Scott Stevenson:They spent, I think, millions training it. And then GPT-4 came out and just completely beat it at finance tasks. So it was a waste. Similar things have happened in legal AI where a number of companies have tried to train their own legal specific models and fine tune them. I do not know of a single one that is still like in use today than any of the major application providers. So it ended up being this like kind of big waste of time. I think it's much, the much better approach is to build value around the models. And I think there's a lot of really great ways to do that. I think like RAG is actually really, really good and actually superior to fine tuning.

26:29Scott Stevenson:Are you familiar with RAG, like retrieval augmented generation?

26:32Turner Novak:It's basically when you take the model plus just like the internet or like external sources, essentially.

26:37Scott Stevenson:Yeah, exactly. So it's like the way I think of it is like, you know, fine tuning and training is kind of like injecting things into like the long term memory of a model or, you know, almost putting into like the evolutionary fiber of the model, like giving evolutionary instincts as well. But if you're asking a model to cite case law for like a litigation case, you don't want it looking at its long-term memory or its evolutionary instincts. You want it to actually look up the information and make it hard citation that you can actually cite. And it's a much actually less hallucination prone method of getting legal specific data to work in these systems.

27:21Scott Stevenson:So like one, you know, relying on RAG, which we did from the early days, like, yeah, you actually get citations that you can trust and inspect. Whereas if you're fine tuning models, you're still going to hallucinate and you have no way to inspect the data. Two, when you use RAG, you can filter. So like we can filter, you know, if you're a lawyer in London, UK, and yeah, you work for a healthcare company, you know, you can actually filter down the data to say, you know, I want to compare my contract to only other healthcare related contracts in the UK. So you can filter the sources down. Whereas when you train a model, you end up with this kind of one size fits all model.

28:03Scott Stevenson:And like a lot of lawyers will complain, well, oh, you know, ChatGPT is too biased towards the US or it's too biased towards like public company contracts. Because those are the only ones that are available to train on. which gets to my third point is like, no one wants to ingest like private legal data into these proprietary models, because there's always a chance it could be it could be spit out again. So by using our approach with like the statistics, people are actually comfortable putting private, allowing us to ingest private data, because it's fully anonymized, whereas people are not comfortable with training models on their their private legal data, because there's the chance that it could be spit back out again.

28:49Scott Stevenson:And that's not something they can accept. For those three reasons, like I think, you know, rag, and like what we're doing with this kind of real time data with market comparison and spellbook is like, yeah, just like much, much more superior, or very superior to to fine tuning for the most part. I think it was like this case where like, the herd just ran in like completely the wrong direction because it and I tweet about this all the time like you know things are hyper legible like the story just sounded right like you know data is the new oil and fine-tuning this models cash as a moat you know it just it's it sounded right but the reality I think is just much more nuanced and and complex and and you know now we see a lot of these companies from like 2022 2023 who went did that fine-tuning approach like a lot them are shutting down and they're probably every two to three weeks, I'll hear from one of these companies who's now they're now looking to get acquired because the markets matured so fast.

29:48They spent a lot of time kind of doing this deeper R &D that they, you know, actually wasn't

29:54Scott Stevenson:that effective. And that's like very core to like our culture at Spellbook is like, I wrote a blog post about this back in 2022 when we launched and it's called like, is GPT-3 too easy? So we, you know, we use the foundation models and back when we launched, people would say, well, is this too easy? Like, where's your team of like machine learning engineers? Shouldn't you be training your own models? And I cited this, this book, have you ever read this or seen it? It's called Playing to Win. It's by David Serlin. He's a professional street fighter player. Have you, have you seen this before?

30:25Turner Novak:I've read your post, but I've not, I've not seen the book.

30:27Scott Stevenson:Okay. It's, it's an amazing book where, you know, this guy professionally played Street Fighter at the highest level. And he talks about, you know, what is different about the mindset of a professional player versus what he calls a scrub or kind of an intermediate or a bad player of the game. And, you know, he said like the scrub basically loses the game before it even starts because they have this totally wrong mindset. And I'm paraphrasing, but like they basically have this romantic vision of the game that like, if they do, you know, play the game this like super proper way, this romantic version of the game that they'll come out on top in the long run.

31:06Scott Stevenson:Whereas the pros basically relentlessly exploit whatever they can to win, even if it looks cheap, even if it's easy. They don't do things because they look hard. Like if they can, you know, if you've ever played Street Fighter, have you ever like called someone cheap that you're playing against in one of these games?

31:25Turner Novak:I haven't really played Street Fighter like competitively. But it reminds me if you ever played like Halo Halo two, you can do this thing called double shotting where you could basically take a shot, but you would shoot two bullets. Okay, I never learned that trick. So all the all the best pro players in Halo two is like, you could literally do twice the damage with one shot. So everyone got good at double shotting. And if if you were like a purist, and you're like, I'm not doing that, like you just you can't beat people who do it.

31:52Scott Stevenson:Exactly, exactly. Yeah. And so I think like, there's this thing in AI, where it's like, you know, people almost, I think engineers in particular who love complexity almost can't accept, you know, how simple these systems can be to add value to customers. And, and there's this attraction to complexity, fine tuning R and D. I think it started, it's starting to die out now, finally. And people are realizing, okay, building on top of foundation models is probably the best approach a lot of the time. But yeah, but yeah, I think, I think like the herd ran, a lot of the heard ran in like the wrong direction.

32:25And it's, it's pretty fascinating.

32:28Turner Novak:Do you think part of it is this dynamic of like, what if open AI builds this or what if Anthropic builds this? Because if you're not building your own model that has any kind of differentiation, like they could just like, you know, tweak ChasGPT to like work better for lawyers or something like that. Like, is that, is that a part of this? Like, and how do you then navigate that as a founder of like building a product that's not in the strike zone? Sure.

32:53Scott Stevenson:Yeah, yeah. I mean, I do think that that is the reason why this kind of narrative took off. But I think like the pendulum swung too far in the direction of differentiation rather than customer value. So you had so many companies and investors focused so much on how do we differentiate and doing really complex and hard things that weren't very useful, you know. And seriously, like so many of these companies are like shutting down and selling off now. But that's the reason it happened. But how do you pragmatically deal with it? Yeah, I mean, that threat is there. I think the vertical AI providers have to work very hard every day to continue to add unique value to these customers.

33:38Scott Stevenson:And I tell our team, we have to be two years ahead at all times in terms of delivering state-of-the-art experiences to lawyers. Two years ahead. Like, we should be shipping things today that, you know, other competitors or other companies will be shipping in two years. So, like, I think you have to have this ruthlessly fast culture, continuously adding unique value. I think the way you add the value is, you know, one through the data. So, like, we have real-time data of, you know, from millions of contracts that we can use to deliver better results to our customers. We also have preference data.

34:17Scott Stevenson:So we learn from each of our customers what they care about. And, you know, ChatGPT and Claude are not really doing that for contracts specifically. You know, the data is super important, but then it's, and the features are important, but then it's like, it's like, how do you fit into the workflow of your customers? You know, lawyers are so busy. They have so much going on. You know, if you don't fit super neatly into the workflow, they're just not going to use it. And the reality is like, you know, Claude's not in Word. It's not, you know, it's not designed out of the gate to give a lawyer value.

34:53Scott Stevenson:And there's a million little friction points because of that. And I think if you, we always say, you know, our goal is to build a toaster, a toaster product. Like we are really good at doing one thing, you know, toasting contracts, I guess. Yeah.

35:08Turner Novak:With a$30 trillion size, market size or whatever. Exactly. Yeah.

