Law Firms: Premium Business… or Expensive Labour?

Law firms look like premium businesses.

Big fees.
Nice offices.
Serious people.

But behind the branding is a very different reality.

In this episode of Good Bad Business, Peter and Fabs break down the Australian law firm model using the MOAT framework - Margin, Operation, Advantage, and TAM, to understand what actually makes a law firm profitable… or painful to run.

We unpack:

•  Why most law firms are still labour-heavy businesses 
•  How AI is starting to pressure billable-hour economics 
•  The hidden operational risk behind reviews, handoffs, and client trust 
•  And whether a new firm can realistically generate $1M in year one

This isn’t about legal theory.

It’s about business structure.

Because premium-looking businesses don’t always have premium economics.

And we don’t want you to get into a pickle.

If you’re thinking about buying a business,
 listen first.

If there’s a business you want us to break down,
 send it in.

And if you got value from this, share it with someone before they sign something they shouldn’t.

Because we don’t want you to get into a pickle.

Transcript

SPEAKER_01 0:00

Burbs, we're here for another episode.

SPEAKER_02 0:02

Indeed, we are, my friend.

SPEAKER_01 0:03

What are we up to? Number five?

SPEAKER_02 0:05

Five, yeah, yeah. Wow. Yeah. Time flies when you're having fun.

SPEAKER_01 0:10

We're definitely having fun, mate. I'm enjoying it. This week's topic is about law firms. God help us. Yep. We're going to piss off every referral partner. Predominantly all of our referral partners are actually lawyers and insolvency lawyers alike. So this is one for them. Please don't take offense to it, all of our referral partners. But nevertheless, I thought it'd be a good sector to do a bit of a deep dive.

SPEAKER_00 0:37

Good business. I picked up good bad business.

SPEAKER_01 0:46

Obviously, AI is affecting most businesses at the moment, and law firms happens to be one of them. So we just thought we'd we'd break it down and uh run over a bit of a bit of a new topic. Yep, exactly. So today we're I want to chat about law firms because they they do look like a premium business, don't they?

SPEAKER_02 1:08

Yeah.

SPEAKER_01 1:08

Nice suits. Remember that Netflix episode?

SPEAKER_02 1:12

Yeah, yeah, yeah. I can't watch it anymore since uh um Amiga Michael became who she became.

SPEAKER_01 1:18

So I love that show, but I got it, I got obsessed with it. Um all of a sudden I started buying new suits and all the rest of it. Yeah, yeah. And then I also got all these suits sitting in the wardrobe, but I use any of them now. Everyone coming in the office with t-shirts and yeah, uh God, no, these things happen. But look, a a law firm, it does give you that perception, you know, that that suits perception where it's up high, you've got these beautiful offices overlooking the city. Um, serious people, uh, big alley rates, which is what they proposed um on that show. But on paper, in Australia alone, it's a $30 billion industry. And that's um B with a B. B with a billion. B for a billion. B for a billion. So here's the real question: Is law firms actually a great business, or is it a labor business wearing one of those fancy suits I've got hanging in my uh in my wardrobe? Because once AI writes the first draft, reviews the documents, speeds everything up, the client asks one simple question Why am I paying for the clock? Exactly. You're you're charging me per blocks, but you've just plugged it in and you saved a shitload of time. So this is the question that most lawyers are gonna get asked.

SPEAKER_02 2:35

They're gonna get asked, and look, going refer back to our industry, the email we write 50 emails, 50 to 5,000 emails a day, depending on when you catch us. And I kid you not, I've had lawyers demanding your client pay like 500 bucks to just draft the same email that I would have drafted for.

SPEAKER_01 2:51

That's true. Because, like, if and if your answer's weak as a lawyer, like if your client says to you, well, hang on, if you've just plugged it into AI and you've got a response straight away, why am I paying for a 14-minute block? So if you've got a really weak answer, lawyers are gonna struggle moving forward because I just happen to spend a lot of time with lawyers. Um, as we probably know, you you certainly know the the history of of a pickle and what we went through to change the the trademark infringements and things like that that you had. Quite the origin story. Yeah, it's quite the origin story. So we're not gonna go down that rabbit hole right now, but look, I've had a lot to do with lawyers over over the years, um, certainly have a lot to do with them now. There's some of our best mates, um, and are also not only a great business to be a part of, um, but AI is gonna disrupt it. Because if there's a a 14-minute block and we've got AI, it's like we just alluded to a moment ago, then why should a client pay for that? And and shouldn't the clients actually be negotiating before they actually engage them? Because if a if a lawyer is using AI, then I want 75% off my bill. Correct. Exactly. Because, you know, you've got AI-driven platforms that are out there at the moment. One of them happens to be Harvey, and that's that's engineered for the legal industry. And it is look, I'll have to, you've got to give credit where credit is due. I mean, it is built on proprietary sort of systems, um, and it's legal models, you know, being the large language models. But I mean, it is layered on top of OpenAI, which sort of handles a lot of the heavy workflows that law firms definitely need. So there is an element of AI in respect to what Harvey actually does, but Harvey, believe it or not, achieved decacorn. Have you heard that term before? No, it's so you've got a unicorn, which is a billion-dollar business, uh, but a decacorn is a $10 billion business. Right. So, and they actually achieved $11 billion valuation in 2026. So it's a massive, massive adoption. So if lawyers are saying they're not using AI, why is Harvey worth the $11 billion? Exactly. And here's one for you. Why is it called Harvey?

SPEAKER_02 5:19

What other Harvey's what's his name? Harvey Specter was it? Harvey Specter from So that's that's where they got the name.

SPEAKER_01 5:23

There you go, you were there. I just thought I'd throw you in the deep end there. Well, well, don't you do what you know? So big law firms are definitely adopting that type of technology. But see, with something like Harvey, the the core position is the ability to act as a digital associate. Yep. And and it's not providing general information. They, if my memory serves me correctly, they they have everything on a secure environment because for obvious reasons there's a lot of secure data that lawyers are uploading. And so lawyers will upload it to what's called the private data, which is the which is the vault. And Harvey uses what is called um, it's called RAG, um, R A G, which is uh retrieval augmented generation, and that ensures that the responses are grounded and it's not based on internet data. So it's verified data, and it's also helping with the workflows. So this is the type of technology that is out there specifically built for the law sector. And so this is where I wanted to do a bit of a deep dive because of the the old law firms that are out there that are still charging per block, you've got to be able to say to yourself, well, if the consumer is well aware, well, they if they're not now, they will be with businesses like Harvey, for example, and using AI and putting forward that adoption. The reality is that what makes you worth the money that you're asking for. And certainly a lawyer can respond to that based on history and experience and networks and obviously the successes that they've had, but they're definitely gonna have to frame that analogy of AI, experience, knowledge, success. The framework needs to get better. So definitely let's get into it.

SPEAKER_02 7:23

Let's get into it, let's do it. So we're going right into so the business snapshot. So learn law firms.

SPEAKER_01 7:34

All right, so let's talk about the origin firms. As far as I'm concerned, it seems like it's two industries in one if you sort of break it down. Australian law has always been two identities: 1919, Morris Blackburn, 1935. Um, there was Slater and Gordon that was created. And then later on, you've got the elite firms, the corporate law, um, high-end deals. And so law seems like it's it's both. It's a premium advisory business and a scaled consumer service. And you know that my brain works on consumer scalability. So that's why I thought this was a really interesting topic to try and sort of implement a lot of what the big law firms are doing into our sector as well, because they've been able to scale really, really well. The big law firms are there, there is a reason why they're a big law firm. So then back in 2007, Slater and Gordon list publicly, and that was an absolute disaster, but maybe we'll save that one for another, another podcast. Law does become scalable, and now, as we mentioned, AI is entering the system. So the question is who keeps the economics?

SPEAKER_02 8:50

That's right. Exactly. Exactly. It's uh, and and now going to like the snapshot of the business, right? So law firms don't sell law. Um, they basically sell their advice, they sell their expertise. But in saying that, and and this goes to why the that uh the public listening of Slater and Gordon would have been a failure, is because we know this, even as a brokerage, to give a parallel to our industry, you need to have data, you need to have, you need to have something tangible there to have an intrinsic value. If you've just got loan writers or legal advisors, your your business is only as valuable as the next advice that you're able to give. And that's and that's the parallels we see between our industry and and the legal side. So they're well, they became a franchise all.