35:14Scott Stevenson:If we can just do that one thing well, and if you optimize your product for that purpose, you just make so, so, so many decisions differently. That would never make sense for ChatGPT to make. There's so many little nuanced decisions that make that toasting experience very easy, simple, and effective for our customers.

35:38Turner Novak:And then there's, because I feel like they're sort of like the seven powers of like just how a business has competitive advantage. And I feel like we've maybe almost forgot about them sort of, but when you were describing sort of some of the different features, when I think of like network effects as a pretty powerful business, and it's just the more customers you have, the more market data you have, the more useful that feature becomes. There's maybe a point where like you have so much data that like one additional point doesn't matter. But like to get to that point, like there are, there is some strength to that.

36:10Turner Novak:like some, some positioning strength.

36:12Scott Stevenson:And then, I mean, it scales more than you would think, because it's not, it's like, Oh, what does it matter whether you have 1 million data points or 2 million data points? Well, it's, do you have data in Manhattan? You know, do you have data in London? Do you have data in SF? Do you have data in the healthcare industry? Do you have data in the aviation industry? Do you have data in manufacturing? Do you have data in energy? When you think of it that way, you know, these are all industries that you have to kind of conquer to deliver the best product to all of the lawyers in those industries. So, you know, it might sound like, well, what's the difference?

36:48Scott Stevenson:Yeah, again, what's the difference between having a million data points and two million data points? Well, it's less about that. It's like, how many industries are you in? How much geography are you in? And do you have a statistically significant sample where you can provide useful insights? And yeah, I think it is legitimately a really, you know, great kind of data network effect that we have.

37:08Turner Novak:yeah i feel and i feel like that's almost like we almost like forgot some of these rules for a while yeah and just like you know i i feel like almost like economies of scale sort of like took over in the sense of like you know the ability to train the models and like having the capital on the balance sheet that you could like utilize which like all this stuff is important but like other stuff still matters too i guess yes yeah i i think so so one thing you mentioned that we kind of like jumped past it, but I want to talk about, you mentioned this difference between kind of like tops, top down and bottoms up kind of like sales cycles in legal AI.

37:46Turner Novak:Can you just talk a little bit more about that and kind of how that's sort of played out in the industry?

37:52Scott Stevenson:Yeah, so I think there's sort of been a divergence of products built in legal AI. There's, you know, products like Harvey and Ligora, which, you know, great, great companies. companies, we don't actually encounter them that much because we're so specialized and we serve as kind of a different customer base. But they have sort of had to optimize to sell to the innovation teams at like the AMLA 100 firms. And back with a previous product, we've done that kind of sales cycle before. And it's very different because these innovation teams are kind of going to push top down across a very large firm and mandate usage.

38:33Scott Stevenson:And they're usually going to bring you this long list of like 50 things that they need in order to move ahead. And it's kind of like a decision by committee sort of thing. And we had an early experience before we hit PMF with Spellbook working with these committees. And they send you in very strange directions and ones that I think are maybe not best for the product. So, you know, for example, at a lot of these large law firms, they operate on an hourly billing model and just decreasing all their billable hours is not a positive incentive. Like the incentive structure is really misaligned with AI.

39:14Scott Stevenson:So you like don't want them to get more work done almost? Yeah. I mean, it's, you know, lawyers make a lot of money and especially in these large firms from billable hours. And so what you found was a lot of the time, what the committee is most concerned about is how do we advertise this to our clients? How do we do a press release to show that we're innovative? How do we just constantly shove into our client's face that, you know, we're an innovative law firm so that we don't look bad, you know, compared to the firm across the street. and you know we we noticed that that was happening and like you know communities would ask for things like client portals well we want our clients to be able to log in and like see the innovation you know firsthand and and what what is that what is a client portal it's a place where like you know a client of a law firm can log in and like you know interact with the lawyer or do the work there and we we actually built one of these in in an earlier product we had called rally because we We also had this request and clients hated it.

40:15Scott Stevenson:They were like, why can't I talk to my lawyer in email? I don't want another login to another website.

40:21Turner Novak:So is this thing like tracking what each piece of work that's done and like automatically puts it in, I can log in and see what you did or something like that?

40:30Scott Stevenson:Yeah, it's like you can log in and like see the documents together or like collaborate on the documents together. But the lawyers didn't really like it when we launched it because they didn't want to like the client to see all their like messy, you know, how the sausage was made kind of stuff. So, you know, this feature, I've seen this feature request a ton, a lot at like the, from the really large firms at the MLaw 100 firms that want to like show the innovation to clients. But, but I, you know, I've mostly just seen it, seen it end up failing. And so that with that experience, you know, we decided to take a really different approach where we're going to sell bottoms up bottom up to the actual end users, the lawyers.

41:09Scott Stevenson:And, you know, the vast majority of our customers, we don't even do a sales call or a demo call. It's like, you know, welcome aboard, Turner. You're going to you are going to use Spellbook today. And in five minutes, you know, you're going to be set up and and actually using Spellbook for some real work or demo work and clicking around and getting value from it. And so that like evolutionary pressure has enabled us to, I think, build a very different product that is much more like, you know, cursor-like in how it's baked into, you know, the user's existing workflow. You know, it's not this grand design thing that you're rolling out across a massive firm.

41:48Scott Stevenson:It's like a really practical tool that is always within arm's reach. That's kind of like a win at the back of the lawyer. And because of that, we have like amazing retention metrics, like, you know, our net revenue retention of like 130 % plus, like we're, you know, we're kind of doing more of a land and expand motion rather than top down. But we have a lot of customers and they're, you know, they're growing their usage, expanding their seats. And yeah, we really like that way of building a product because it, it subjects you to kind of different influences, the influences of the actual end user who's going to be using this thing to get work done.

42:26Turner Novak:So there is quite a few different kind of like legal AI software products that have gotten like a lot of like a revenue. Like there people are, people are using it or whatever. Like what is kind of like the, the, the market leaders in like some of these different like a legal AI categories kind of look like, cause I think we talked about two Harvey and Lagora, but I think there's like a lot more, I don't know. Is there like a, is there like an easy way to kind of like educate people for a couple minutes on like what's kind of working in all these different like subcategories of legal?

42:55Scott Stevenson:So yeah, Harvey and Lagora fairly, fairly similar. I started with the law firms and like the AMLA 100 types of customers, like what I've been kind of talking about.

43:04Turner Novak:And what is their product? Like what do you use when you're using it?

43:08Scott Stevenson:I would say it's a very broad platform that's, you know, broadly kind of like chat GPT for law. They have a number of different things they do, but it's quite broad because they're rolling it out to like a whole legal team that might include litigation teams and transactional teams and so on. And so it's kind of like, you know, if you imagine, yeah, like tuning, you know, ChatGPT or Claude, you know, for a legal use case.

43:36Turner Novak:Is it kind of like a whole sort of like operating system to run your law firm on?

43:40Scott Stevenson:The work, yeah. I think that that's more what they're building. It's like this all-encompassing kind of operating system kind of thing for a firm. but very tuned towards the law firms. Whereas, you know, we've had really amazing product market fit with the in-house legal teams who don't care about the billable hour, who, so this is the other type of customers, is the in-house legal team. And they're starting to sell to that customer base too. But, you know, it's very different because they don't care about the billable hour. They don't care about showing the clients, you know, their clients, you know, the legal innovation they're doing.

44:15Scott Stevenson:they really just want tools that they can switch on and deal with this like hair on fire problem of like I have too many contracts to deal with I need to clear my queue I need some way out and so you know that's that's kind of where we've really been shining is in that that segment in terms of other companies it's probably like personal litigation type of stuff oh yeah there's litigation you know there's like even ups done really well for instance that's personal litigation Personal injury. Personal injury.

44:46Turner Novak:Okay. There's that one, Even Up. People that know Even Up listening to this, they can write in the comments what Even Up does. I've definitely heard of that one before.

44:53Scott Stevenson:Yeah, Even Up is kind of AI for personal injury cases.

44:57Turner Novak:Okay. And then there's like other, there's sort of like corporate advisory type stuff. Like isn't there a company called Hebbia?

45:05Scott Stevenson:Oh, yeah, yeah, yeah, yeah. Hebbia was pretty broad at first, but from what I see, they're really trying to take the position of being for finance now. So, yeah, they've gone kind of very deep down that angle. I don't know if they're still running. They're kind of like legal arm anymore. Yeah.

45:23Turner Novak:And is there is there like a couple others? Or there's like maybe like a longer tail?

45:28Scott Stevenson:There is a very, very long tail. I would say other small, smaller companies that are have launched. Like Sandstone is one that does kind of in-house enterprise legal that they've launched pretty recently. And then there's like a really long tail of other startups doing similar things that, you know, I think a lot like this vertical has matured. This AI vertical has matured so, so fast that it's been, I think, really, really hard for this long tail of companies to catch up to the point where like, you know, it's almost like every three weeks now. one of these small companies comes to us, you know, looking for maybe an acquisition or something like that.

46:12Scott Stevenson:And they've built like decent customer bases, but you know, it's like, it's shocking, like how fast this vertical has moved.

46:20Turner Novak:Oh, so you have, you have you guys done any acquisitions or exploring some or?

46:24Scott Stevenson:We are actually looking at two now. And yeah, part of our strategy this year is definitely to, you know, roll up some of these smaller companies that couldn't, couldn't quite get a foothold. The market moved a little bit too fast. They've built, you know, maybe something similar or more lightweight. They have a little bit of a customer base. And like we look at the math and it's like, you know, we can spend money on Google ads or we could just, you know, acquire a bunch of these small companies. So I think like there has, you know, there is consolidation happening for sure. Like legal AI has been very hyped.

46:56Turner Novak:And that's fascinating because you think like there would probably not be as much consolidation like this sort of early into a hyper growth. market.

47:06Scott Stevenson:Mm-hmm.

47:07Turner Novak:But it's probably that there's, it's just like these like massive fluctuations in like product capabilities, adoption.

47:14Scott Stevenson:Yeah. I mean, it's so fast. I mean, the speed you have to move to keep up with the market is really, really fast. Yeah.

47:22Turner Novak:Have you guys found like, were there like certain times where the models would maybe like open AI or Panthrobic would release new model and just suddenly like Spellbook worked so much better or kind of like, I know a lot of people have kind of had those.

47:33Scott Stevenson:Yeah, I mean, that definitely has happened. Like, we started building Spellbook on GPT-2. So, like, that was tough. GPT-3 was a little bit better. Like, and yeah, it was funny when ChatGPT came out, everyone was like, oh, my God, what's going to happen to Spellbook? Like, everyone, like, this was 2022. Everyone was worried about it. And when ChatGPT came out, you know, our growth just kind of exploded because, like, the model started getting better that we were using. and like lawyers were getting their feet wet in these like kind of generalized AI experiences and then searching on Google, like I want Chattapd for lawyers.

48:09Scott Stevenson:Like that's literally what they search and then they would find spell books. So every time like these model, the generalized models and platforms have launched new things or gotten better, it's generally been very good for us, both from like a capabilities perspective and, you know, in terms of, yeah, Yeah, just getting lawyers interested in AI enough to look a step deeper. One mantra we have at Spellbook is like, that I think a lot of other companies have maybe gotten wrong is like, and I think it's important for everyone to think about when they're adopting AI is we say, you know, it's time to cut trees with a blunt, chop down trees with a blunt axe.

48:53Scott Stevenson:Like there's the Abe Lincoln quote, like, if I had six hours to chop down a tree, I'd spend like five hours sharpening the axe. And I think a lot of engineers, a lot of knowledge workers, we're used to the idea of like mastering a tool and then getting dividends from that mastery. But the reality in AI now is like, there's no time to master anything. Like every six months, a new tool or a new model comes out. And, you know, we've had to teach our engineers like, look, we can't sit around optimizing around, you know, GPT-3 because GPT-4 is going to come out in six months. And we can't try to master this thing.

49:30Scott Stevenson:No one is going to have time to master this thing. So a really important part of our culture is, there's no point in sharpening the axe when the chainsaw is coming out tomorrow. And so what we teach our team to do is drop the axe, pick up the chainsaw, stop and keep moving on, marching forward, implementing new models, new techniques, delete old code very quickly when it's not needed. And as an engineer, I think it's like an unintuitive culture. I think one of our advantages is also just the culture that we've been building these products since 2022. And our team has learned how to do this. Which I think the natural instinct of a lot of experienced engineers is in the opposite direction.

50:15Scott Stevenson:Like I'm going to build a really complex, robust system around this model. But then the next model comes out in six months and then, you know, you just have to delete all that code. So it's, yeah, it's an interesting time to build software.

50:30Turner Novak:Yeah. Because I mean, I guess, isn't there this risk though that the models don't get better? Like the chainsaw doesn't come out, right? Yeah.

50:38Scott Stevenson:And then it's like your axe isn't sharp and you're screwed. Yeah.

50:41Turner Novak:I mean, I guess they can kind of go both ways.

50:43Scott Stevenson:Yeah. Yeah, I mean, there is that risk, but yeah, I mean, I don't think we're there yet. I don't think we're seeing this sort of plateau yet.

50:51Turner Novak:That's right. And is there like a reason that you have that specific viewpoint? Like, what are you seeing to make you so confident? And then maybe how does that relate into how sort of like the AI software is going to change? Like when you look five years in the future, like is there still just like so much more room to like use current capabilities to make the products and the features so much better?

51:15Scott Stevenson:Yeah, I mean, so, I mean, I think you just look at the trajectory. It's like, you know, I'm Canadian, like you're skating where the puck is going. You know, that's, and the thing is, the puck, you know, if you just draw a trend line, the puck is moving way faster than it ever has before. We've never seen technology advance at this pace in our lifetimes. So a lot of people, you know, and just trying to do the math of like skating where the puck is going, they're thinking about the puck speed that you know you might have had in like 2015 but it's actually this like really accelerated speed and I think it's just like the math of like what angles you want to go at how ambitious do you want to be like if you're if you are looking at the trend of where things are going and you and you point your angle to meet the puck at the right place like you're going to be a lot a lot more ambitious and we've done that again and again like what we do is we when we start building a feature or product, we aim to build things that are not doable today.

52:10Scott Stevenson:That and that's like a really, really important feature that I, you know, is a hard thing to tell your engineers. It's like your goal is to start building something now that will not work today. It will work in six months when the models get better. That's that's how you actually time the building of these features, because if you build something that's achievable today, it's not going to be that impressive, you know, in six months. So

52:33Turner Novak:How do you know what's like an okay degree of like, not quite yet possible, but will be possible soon?

52:40Scott Stevenson:Yeah, I mean, it's, it's intuition, you know, it's like, it's like shooting a basketball or something, you know, it's, you get a feel for just watching the technology. I think like being plugged into, into X is like really good for just sensing the velocity. I think like X is generally like two years ahead of like, you know, like LinkedIn on this stuff. So if you're plugged in there, you're going to be seeing what researchers are talking about. You're going to be seeing what engineers are hacking on. And if you understand how the tech works, you'll see that like there's a sequence of advancements that are will inevitably be made that are going to make things easier.