SPEAKER_01 9:33

Yep, that's right. Because you you've got all these locations right across the country, and each franchisee is essentially paying you know rent and franchise fees on those locations. Yep. They're not necessarily company owned. Yeah, they also scaled to the UK. They bought another law firm there as well.

SPEAKER_02 9:51

But yeah, like you're gonna watch the margin of the city. Yeah, that's right. So, but and and effectively that they're selling risk transfer, judgment, processes, outcomes, the revenue models are based on hourly billing, fixed fees, retainers, no win, no fee. And the clients are pushing back on time-based billing. But in saying that, there is opportunity as well, in my opinion, because if you're able to say automate it, if you're able to integrate it with AI more, you can actually create intrinsic value in that law firm because now you've got a system that's collecting all that data, all that legal precedent. So you're no longer relying on it. So I can actually see some opportunity for the legal firms if they're able to automate and integrate with AI.

SPEAKER_01 10:30

Yeah, no, agreed. But if you're going down the franchise model, which is what uh Slater and Gordon did, then what is their intrinsic value? Exactly. Where is their equity branding? Because what makes that business different from your local practice? Correct. And if you're running a franchise or model, then you've got the additional expenses, and that's where it starts to sort of compound into the margins, as we've mentioned, you know, on previous podcasts. A franchise or like McDonald's, for example, owns the real estate. They made $38 billion last year and they didn't sell one hamburger. That's right. They got it from the real estate. Yeah. So, you know, what is what is the model?

SPEAKER_02 11:16

Yeah, that's right. And and I wonder if we could, you know, how like funds managers, for example, right? What they sell, what their intrinsic value is, we've got a specific algorithm or strategy of investment, right? You see it all the time. They'll hire analysts and they've got this engine room, which you know they can run for 20 years and no one will know what the secret source is. I wonder if, with this obviously, consolidation of technology, the automation of it, if you can get legal firms becoming something like that, right? So you get a legal firm that's got this unique AI system that's able to record all its own unique legal precedents, its own kind of judgments, and then it's able to create and cater for a very unique kind of legal advice. I wonder if the automation could do that, you know?

SPEAKER_01 11:57

Yeah, well, and we're gonna go down a rabbit hole. Look, it's it's an interesting point. Yeah. Um, and it's something that you and I talk about on the daily in regards to the intrinsic value that a pickle has. Because every time we remove a cave it, every time we deal with a business, we know where the business is located. We know how much debt we've been able to negotiate, um, we know the type of business it is, the type of sector it is. So that is all proprietary data that a pickle has. So therefore, that proprietary data, why can't we train our own LLM? And so to touch on that point, it's relatively easy for a law firm, and I'm not sure, and I'm sure the big firms are thinking this way, but if they're to plug in Claude, for example, do some mods, filter out um some of the data sets, then look at the actual vector database, so being the inferencing um that it's responding to, and do some error checks, have those inputs that are actually put in their own data vault. This obviously drives very, very strong sufficient data for the insolvency space, whether they're in insolvency law, whether it's corporate law or whatever their speciality is, and therefore they can then on sell that data because it's proprietary to them. Like some of these law firms have been around for over a hundred years, they've got a hundred years worth of data. Why aren't you training the vector space and their own um LLM? And yes, look, training an LMM is not a cheap exercise, but they've got the budgets to be able to do it. Now it doesn't have to be one of the frontier labs. It can be um open source, for example. Um it can be something that's less sufficient, which is what the Chinese call the lightweight models, um, which and like you've got Meta, for example, that has Lama. Llama is obviously their LLM, um, but you can you can train it locally. So, you know, if if businesses are not thinking that way, especially and it frustrates me with a with a law firm because you've been around for over 100 years.

SPEAKER_02 14:20

Yeah, exactly.

SPEAKER_01 14:20

You've got an absolute shitload of data, you need to be training that data. That's right.

SPEAKER_00 14:26

Yep.

SPEAKER_02 14:28

And um so basically, yeah, so we were at the yes, a review of um, so are we at the moat stage? All right, I mean, and yeah, look, and let's go straight into the uh moat on that note. Yeah, now that I've gone off my tangent.

SPEAKER_01 14:44

No, I love I'm off my whole high horse language model for you.