53:16Scott Stevenson:And I think like a lot of the advancements now are not even at the model level necessarily, but it's just in terms of figuring out the right techniques for like, you know, how do you schedule an agent to operate in the background rather than needing to be prompted? You know, how do you, what techniques to use for planning and how do you implement planning for long range tasks? These are things that are rapidly being iterated on and you can pull those into your software very easily.

53:44Turner Novak:Are those things you guys have thought about at Spellbook?

53:46Scott Stevenson:Yeah, yeah, a lot.

53:48Turner Novak:Yeah, quite a lot. Is that, is it there yet? Like, is it in the product right now?

53:52Scott Stevenson:Like, like planning for long range tasks or like the scheduling of?

53:55Turner Novak:Yeah, like agents doing stuff. agents doing stuff.

53:58Scott Stevenson:Yeah, yeah. I mean, so we definitely have agents that can do very long running like drafting tasks today.

54:04Turner Novak:So what's an example of that?

54:06Scott Stevenson:Yeah, I mean, one example would be like the financing transaction, like VC financing transaction example I gave. So you have a term sheet and you need to draft like a full set of NVCA docs. That could be like a thousand edits. Like it's actually quite detailed. I don't know if you've seen like the full template set, But like before Laura has edited it, but it's like there's tons of optional language. There's math you have to calculate. It's a very deep problem. And so that's something that we can do quite accurately as kind of a long range task. And it's not just like filling in the blanks. It's like you're cross-referencing these documents, making sure they're consistent, doing math, making sure that the math adds up, like you're calculating share prices and things like that.

54:51Scott Stevenson:So that's kind of, I think, where the state of the art is. But that product, so it's public associate is our agent product. We started working on that, like I said, before it was possible. We launched the first version of that product in like alpha, like almost two years ago now. So like it was the first long running, you know, agent for multi-document legal work ever launched. And it didn't really work when we launched it, you know, but then the models got better and better and we got feedback. And then, you know, today it works really, really well. So, yeah. And then the next thing that we're really excited about is like agents working in the background.

55:27Scott Stevenson:I think, you know, when people think about like, you know, is AI overhyped or not? The biggest thing on my mind is, you know, the way most people use AI today is, you know, you put in a prompt and it works for like five minutes and then you get an answer back.

55:43Turner Novak:It's just kind of like better Google.

55:45Scott Stevenson:Yeah, better Google. Yeah, better Google. But the thing I think about is like, imagine, you know, having an employee like that. You know, you go to the employee, you ask them a question, they work for five minutes, they give you an answer and then they do nothing. You know, that's like the worst employee ever. That's what we have today. It's like the worst employee ever who's who, if you go over their shoulder and ask them a question, they'll work hard and give you something. But after that, they do nothing. You know, they just kind of sit there and, and, you know, there's such an easy gap for us to jump to say, well, how do we make these agents work in the background all the time?

56:15Scott Stevenson:like an actual employee, you know, pushing the boulder forward without us. You know, I think that is going to be like a 10x for AI and agents. So I don't think people understand how impactful this technology is going to be. Because when you have like an AI co-worker in your Slack, who's like, I don't know, fishing through your emails, finding work to do, looking at what your clients are asking for, say, if you're a lawyer, you know, that's going to be just this massive, you know, leap forward in productivity. So that we're working on that now for like, basically getting to the point where, you know, your spell book agent can be in Slack as like this artificial, you know, legally competent coworker that you can delegate stuff to.

56:59Turner Novak:Yeah. Cause I can think of from like the VC perspective, it's like, you almost create this like thing where like you figure out the first engineer at Spellbook leaves and they, their LinkedIn says they're working on a stealth startup and my tool automatically messages them and like gets on a call. And even like, it's like an AI agent that's like doing the call. And then I get all this information and I'm like, and I just get an email. It's like, would you like to invest or not? Or whatever. Like, that'd be pretty incredible if it does that. I don't think we're fair.

57:31Scott Stevenson:I don't think we're that far from that. Yeah. I don't know how we're going to deal with the noise of like AI agents calling and emailing everyone all the time. I mean, I get that a lot. Like, you get all these, like, spam emails. I do, yeah, yeah. They're not very good yet. Yeah.

57:43Turner Novak:And I don't think, like, if I'm that founder, am I going to, like, do a call with a VC AI associate? I probably wouldn't, though. But I might do a call with, like, the guy who is, like, making the investment decision or whatever. But in a sense, you do, like, maybe skip through some pieces of the process and just more efficiently get to like, it's like humans making decisions ultimately based on information from the AI. So maybe you do skip some, save some time. I'm not sure. I do go back and forth of like personally trying to wait out, like what are the ways that I just completely lean into AI and what are the ways that I just completely like just choose to not do it at all and just lean more into like podcasts is like interesting, like meeting in person, hanging out for two hours.

58:36Turner Novak:There's like no technology really in this aside from like cameras and mics. And we're just talking and communicating about this thing. And like that's probably like a good way to just like get to know you better and like build a relationship versus like, I don't know, we could have like had our AIs exchanging information or something, but like... That's not interesting at all, yeah. Yeah, this is like you're almost like transcending or like above the technology in a way.

58:58Scott Stevenson:I don't know. That's true, that's true, yeah. I think about that so much with like writing and like tweeting, like to me, like AI writing, like creative or informational writing is like so obvious still today. Like the GPT isms that everyone makes fun of, like, you know, it's not this, it's that thing. Like, you know what I'm talking about? Like kind of. Yeah.

59:19Turner Novak:I honestly don't even catch it because I don't do enough AI writing. Like I don't use it enough for writing.

59:25Scott Stevenson:Yeah. There's like all these tropes that you just pick up on and like, you know, the minute, the minute I see them, you know, I feel like, oh, this is if it wasn't worth the time for this person to write this, then it's not worth the time for me to read it. Because in a way, I think like, you know, the fact that, you know, we're taking the time to sit here and have this conversation, or the fact that someone's willing to actually sit down and write something themselves, indicates that like, they thought it was important enough to invest that time. And that means, you know, for the reader or for the viewer, like, wow, that's an indication that this might be worth my time as well.

1:00:01Scott Stevenson:And it's kind of like a proof of work, like with Bitcoin and stuff like that. I think of it's like, you know, writing is sort of like this proof of work. And you know, the minute someone like at Spellbook sends me like a recommendation of something we can, we should do. And like the whole proposal is obviously written with AI. I'm like, I can't, like, this doesn't, I can't trust that you thought, actually thought about this enough, you know? So to your point, we're to not use AI, I think. Writing is an area I'm like creative writing, writing recommendations, something I'm careful with. Luckily, contracts are very like...

1:00:37Scott Stevenson:They're not meant to be creative whatsoever. They're very formulaic. Lawyers are not trying to be original. And so that's one reason I think also why legal AI has taken off so much, especially in transactional work is like there's there is no desire for really originality in contracts. People want to use standard language.

1:00:57Turner Novak:Yeah, because because you think about it like a business contract is almost the it's like the programming language of business, I guess. Yeah, right. If you want to really get philosophical about this stuff.

1:01:08Scott Stevenson:Yeah, exactly. It's just like code. So AI for code and AI for legal, I think in both of these areas, you're not trying to write creative original code or creative original contracts. You're trying to make functional documents. And that's why I think AI works really well in those areas.

1:01:27Turner Novak:Yeah. And I want to talk a little bit about maybe early spellbook stuff because you have some interesting history of the company. But going back even further than that, you, your first company that you started, you like made an instrument, like invented an instrument. Yeah. So what was that? And then how did it go?

1:01:43Scott Stevenson:Yeah. So that was my, you know, first kind of like ill-advised startup. Yeah. I was super into electronic music, electronic, like grew up, I studied computer engineering, but like I grew up, you know, making electronic music, DJing, things like that. and you know I met this composer who was composing these like awesome you know pieces for of classical music but incorporating these like electronic elements and he was like it's really frustrating that there's no electronic instrument that like fits into that atmosphere that the audience will really appreciate like if you go with like a DJ turntable to like a classical concert like people are like I don't really understand like is this person just like hitting buttons or whatever like

1:02:25Turner Novak:they press play, but then they're like, yeah, it's like, are they actually doing anything?

1:02:28Scott Stevenson:Are they not? And so we conspired to build this like instrument that would allow an electronic performer to like really show the cause and effect of what they're doing for these sorts of electronic performances and, you know, shape kind of like guitar or something. It has this like beautiful wooden frame, kind of like an acoustic instrument.

1:02:45Turner Novak:And you almost like hold it like you would an accordion, like sort of in front of you like this.

1:02:50Scott Stevenson:You hold it on your lap or whatever. Yeah. Yeah. So it can face towards the audience. You can use it in like a desktop mode as well, But it has all these lights and things, so it kind of makes it obvious. And there's like buttons and like sliders kind of or something. Yeah. Button sliders has like a synthesis engine. It has like a drum machine. You can use it in all these sorts of ways. And that was like my first very naive startup when I was fresh out of college. You know, very little money.

1:03:14Turner Novak:So you like created this thing, made it, and were like kind of mass producing them, but not.

1:03:19Scott Stevenson:We started producing them, selling some of them. We ran a Kickstarter. You know, a couple of things happened that kind of like, you know, kind of three things happened that, you know, kind of set it off course. One was I got a really big legal bill. So you know, one of my first bosses, this guy Wally Haas, who ran this company Avalon Microelectronics that I worked for is my first one of my first internships. He invested 20K in the company is, you know, and that was a lot of money for me as like a broke student. And you know, he was like, you know, this will get you to your next milestone. And then one day we got like a 10K legal bill by surprise that took half that cash out of the bank account.

1:03:59Scott Stevenson:And for me, that was an enormous amount of money. It was half the angel check we had. And the amount of value we had gotten from that seemed like very, very little. So at that point, I started thinking about, okay, I think there's like a way bigger problem to solve than like electronic music instruments. So, you know, that is where the idea for Spellbook came from was that like frustration. But some other things that happened like through that experience, like a hero of mine is this guy, Roger Lin. He built one of the first digital drum machines in the world. So if you listen to like 80s music and hear that like snare drum with like the big echo, you know, that might be like a Lin drum.

1:04:40Scott Stevenson:And, you know, I met, I got to go to NAMM, which is this big trade show for musical instruments. And I got to meet Roger Lynn, this like hero of mine and, you know, awesome guy, super friendly. But his company was still only two people, you know, and he had dedicated his life to building his instruments. And I was like, I really, I'm really glad he did that. And I really appreciate it. But like, I don't know if, you know, electronic instruments are like, you know, what I'm going to dedicate my life to. I don't know if it's a great market. And, you know, the TAM for niche electronic instruments is, you know, pretty small.

1:05:17Turner Novak:Maybe a million dollars, maybe a little more or less.

1:05:20Scott Stevenson:Yeah, these are like, they're expensive to build. And yeah, I learned a ton about like producing hardware. And it was really fun. But maybe we'll get back to it as a hobby someday. But like the legal market, you know, having that pain myself. And just seeing the size of the TAM of like, you know, how many people touch contracts every day. It's gigantic. You know, like companies like Harvey and LaGora really focus on the lawyers and the lawyer market. And I've talked a lot about lawyers, maybe 20 million lawyers in the world. If you think about how many people, you know, kind of touch contracts, it's way, way, way beyond that, that 20 million number.

1:05:57Scott Stevenson:So I was really inspired through that experience to start Spoblock. Yeah.

1:06:03Turner Novak:Like every salesperson, when you close a deal, there's a contract related to it that you probably touch. Yeah, exactly. Any, any, I mean, it's really any kind of like business transaction that happens. There's like, you know, there's some handshake deals. Maybe you don't sign a contract, but like most people do, even that the handshake deals that I do will do like a one page contract, which is just like, we won't screw each other, but like, we're just making sure that we can't screw each other basically with our, with our very rough contract. Yeah.

1:06:33Scott Stevenson:And even like an email can be considered, you know, a contract, you know, legally as well. Yeah.

1:06:38Turner Novak:And so you were like, holy cow, I paid half of my bank account for this legal bill. I'm going to try to fix the legal market. Like how did, what happened from there?

1:06:48Scott Stevenson:Yeah. So I actually, you know, I stood on the idea for a while. I kind of, I worked at a network monitoring company and was the director of engineering there. I'm kind of building out that product for a while. And then I was kind of working on this in the background, trying to figure it out. And we went through a ton of different iterations. Like first in my mind, like smart contracts were big and I was like, oh, like maybe like Ethereum smart contracts will like be this automated type of contract we can all use. That really obviously didn't, wasn't going to work for a bunch of reasons. Like we showed like blockchain smart contracts to a lawyer and they're like, I will never use this.

1:07:26Scott Stevenson:So, you know, that, you know, we threw that away pretty quickly. But where we kind of landed was we had this product called Rally. It was a template based product. So there was no AI at the time. We actually launched in 2018 originally. And we sold that to about 100 law firms. And basically what it let them do is build these really advanced legal templates. So if you're doing a bunch of NDAs or sales agreements, you can build a template on our platform with, which goes, it does a lot of things that you wouldn't be able to do in a normal templating engine. It's built on Word, it can ingest legal data, and then you could kind of spit out contracts much more efficiently.

1:08:02Scott Stevenson:But for a bunch of reasons, like it didn't, And there wasn't real PMF there for a long time. We were able to do 100 sales. We had raised some money. Our board was like, let's get on with it and scale this thing. And we're like, no, we don't think we have PMF, a product market fit. Our view of product market fit is basically the customer is pulling the product out of your hands faster than you can keep up with. And until we hit that, we're not scaling the company. So we kept the company super lean for a really long time as we built that out. and we actually launched like over 100 landing pages.

1:08:37Turner Novak:This is like a three-year period. Yeah. I think you posted this one chart where it was like, it was just like the revenue, I think, of the company. And there's a point, it was maybe 2020, where you're like, we lost half our revenue or something. What happened there?

1:08:53Scott Stevenson:Yeah, so we built out the platform. We did sell to kind of like the big law innovation committees at first. And like, you know, We had some very lucrative early customers. And one of those customers, cheering them, was half our revenue. And we literally lost half our revenue overnight. But we felt that that wasn't bringing our product in the right direction. Again, what I talked about earlier, working with these innovation committees, they're not the actual users. And so we're like, you know what, we're going to start selling to these small firms, solo lawyers to start and snowball our way from there.

1:09:29Turner Novak:So this is around when you started to test like a landing page. So you launched a hundred landing pages in three years. What does that mean on like practical, like you were basically doing one every like two weeks roughly.

1:09:40Scott Stevenson:Yeah, exactly. So we launched one every two weeks. Sometimes we would actually launch a product variation with the landing page. So like at one point, you know, our view was like, if we roll the dice enough times, eventually we'll figure this out and find product market fit. And we're optimizing for the number of app ads we could do. At one point we launched Shopify for law firms. So we took our templating engine and we put a store on top of it. So like a law firm could stand up like a Shopify store, like need an employment agreement, like click here and then it'll use the template to like spit one out the back end.

1:10:08Scott Stevenson:Like we tried an absurd number. We launched the client portal and we built landing pages for a ton of these different

1:10:15Turner Novak:angles and things that we were trying. So what's the importance of launching a landing page for someone who's like, what is this even? What is the landing page?