SPEAKER_02 14:47

You're a passionate man about AI and honestly, and me, I'm not as technically understanding, but I do understand the practical elements and I do appreciate them. And one thing I will say is considering you're half my age, what's the story? Exactly. What's going on? I'm a bit out of touch. I'm uh I'm an old soul, you know. You're an old soul. You're an old soul. But you know what it is? Um, well, you know, we've all spoken, I've spoken to trainee uh graduate lawyers, and one of the biggest things they tell me is a lot of them get deterred from even going into the legal industry because they spend, you know, most the early years, their first five years, just doing 18-hour days. So I think AI actually has an opportunity to make the entry into the field much more palatable. So yeah, so um on that note, so let's go into M for margins. So let's have a look here. So industry margin average about 15%. People are the biggest costs, namely because the barrier to entry is so difficult as well. Revenue per employee is around 240,000 to 290,000 per annum. And it's not software, it's leverage labor. AI pressures pricing, margin depends on the structure. So I personally I'll say, you know, in terms of the the kind of margin, you know, there is lower overheads, not capital intensive, just like most service-based roles. However, um, yeah, the the the higher the the the sheer reliance on labor, I would probably give it a I'd so say five out of ten and even in the middle kind of range there.

SPEAKER_01 16:10

Yeah, I'm I'm at a one because it's that element of perception that a law firm needs. So their costs, their costs are huge. If you're at a fancy office in the city and you've got massive overheads in relation to the staff based on experiences, and now you've got AI that's eating into your lunch, your margins are going to be reduced substantially. That's right. That's right. Well, and and if it hasn't now, it's going to be.

SPEAKER_02 16:39

Well, what do we always say, you know, uh parallel to financial services, we say, you know, is is it not worth paying the premium to automate something so you have time to do something more, something more extensive, maybe spend more time on less time on hiring um quantity of lawyers, but hiring quality of lawyers or focusing more on bringing on referrals?

SPEAKER_01 16:57

Which is exactly what the moat strategy is again, the moat strategy is all about. So, what is the advantage? And is the advantage adopting AI, a platform like Harvey, for example, and automating the mundane, which means that obviously the the law clerks are gonna struggle to find an opportunity. They might even lose their job, but in most instances, they're gonna be reskilled. And if anything, there's gonna be promotions that are come that are gonna come out of that. Because if the law firm is saving on employing multiple law clerks because they've been able to automate that through Harvey, for example, well, then you're upskilling your law clerks. That's right, exactly. Well, quality, not quantity, effectively. So, yeah, automate. That's that's the key. So yeah, um it's it's a low one for me, mate.

SPEAKER_02 17:46

That's fair. That's fair. All right, well, over to you for operation.

SPEAKER_01 17:49

So if I'm gonna score the operation, I'm gonna be low again. I'm gonna be at a at a two because I'm I'm just not seeing where the ideal process is in respect to what law firms have done in the past to where they are now and what can be adopted via the AI automation. So I'm just I'm not I can't reconcile the two for some reason. I don't know. What do you think? Yeah, it's a it's a low one for me.

SPEAKER_02 18:20

I actually agree. I would even be even a bit harsher than you would say one out of ten, namely because yeah, what you've said is 100% correct. But also, we know that the courts in general, the legal system, it is quite archaic, right? They're not very open to innovation, and rightly so. They need to be a bit more conservative. I agree with that. And the problem is it's like, you know, what's gonna happen? You know, what what role? And furthermore, forget the lawyers, what's gonna happen to the paralegals? All these paralegals that were the engine room, you know. I know lawyers that without their paralegals, they couldn't function. But then if that's replaced by AI agent, but then you've got the dual transition of the customers are more happier with the automation of the AI because it's gonna reduce costs, gonna make it more efficient, turnaround times will improve. But then if the courts are resisting it, and I think in America was it recently, there was a court case where it turns out that they try to use an AI to represent them legally, and then the court found out and the judge absolutely like stripped the guy apart, right? Legally speaking, probably so.

SPEAKER_01 19:15

So yeah, because like in situations like that, it's like in in our industry and like in any industry if if you have a client that has taken you to task and you now have to go to court based on the advice that an AI chatbot has given you. If there isn't a human in the loop, then who's responsible? The person that's responsible is the person that has the qualification, it's the person that has the license. If if we go to if we go to court because we've given the wrong advice to a client, are we gonna go up on on the stand and go, oh, sorry, Your Honor, it wasn't me. It was the AI chatbot.

unknown 19:56

That's right.