1:10:21Scott Stevenson:Yeah, landing page is a single web page that usually has no links to anything else. that has a single message, an image that is trying to get you to, you know, sign up for a product or sign up for a wait list or something like that. So it's a, you know, and often, you know, you drive traffic to these through advertising or through social media or through campaigns. So you're not, it's not like people are landing on your homepage and finding it. You're finding a way to drive traffic to it. So what we would do is launch these like little ad campaigns with maybe a thousand bucks or something like that.

1:10:52Scott Stevenson:And then we would drive traffic to the landing page and we would see what's the cost for conversion. Like how many dollars do we have to spend in ads to get a lawyer to sign up for the product from this landing page? And we literally tracked that over 100 landing pages where we could see, you know, okay, we ran this experiment with client portal and like, you know, that cost us, you know, $100 per lawyer. And then we ran this experiment and that cost us$500 per lawyer. And you really get a sense of like what's resonating and what's not. And you really learn like, how do you get a message, you know, straight into someone's past, their like blood brain barrier into their brain, you know, really, really fast.

1:11:31Scott Stevenson:And yeah, then eventually, you know, we launched, you know, the AI product was just another landing page. You know, we were like, okay, GPT-2 was around. I'd use GitHub Copilot for coding. And we were like, oh, this is cool. Like, we're going to try launching like GitHub Copilot for lawyers. We're going to basically just launch another landing page with this. And what did you call it?

1:11:52Turner Novak:Like what was like the buzzword? Like, cause ChatsDBT had not launched yet, right? No, it had not.

1:11:57Scott Stevenson:Yeah.

1:11:57Turner Novak:So what did you, how did you describe it in like a couple words?

1:12:01Scott Stevenson:Well, we called it, we did call it spell book and we, you know, the big thing we had was an image. The thing that we would do in a landing page is there's a headline and there's an image and the thing, and we would design the image or GIF or video, you know, our thesis is like Like your image plus headline has to deliver this visceral sense of value in five seconds flat. Like, and I think our headline was just like, you know, draft and review contracts 10 times faster. You know, Spellbook uses GPT-3. People kind of knew what GPT-3 was on like LinkedIn and stuff. Like it was a little bit of a buzzword.

1:12:38Scott Stevenson:Even before chat GPT, people were talking, people were curious about what is this GPT-3 thing. and, you know, Spellbook uses, you know, GPT-3 to, you know, surgically redline your documents or something, you know, something like that. And then we had the, but the important part was the image, you know, we had an image of like the word window and someone just like, you know, clicking, you know, draft and it just like drafts a clause instantly, you know, and that was like the sort of magic moment that once someone saw, once a lawyer just saw, you know, hitting a button and drafting a clause, you know, from like a headline.

1:13:15Turner Novak:Versus having a hand type or copy paste from somewhere.

1:13:18Scott Stevenson:It was just like cut through. And like there's this, have you ever read this blog post? It's find the fast moving water on NFX. Have you ever seen it? I don't think so. What is it? It's a really good blog post. I forget who wrote it at NFX, but, you know, the author talks about their first experience of seeing Cabulous, which was like a precursor to Uber. and you know seeing this app for the first time I think someone else had it like his friend had it or something and he was like looking over their shoulder and he was like seeing this app for the first time was like a I had a neurochemical response to it like my pupils dilated like blood rushed in my head and I just saw that the future ahead of me was going to look very different than you know what it has been when it comes to like transportation like and I think that was like the magic moment we were trying to hit with the public.

1:14:11Scott Stevenson:And we hit it. And within three months, we had 30 ,000 waitlist signups. Within three months, we had more revenue from that product than the other three years we had selling everything else. So it was when you hit that kind of like resonance, you feel it, you see it in the numbers. It's also unmistakable. I don't know if you need to actually measure it as much as we did, because when you really hit it, it's like PMF resonance. It's, it's, it's, yeah, it's unmistakable. You will know it. Yeah.

1:14:41Turner Novak:It's like, it's like something that you can't quantify because it's just vastly like order of magnitude or more of just like way more resonance and usage of the product.

1:14:52Scott Stevenson:Yeah. Signups. I mean, yeah. I mean, I mean, we couldn't measure it too. Like, you know, the other things we were selling, the landing page might've cost us, you know, a hundred or$200 in, in ads to first sign up. When we first launched this, it was like$5,$10 signups, almost like an order of magnitude cheaper in advertising to get someone to sign up.

1:15:11Turner Novak:So you basically pay$5 to get someone to sign up, and then if they convert and use it, they may pay like$10 a month or something, and you might get like a 20 % conversion rate from like free signups to paid. And then you could do the math of saying like, okay, we need to acquire a fully converted user, cost us about$25 and they pay us$10 a month. So within three months, we are making money off of that customer.

1:15:35Scott Stevenson:Yeah, that would be like your CAC payback. Yeah. And I think our original price was like$49 a month. Today, it's more like$500 a month. So like, you know, VCs also talk about like, oh, our price is going to go down. Actually, our price has only gone up as we've added more and more value, you know, to our customers. So, but yeah, that's how you can do the math. And it is measurable. And like our, when we hit that moment, like our salespeople's calendars were just completely blocked. Like every day of the week, they would have eight sales meetings, you know, or onboard onboardings. Actually, we didn't do sales meetings.

1:16:06Scott Stevenson:It's just onboardings, you know.

1:16:08Turner Novak:So what were some of the other biggest things you think then you learned over that like launch, launch or landing page period? And then just like this thing worked and you just had to like start going. Like any lessons?

1:16:20Scott Stevenson:Yeah. I think the biggest lesson for us was, or what I would tell other founders is, yeah, if you keep beating, resilience is incredibly important, obviously. If you believe in something enough and keep beating your head against the wall long enough, you will probably figure out something eventually. And if you keep your burn rate low, you can make money last a very long time if you're scrappy. But the secret is you have to really believe in the problem. The problem, if you don't believe in the problem, you know, we, if we didn't believe in the problem that we were solving, we never would have, you know, gone through a hundred landing pages for a year, years of grinding with the super small team with very low salaries.

1:16:58Scott Stevenson:Like the reason we were able to do that is because we believe that, you know, legal efficiency was unsolved, that software still had not really changed how law was practiced and that a lot of people needed legal services, you know, less expensive, you know, a lot of in-house teams needed more efficiency on their contracting. And we just really believe in both one, this problem is huge when you think about the scale of it and two, that it is unsolved. And if you believe those things, you can be very resilient.

1:17:32Turner Novak:So then when this thing really started to work, did you, you did not have a product yet? And were you just like, oh, we got to like actually build this?

1:17:39Scott Stevenson:Oh, no, we did. I actually, I built the product, a really crappy version of it that like our Eng team tore apart. But like, I built, first we thought this was going to be like a lead magnet. Like we didn't even think people would, this would be like a product. We were like, we thought it'd be like a cool marketing splash. Like, oh, first company to bring GPT-3 to lawyers. Like, wouldn't that be like a cool press headline? And then people will convert to our other product. That's what we thought would happen.

1:18:03Turner Novak:Really? Okay. And then eventually this became the product, right? It is the product.

1:18:09Scott Stevenson:Yeah. Yeah, it is 99.9 % of our revenue is from Spellbook today, a very small amount from our previous products. But yeah, we actually built the prototype on Replit. This is before Replit had AI features. I built it in like a couple of weeks to like evenings and weekends. It wasn't a board goal. Like no one really knew it was being worked on. It was just like a fun little side project. Yeah. And I built a really crappy version of it on Replit. And Replit ended up using us as a case study after. Because the thing I said is, it was such an amazing platform because I had this bolt of inspiration and it would have perished.