SPEAKER_01 19:57

The magistrate's gonna go, well, hang on a second. Well, who's whose license is it? You're responsible. So there needs to be that cross-checking in respect to the data that is that is inputted. What output has the AI in return given? And who is the human in the loop that's actually certified the data that the chatbots have given? Yes. And so, and without that, and that's not going to change. It doesn't matter how much we talk about AI and technology, who's the human in the loop? That's right. Has the authorized representative signed off on it before it's actually given out the advice? Yes. Yep. Yep. That's right. Which comes back down to the data. If the if the law firm has been around for over a hundred years, you've got all this data sheet. So you should be able to train the vector space relatively easy. Correct. But the the certified authority, the human in the loop, needs to sign off on it.

SPEAKER_02 20:48

That's right. Yep. And that actually is a good segue into the next part, which is A for advantage. And you're right. And look, I would even, I'm gonna go a bit more on the machine side of things. I mean, look, let's be real. In what sense has the legal system been able to even quantify and analyze common law precedents, right? Before AI, there was no mechanism that could quantify qualitative data, right? Before we never had a system we could do, we could analyze numbers well, you could put numbers in the spreadsheet, but you couldn't get a computer to analyze, for example, you know, the trends of common law, but now you can. So what I wonder is is, you know, how did that human is is how is that human going to stack up against a machine that can process 100 years worth of legal precedents and laws without the hallucination that we see that happens with AI, and then effectively come to a much more comprehensive understanding? Because I I personally don't see an argument against it from the human side of things, right? So I think from an advantage perspective, I mean, AI has the capacity to really shake things up. But I think in saying that, it might be to the detriment of the advantage of the independent solicitor. Because what's gonna happen eventually is I mean, if AI gets good enough, Why do we need to qualify people with put them into uni for five years and get these, you know, these um uh these lawyers that are coming into the play when a machine can probably understand the common law better, right?

SPEAKER_01 22:12

Yeah, but who's a licensed leader?

SPEAKER_02 22:13

Yeah, and that's where it comes back down to Detroit. Yeah, yeah, yeah. So there still needs to be the fact checking. Well, and also what that what does that mean then? Then I'd imagine, tell me if you agree. I think from an advantage perspective, that just means that the number of lawyers, it's gonna get much more competitive and there's gonna be less room for lawyers um in this space because you're gonna find that one agent, one uh principal can sign off on what say multiple paralegals would have done, like say, say, sorry, what the AI would have been able to assess is the amount that what multiple paralegals can do. But if you only need one expert to just sign off on it, well then how many lawyers do you need in the law firm, right? So I think from advantage, I'm gonna give it probably a one out of ten as well because I just I see with automation, there's gonna be less advantage for lawyers.

SPEAKER_01 22:58

Yeah, I I just I disagree slightly.

SPEAKER_02 23:00

Yeah, tell me.

SPEAKER_01 23:01

Because the the process that a lot of law firms have had over the years is obviously the partnerships. So the more partners they have within a law firm, the more revenue that each of those partners are bringing in. Correct. Yeah. So if one if one lawyer can bring in one to three to five to ten, if some of the some of the big lawyers are bringing in 20 plus 20 to 30 mil uh per um rev revenue per per individual partner. So if you've got 15, 20 partners within the one law firm, well then that that business scales quite quickly based on that one individual. Yeah, yeah. So the more individuals you have that are well connected, the more revenue they're bringing into the business. But then that's a scaling issue now. So then how do you automate that? Exactly, which goes back to your point, then you're bringing in the law clerks and everything else to be able to support the um big revenue earners.

SPEAKER_02 23:58

Yeah, that's right. But then the question is, yeah, but then the question is like, you know, with with a if you're able to say AI is able to automate it effectively. I wonder, then it becomes the question of do I get more people on board and increase my labor costs, or do I reduce my labor costs, maybe lose a few referrers, but in saying that my margins thin down. But I guess AI is impacting the, I think that's the trade-off for every company, every industry is deciding, and that's the fundamental point about AI, right?