1:18:51If it weren't for a platform like Replit, I would have just not found time to work on

1:18:56Scott Stevenson:it and it would have just died in the shower where I had the idea. But because ideas are perishable and you have to rapidly chase them down, because I was able to deploy it really fast. Yeah, we actually did have a product, you know, on day one. It was it wasn't great. The engineering team ended up having to like basically rebuild it from the ground up.

1:19:16Turner Novak:But well, and I think the other interesting thing that I've heard you say before is that you were never like attaching these experiments to kind of a legacy product like you. This is called Spellbook. It was a different sort of product.

1:19:27Scott Stevenson:Exactly. Yeah, I think that's that's one of the things I would have done even more differently. Like we learned this towards the later phase. like, you know, I think there's this instinct for founders pre-PMF to just keep stacking on features and like the list and the website gets longer and longer and more complex. That does not make it easier to pitch your product, especially in the earlier days. What you want is a pointed, easy to understand thing. So with Spellbook, you know, our goal was like, we're not going to talk about any of the other things we built for the last three years. We're only going to talk about this in a really pointed way to make it super simple for people to comprehend.

1:20:02Scott Stevenson:And that was part of the success. Yeah. So yeah, my advice to founders launching these landing pages would be like, just focus it on one thing. Don't, don't list off everything you've built because chances are if you're pre PMF, like 80 % of what you built doesn't matter and you can delete it. And there, there will be like one or two things that matter.

1:20:19Turner Novak:Yeah. I think the other thing too, is like, you, you think there's like all this like historical context and history of the product, but literally 99 % of people just, they're seeing you for the first time and they don't give a shit about the history.

1:20:30Scott Stevenson:No, people are so busy. They are not paying attention. Yeah.

1:20:34Turner Novak:Yeah. Like one thing I think I need to do more is just generally tweeting like, hey, by the way, like I have a, I'm also invested in startups, like, you know, whatever. Because I like don't do that enough. And there's like so many people that like they've literally, they've like followed me for like years and they're like, oh, I thought you were just kind of like a meme account or something. Like, I didn't know you actually were an investor. They're like, oh man, I got to like, so sometimes it's like that of like just saying the message over and over again, like reminding people the thing that you do almost.

1:21:01Turner Novak:Yeah, this is one of the things I've been learning

1:21:03Scott Stevenson:from Keith since KB's come on board. I think that's something we were going to chat about.

1:21:10Turner Novak:Well, I wanted to ask you, so you guys really took off. You started going really quickly. So what happened? Did you, like, was it instantly you're like, okay, we have PMF, like we're leaning into this. Was there like a debate of like, like how do we need to figure out this like legacy old thing? Like, how did you manage that?

1:21:28Scott Stevenson:It was funny. Like, no one knew we were working on this thing. The board had no idea. You know, by the time we got to our board meeting, it was just, it was almost like a mic drop moment. It was like, we launched surprise. We launched a new product. Here's the growth chart. And it was just like immediate consensus. It was like, wow, like everyone's like, you need to go chase this thing. Don't worry about the other stuff.

1:21:50Turner Novak:And you'd been holding off, right? Like, weren't your investors kind of like, we have PMF, like we need to start scaling. Yeah, yeah. But at this point, it was like, we truly do.

1:21:59Scott Stevenson:Yeah, yeah, yeah. I mean, yeah, it was a pretty amazing moment for the team. Because we were so, like we went through like the 2020 era, like the Zerp era, pure Zerp era, where like people were scaling way too fast pre-PMF. And like we really, really resisted it. And sometimes we felt, you know, insane for doing it. Are you missing out? Because you didn't do like a remote, like video calling tool or something. Yeah, yeah. Yeah, and like, and like for not scaling the company, like everyone around, all the companies around us were just scaling, scaling, scaling, whether they had PMF or not, there was a lot of capital sloshing around and our investors were feeling like the pressure also, I think to some extent to like, okay, let's get the show on the road.

1:22:42Scott Stevenson:And it was a very validating moment for the whole team to be like, this is what we've been talking about. This is real product market fit. And the board was like, yeah, you're right. Like, I'm glad we waited, you know, for this moment. And then it was just very fast consensus. It was like, we have to scale this now.

1:23:03Turner Novak:I think what I first came across you, like, maybe I'd seen you, but it was like a non-conscious, like consciousness, like, oh, that's like you tweeted just, it was like a graph of your like ARR growth or something. And you're just like, you were like going to San Francisco to fundraise. You just like posted the growth. I remember seeing it. I was like, oh, nice. That's pretty cool. And I just like retweeted it because I was like, thank you. I hope that you have a great fundraising round or whatever. And so how did that process kind of go of like raising this?

1:23:33Scott Stevenson:I think that one specifically was a Series B. That was our recent Series B. Yeah. Okay.

1:23:37Turner Novak:So just like for take us like kind of like beginning to end, just like how that process went.

1:23:41Scott Stevenson:So, I mean, I will say it was the easiest raise we've ever had because it's just numbers and metrics at this stage.

1:23:48Turner Novak:Were you just like kind of showing the spreadsheet?

1:23:51Scott Stevenson:Basically showing the numbers, showing the growth. And it's been really good. And it's much easier than the pre-PMF phase where you're like, you know, convincing of just on pure vision, at least for me, it's easier when you kind of point to the numbers. But yeah, we started the raise, I made a tweet and I was like, hey, My goal of raising is usually to compress it into as tight a time as possible because I want to be working on the product. I want to be working with our customers. No offense to like, I like hanging out with VCs, but to move the company forward, I have to be constantly working with our team and our customers.

1:24:32Scott Stevenson:And so you want to compress it and you also compress it in order to create deal tension. Like if you dilute the deal tension across six months or whatever, you know, it, no one moves, everything is sluggish. And so I tried this, I made this tweet, I tweeted our like growth chart. And I was like, yeah, raising our series B, you know, I'm going to be in New York this week and SF the week after, you know, and that was it. And so it went viral. I maybe because of the chart and And like we got like bombarded. My email, I still have emails I've not responded to from that moment from funds who reached out.

1:25:12Scott Stevenson:And yeah, I set up in a hotel in New York for like a week. And yeah, most of the investors all like came to us. So like one of the strategies I learned from another founder is like, you know, try to, if you can sequence your meetings, you know, we rented a boardroom in the hotel and we just said, you know, come here, we're taking meetings this week. kind of dictate the schedule. So we had a lot of investors come by there, visit a couple offices, and then did the same thing in SF, basically. And then we got in touch with Keith, and he was like, you know, I only take in-person pitches. And he was in New York, and I was in SF, and I had to fly all the way back and cut the SF time short.

1:25:55Turner Novak:And wasn't he, like, the only one who didn't come to you? Basically, yeah. Basically, yeah.

1:26:00Scott Stevenson:Yeah, I also went to the like the KB office as well. I'm assuming I can't counter there as well. Yeah.

1:26:09Turner Novak:And what was kind of the difference? I know you said there was a pretty big like mindset difference almost between East and West Coast investors. What did you kind of experience there?

1:26:16Scott Stevenson:Yeah, for me, it was very night and day like the for the most part, you know, besides Keith, who I think is a very like he's smart on the numbers, but he's unique in how qualitatively he assesses the world. He's very willing to bet on qualitative things. But the New York investors are extraordinarily quantitative. And it seems like they almost have all converged on the exact same spreadsheet that they use, the exact same metrics, exact same benchmarks, exact same spreadsheets. And to the point where it's a little bit absurd, because if everyone's looking at the same spreadsheet, then where's the alpha?