SPEAKER_01 24:26

You're gonna you're gonna have the hyper-specific models that have been trained specifically for an industry. Correct, yeah. And so therefore, a lot of the a lot of the staff, the the low-hanging fruit that you would typically hire, you're actually not because you've got the hyper-specific model to be able to assist with the automation of your business. That's right.

SPEAKER_02 24:53

Yeah, yeah. And then it looks happened to people. I know in tech it's happened already where companies are literally looking at balance sheets saying, Yes, I can have, for example, I'm just using example, I can have 10 university of um Sydney uni uh graduates that are the top achievers in their field, but in the end, they'll look at the balance sheet and say, it's much easier for me to cut off that staff, keep one of them as opposed to 10 and automate the rest. So it's a trend that we're seeing, but you know, time will tell.

SPEAKER_01 25:20

I think you know, humans humans are a not only are emotional beings, but we're also creative beings. Yes. And so that means that if you have a domain-specific knowledge that nobody else has, you now can sort of vibe code bits and SaaS software and build out your domain-specific knowledge where you couldn't do that in before because of the cost that was involved. So there's going to be a lot of new businesses that are being created specifically for that. So you might lose your job as a law clerk, but you might have specific domain knowledge and you can go out there and build out your knowledge based on uh the law clerk processes. And that's a thing. And before you know it, you've got you know an AI platform that you could probably sell back to the law firms. 100%.

SPEAKER_02 26:11

And was that that's a throwback to last episode, right? About specialization. Don't worry about it. Specialization. In this environment, you cannot be a generalist because the AI is the true generalist now. Now that's the generalist, and the humans are the specialists, which is exactly what we talk about all the time.

SPEAKER_01 26:26

Who's the specialist?

SPEAKER_02 26:27

Yep, yep, yep. You know what? And I actually found that I read a post online about this. Um, this is a very uh small tangent uh audience, as I assure you. But basically what they're saying was they that's unlock us. That is but it's crazy. Like, think about this, right? Philosophy was essentially a philosopher, was the scientist of the ancient world, right? And then all of a sudden, philosophy turned into so many different branches, ranging from tech to humanities to all kinds of fields. So that's what I believe AI is gonna do to us, akin to the way the blacksmith was the guy that he dealt with all the different metals, all the different minerals, and created something new, but then the blacksmith eventually divided into so many different trades over time. So we'll probably see that with AI as well.

unknown 27:09

Yeah.

SPEAKER_02 27:09

Yeah. So what have you got for the uh for advantage? So I've got a one out of ten, I'm gonna be harsh again just because of disruptions.

SPEAKER_01 27:17

Yeah, the I I'm much at the same because there isn't really the advantage, the only real advantage is what we're talking about before, where you've got hundred years worth of data and you've trained your own hyper-specific LLM.

SPEAKER_02 27:34

Exactly. And and and 100%. And the only reason I would say even that advantage would be a potential disadvantage is because we don't know how it's gonna disrupt people. It's like they always told people if technology is really good at doing your job, you better be concerned that that job is gonna get taken out. And the biggest example is forget everything else. Look at what happened to the coders. Back in the days, learn to code was a hashtag that was trending because it was mocking Midwest American truckers and all the blue-collar guys because saying, Oh, you guys need to learn how to code. Now all the coders are the ones that need to learn how to, you know, do something hands-on.

SPEAKER_01 28:07

Yeah. So but even even with the the coders now, they're able to code faster because they they understand Python, they understand a lot of the coding that's out there. So at the end of the day, you and I could probably vibe code something because we understand English. That's right. So you can prompt it to be able to understand what you're trying to build. But at the end of the day, what's the execution? Yeah, the execution is that you've now got to get that SaaS software live. So you need to be able to understand certain codes to be able to prompt the AI to be able to complete it.

SPEAKER_02 28:41

Yep, that's right.

SPEAKER_01 28:42

Otherwise, you've just got a fancy website that doesn't do anything.

SPEAKER_02 28:45

That doesn't do much, yeah. Yeah, but then in saying that, what is that doing? That's giving the everyday person who would have otherwise relied on the coder, it's giving them much more direct access. It's taking out the middleman, i.e., the traditional coder, and it's giving them more direct access because now obviously the AI can go off your layman terms to then translate into coding.