1:26:53Scott Stevenson:If everyone's looking at the same thing, they're going to pay the same price. Everyone's going to bet on the same companies. I found And there's very little emphasis on the qualitative side with a lot of the New York investors we pitched. Whereas in SF, there was a much deeper focus on the qualitative vision of the company and how is the future going to play out and why and kind of like the idea maze, kind of exploring the idea maze and understanding where things are going to go. Very different.

1:27:25Turner Novak:Yeah, very different. And then you end up Keith and KV, Coastal Ventures, led the round. What has it been like working with Keith just over the past couple of months? Yeah, incredible. So yeah, we've had a few board meetings and yeah, Keith's an incredibly sharp investor.

1:27:45Scott Stevenson:He's been in the weeds. Like he really understands how things work on like a deep, deep kind of like CEO level. And yeah, like we just learned a ton. He obsesses about performance and like what best in class people look like. And you just learn so much from someone like that. One of the biggest things I've learned from Keith and that I'm learning is like how to communicate really well. Like this guy is like a nuclear grade communicator, like how incredibly concise he is, how he's able to, I think he's really good at like counter positioning companies and opportunities to cut through the noise in a way that other people really struggle with.

1:28:34Scott Stevenson:and he thinks about it deeply. It's not like by chance or he just happens to have this talent. It's like he's consciously very good at thinking about how to get the message to listeners and how to create a movement with a message in a way that I think it's one of the hardest skills for anyone to learn. It's a marketing skill. How do you get your message out into the world spreading? To do it, you have to have a really simple message and you have to repeat it a lot. Like you were saying, like, you know, people not knowing you're an investor, like you have to find a way to repeat that message or get that message in front of people because people are so busy, so distracted.

1:29:11Scott Stevenson:They have no time. Like getting someone to read more than five words is an extremely hard challenge for the most part. Getting someone to watch more than a, you know, five second video clip, you know, is pretty difficult a lot of the time. So, I mean, he's just so aware of that and so good at dealing with it. Like the example I'll give you, we did like a series B announcement video. And, you know, I think we booked like 45 minutes for Keith to come down and talk about like his view of the company and the opportunity. And we're sitting down, you know, in front of cameras kind of like this. And, you know, Keith sits down next to me and he just hits his lines.

1:29:49Scott Stevenson:He's just like, boom, boom, boom. He has five bullet points about why, why this opportunity is incredible. You know, why, you know, the contract opportunity in particular is really special and how the market data that we're collecting is going to change how contracting is done. I mean, he just, he communicated everything in like five minutes and then he was like, okay, I think we're done. Like he couldn't think of anything else to say. Like that was it. And like that's the pattern that you notice in the board meetings too. It's like, you know, you'll ask for feedback and he'll say it in like one sentence, you know, and then he'll be quiet and it's like he's so good at getting to the heart of the matter and then letting the core message like breathe and be received.

1:30:34Turner Novak:Are there things you've changed about like marketing or messaging over the past couple months then?

1:30:41Scott Stevenson:Yeah, definitely. I think focusing more on delivering, you know, our core message, what we're all about, is and repeating it, you know, to the point where it can become kind of boring for the speaker. You know, I think just doing that more and doing it better.

1:31:00Turner Novak:One last thing I wanted to ask you about. So you, I feel like you're maybe like, you know, bleeding edge of AI, quote unquote, like, personally, kind of what is your like personal AI stack look like? Like, what are you using? Like, what kind of products and like, things are you taking advantage of? I've tried a ton of stuff.

1:31:18Scott Stevenson:You know, obviously, I use like cursor and cloud code on the engineering side for mainly building prototypes and things like that. The product I'm loving right now is Twin. Have you tried Twin.so? Have you ever seen it? No, I've never. I mean, it's on the surface very simple, but I think they just got like this generalized agent formula really right. So how it works is you can go in there, you can say, I want to build an agent. You build an agent by prompting. You don't have to like write code or connect together boxes or anything like that. That's always the most frustrating.

1:31:51Turner Novak:There's one called like N8N, I think. Yeah, I never ended up getting it to work.

1:31:55Scott Stevenson:So I was like, I don't care enough to figure this out. Yeah, the way this works is you're almost like vibe coding these agents with a couple prompts. And it's really good at scheduling the work in the background. Like what I was talking about earlier, it's like you don't want an agent that you have to prompt to do work. You want it to work on its own. And so I built probably about five agents with Twin now that I use daily. Yeah, one of them is my Canada recruiting scanner. And what it does is every morning at 8 a.m., it scans Twitter. We're generally, we hire in the U.S., but, you know, most of our team is in Canada.

1:32:31Scott Stevenson:We basically scan all of tech Twitter and find, like, Canadians who are tweeting about AI and looking for engineers, you know, designers and interesting people saying interesting things. And that fills our queue for, like, recruiting. And that's been a huge help.

1:32:45Turner Novak:Do you reach out to them or is, like, the team, like, what's the process for that?

1:32:49Scott Stevenson:Yeah, I mean, we manually kind of evaluate like myself and like our hiring managers will then look at the candidates and then we will do the reach outs ourselves. We're not at the point where the agent goes and reaches out yet. And we're kind of like tuning the quality and the filtering and stuff like that. So, yeah, that's been really useful. We have another agent that does a similar thing where it's just like digests all the feedback from every channel, from Slack, from email, from HubSpot. And like we'll basically summarize, you know, all of our product feedback, you know, every day in Slack.

1:33:18Scott Stevenson:Like, why are people churning? Why are people expanding? Things like that. So that's probably the biggest one, Twin.so. For me, that's the thing I'm loving the most right now. It's just so easy. It's so fast. I think you also, you want something that's like so easy that like, you know, if you have, you want to make it so that like, if you're dealing with a problem, like, oh, I need to, you know, find my next podcast guest, you know, that it's, you know, so fast to set up an agent to do that, that it almost takes you no extra time. So it's like, oh, I'm going to go look for podcast guests or search Twitter or something.

1:33:50Turner Novak:You almost want like the process of creating the agent is actually faster than just going on and doing the thing.

1:33:57Scott Stevenson:Actually, yeah, that's right. Yeah, it's actually faster. And that's like twin is the first thing that's actually hit that level for me where it's like, I might as well create an agent and it actually works. Yeah. And I love the other thing I love. It has so many integrations, so it can suck in from so many things and then it can pump into Slack. Like, I think the thing I always think about with products, like the hardest thing is getting people in the habit of actually using them and getting the products in people's faces and like getting our team to like go to a new agent product and changing their habits is really tough.

1:34:27Scott Stevenson:But if we can pump the agent output into Slack, you know, into channels that people are in, you know, the usage is much better.

1:34:35Turner Novak:Yeah, that's pretty cool. Well, I'll throw a link in the show notes. People can check it out. I'm going to try it. I'll see. I'll let you know what I do with it. Yeah, but this is a lot of fun. Thanks for coming on the show. Thanks for having me, Turner. And thank you for listening. A quick thanks again to Flex for supporting this episode. Upgrade your spending to Flex Elite to get$1 ,000 on your first card using code Turner with the waitlist link in the description. If you enjoyed this conversation, please like, comment, subscribe, and share this episode with a lawyer friend who is still manually reviewing all their contracts by hand without AI.

1:35:05Turner Novak:Make sure to check out the back catalog of over 100 episodes with the founders of companies like Robinhood and Mercury. Tune in over the next few weeks for guests like Mike and Akil at Footwork, Chris Hodgick at Panover Park, and Sofia Amoruso, founder of Nasty Gal and Trust Fund. If you want to miss any of these, subscribe to my newsletter, The Split, linked in the description. Get each episode, plus a transcript emailed directly to your inbox every week. Thanks again for listening. See you next time.

1:35:41now