SPEAKER_01 29:04

So, like in in I was reading a report the other day, like with um NVIDIA, so Jan Strong, his AI coders, he's frustrated if he's paying them $500,000 a year on on average in in wages, and they're not using up to two to three hundred thousand dollars US in tokens. Oh wow, because they're KPIs now. So a coder's KPI is based on the amount of tokens that they spend using the large language models, using the using the vibe coding. So that enables them to be able to develop faster. Wow. So with a lot of the engineers that are out there right now, their KPIs is based on their token spend.

SPEAKER_02 29:45

Right. So then once again, I guess the coders are infected with that same dynamic of they're gonna all have to specialize in very, very specific fields as opposed to just being hired as a as a frameworks change.

SPEAKER_01 29:58

Just the frameworks change. Yeah, yeah. So you still gotta you still gotta understand code, but your your skill is based on how fast you can prompt the AI to build the code. And so that comes on your token spin because most engineers now have a budget based on their token spin.

SPEAKER_02 30:15

Yeah, yeah. Well, that's actually um, yeah. Well, you know what? Another good segue into actually total addressable market, because this is gonna actually play a role in we're gonna talk about automation once again. And um, yeah, over to you, Pete.

SPEAKER_01 30:28

So, Tam, we've we've already we've spoken about it. It's a $33 billion industry just in Australia loan, but it's highly fragmented. Uh, it's easy to start, but it's very, very hard to scale. If you're a one-man band, you're gonna struggle in the in this sector. You need to be able to bring on the partners, you need to be able to understand systems, you need to be able to understand AI to be able to scale a law firm. So, TAM for me, obviously it's gonna be a high score. It's a massive, it's a massive market. We're not gonna go, I'm not gonna go down the path of what we just sort of spoke about in re in relation to the intrinsic issues that they have, but purely based on on TAM, it's a high, it's a high score for me. Um, it's a seven, it's a seven out of ten because it's a $33 billion industry alone, just in Australia.

SPEAKER_02 31:16

Yep, yep, definitely. And look, everyone does need legal advice and we we rely on it for our we for our clients a lot to seek that independent legal advice as well. Um, I'll I'll go on the middle line, five out of ten, although I do agree, uh, but I would say also the total adjustment market could really be eclipsed by the potential innovations and the efficiencies that will come from obviously the the future AI integration and automation. But the big law firms are just gonna eat everybody's lines if they start implementing it.

SPEAKER_01 31:42

So all the one-man bands and everybody that are out there at the moment, the small law firms will basically just disappear.

SPEAKER_02 31:47

Yeah, all the old school lawyers that we've dealt with, you know, the guy he's been operating since.

SPEAKER_00 31:52

He's got a paralegal that's from another age, you know. You walk in there, you know, he's smoking on his table. He's just he's button, he's button out his cigar.

SPEAKER_01 32:01

He's got the bookshelf, he's got the bookshelf on the it's covered with books every encyclopedias that haven't been that haven't been used for the past 30 years. You talk to him about you know your file and he he brings out his little filing cabinet on the side, and your your file's right on the bottom of the pile.

SPEAKER_02 32:16

Yeah, yeah, exactly right.

SPEAKER_00 32:17

They're gonna disappear.

SPEAKER_02 32:18

Yep, yep, 100%. So basically, let's do the moat uh score. So we've got um uh six out of ten for me for margin and one out of ten um for margin for Pete. And then for operations, we've got one out of ten for me and two out of ten for Pete. So we what have we got? We've got nine in total, and then for advantage, it's one out of ten for me, and two out of ten for Pete. So it gives a total of three, and then for TAM, it's five out of ten for me, and seven out of ten for Pete. So we've got a total of 24 out of 40, which then translates into 60%.

SPEAKER_01 32:59

So, you know, not the just the pass.

SPEAKER_02 33:01

Yep, so just the pass because we know it's valuable, we know the barrier to entry is difficult, we know it's actually a very valuable asset to have. Of course it is. Um, but yeah, but effectively, just like anything, there it is due for some kind of disruption, some of it positive, but some of it could be to the detriment of the players.

SPEAKER_01 33:16

Yep. To get anything out of this today, use it, use your data.

SPEAKER_02 33:19

Yeah, 100%, 100%. And and and don't be shy of of you know changing your ways, don't be stuck in old ways. It's very hard to get out of it. It can feel I know I know a lot of people, especially from the older generations, you know, the more boomers, you know, they tend to see it as a threat, but you shouldn't see it as a threat because the reality is, yeah, that's the way the world turns.

SPEAKER_01 33:37

Definitely have to embrace it.

SPEAKER_02 33:38

Yep, they will. So, yeah, 60%. So, yeah, let's look at the $1 million test. So, can you earn a mill, or what do you need better yet, to um to earn a mill in the first year of operating a leak uh law practice? So we've got $3,500 on average per matter, 286 matters would be needed. So that's 1.2 per day. You need four people um in terms of their output, in terms of uh manpower. Um, so you know I I'd say it's it that's quite a challenging feat for the first year, especially because you're so reliant on actual referrants and clients and generating leads, as we know that's one of the biggest struggles in our industry as well. So, what's your thoughts?

SPEAKER_01 34:16

Look, I mean it comes back down to like solo, it's unlikely. Um, team, yeah, definitely possible. The verdict, it's possible, but it's very, very hard, which is why the partnership um analogy, which is what we're talking about before, which is why it works. You know, you bring on five, five to ten skilled partners, your business scales.

SPEAKER_02 34:32

Yep, definitely.

SPEAKER_01 34:34

So final verdict. Um look, it's not it's not passive, it's not easy. Um, you can't you can't automate a law practice and um go on holidays. Yeah, that's right. Um, you've got to be there. So it is very, very labor labor intensive. But which goes back down to the whole point of this podcast today, you know, if you can systemize the specialized domain expertise that they have and bring down some of the cost that they would typically charge, it's possible.

SPEAKER_02 35:08

Yep, yep, definitely. And look, I and I think look, it it's a it it just because it is low capital doesn't necessarily mean it's low risk, um, because the risk does come in time, people and trust. And honestly, I I'm I don't know about you, Pete, but I'm seeing a lot of similarities in service-based roles in general, right? Like, for example, I know people that can run a let's just say a burger franchise, right? They don't have to necessarily have a passion for burgers because they're just selling the item itself, right? You're kind of selling a product, you don't really have to pitch it, people are buying it, and that's it, right? But when it comes to service-based role, you have to have a lot of passion. And I don't say passion in like, you know, a cheesy, romanticized way. I mean genuinely you've got to have the passion to the point where when you know shit hits the fan, you are prepared to keep going. If you don't like industry or if you're entering a service-based industry just for the money, you're gonna hate it because until you're able to deliver for all the challenges, you're not gonna get paid. So I think there's a big difference now, with you know, a big contrast between this year.

SPEAKER_01 36:04

It's it's a people-based business. Because at the end of the day, when when you employ a lawyer, you're not employing a chatbot, you're employing a lawyer. What technology they use in the background is irrelevant as long as the business or the person or the individual gets the outcome that they're trying to achieve. Correct. How they achieve that is irrelevant. Yeah. But it's a people-based business. Yeah. Unfortunately, with people-based businesses, it's very, very difficult to scale because you need a multi a multitude of people to be able to generate additional or or or high leverage. Um, one, you become high leverage, but two, to be able to justify that is based on the amount of partners that you have within your firm.

SPEAKER_00 36:45

Correct.

SPEAKER_02 36:45

Yep. So that's about it. Um, we know that uh, you know, the disruptions are coming into that industry as they're coming to most. Um, and you know, the jury's out, no pun intended, as to where it will land. Oh, hey, we're pretty good with the pickle pun. So yeah, yeah, yeah, there you go. I knew that.

SPEAKER_01 37:03

I knew there had to be a pickle pun in there somewhere because we just love our puns in this business. Yeah. So if they want to get a big deal, yeah, they need to be sort of throwing out another one. I thought you I can't just allow you to have the problem. Yeah, you can't. Yeah. Um, if they want to do the big deals, they've got to be able to scale. You can't scale individuals, you can scale people if you automate the processes and you understand the data, which is what a lot of law firms have, and that's where the value is. The value is within data. It's how can you utilize that data and how can you bring that to the forefront of technology? Yeah. So the close run the numbers because we do not want you to get into a pickle. Starting off.

SPEAKER_02 37:50

Yep.

SPEAKER_01 37:51

Thank you, sir.