Episode 018: Season 2, the $75 Consult and the Frankenstein Stack

Welcome to AI Tools for Practicing Lawyers. Practical, no nonsense guidance to help attorneys put AI power to work in their practice right now. This past February, when we launched the podcast, the biggest question lawyers were asking was, Should I even be using AI? Because I I'm a little afraid of it, to be honest with you.

Five months later, that's no longer the conversation. Heather recently attended the Maryland Legal Summit and she was doing a panel. And when the audience was asked who had used AI in their practice, almost every hand went up. And that that changes everything because what we used to think of as a Flintstones lawyer or a Simpsons lawyer or Jetson's lawyer is changing as AI becomes more widely adopted.

So welcome to season two of AI Tools for Practicing Lawyers. Season one was about learning AI. Season two is about something really that's going to be harder. It's going to be about teaching AI your firm so you can really optimize its use.

And today is the first step. Welcome to AI Tools for Practicing Lawyers. I'm Ron Drescher. I'm Heather Gardner.

And today we're joined by someone many bankruptcy lawyers have known for years. Jen Grondall Lee, founder of Lawyers Success Network. Jen has spent years helping lawyers build better bankruptcy practices, improve systems, rethink pricing. That's a dangerous subject in the bankruptcy world and innovate inside their firms.

And today we're going to talk about where legal practice is headed in the AI era. Jen, I am really genuinely thrilled to have you on the show to launch our second season. Why didn't she just tell our listeners a little bit about yourself? Sure.

Thank you, Ron, and thank you, Heather, for having me here because I do love my technology and I do love bankruptcy. So I am a bankruptcy lawyer, total bankruptcy nerd. Um, I really love the practice of bankruptcy, but I like the practical side and the education of what we do with debtors. And so I consult with law firms on fixing their workflow systems, charging what they're actually worth.

I know that's a crazy thing in the bankruptcy world, and just helping overall practices run smoother and be more efficient. And so AI comes into play these days because there are so many things you can do with that. So I had a lot of fun doing it. Of course, I prepared for this episode by brainstorming with ChatGPT.

But I want to ask you a different question that that never even came up. Why is there so much tension in the bankruptcy world about pricing and fees? I think there is a lot, a lot of it it comes down to our clients don't have a lot of money, technically speaking. I always say that our clients have plenty of money.

They have a resource allocation problem. But I think some of it is that's the mindset of how it's always been, which bleeds over into technology and AI. That's how it's always been done. You have to be a struggling bankruptcy lawyer because you represent struggling people.

And so I fight that mindset quite often because what we do is very valuable. And also our clients can afford good legal services if positioned properly and explained properly to them. So how much of that anxiety about fees comes from courts and the US trustee? A lot of it does come from courts and the US trustee.

I think it's we're undervalued in the profession. I don't think people exactly understand all of the hand holding that debtor attorneys do when it comes to getting our clients through what the courts probably see as a simple process. Our clients don't see it as simple. And the good bankruptcy attorneys out there make it easy on the courts to see it as simple.

They don't see what's behind the scenes. And so I do think that there is a lot of scrutiny on it. There's also a lot of bad actors, not a lot. There are bad actors out there.

And so I think that brings scrutiny on those of us who are doing a great job for our clients because they're trying to make sure that people aren't getting scammed and there's not predatory services and that kind of thing out there. So it's kind of a catch-22 sometimes. Heather, you spend a fair amount of time with lawyers who are somewhat new to the bankruptcy world. What kind of conversations do you have with them about fees?

I honestly do have attorneys ask me about setting fees when they're new to bankruptcy pretty regularly. And I always tell them my best suggestion to them is to talk to other bankruptcy attorneys that are more established in their area, uh, their local bankruptcy bar association, should they have one, or if their state bar has a bankruptcy section such as ours, those are great resources, as well as just doing your own research, but that they have to set their fees themselves, of course.

All right. So let's talk about a topic that I know, Jen, is one of your favorite topics. You've probably talked about this 10 million times, but you have been a very powerful advocate for dispensing, doing away with the fear-driven practice of giving free consults to bankruptcy clients. Now, how did you start doing that and how has that worked for you?

So, how I started doing it was I was working with a marketing coach a few, like this was 15 years ago. And one of the things he got me to think about is what impact does a client have if they come see you earlier in the process rather than when it's too late? How do you impact their trajectory basically? And he's like, isn't that a valuable service that you're offering if you can create better without them having to go through a lot of stress?

So he's like, How about you just charge $75? Try just $75 for a consult and see how it goes. And so I started out at $75 for a consult and set it up as I'm giving you valuable advice. It's not selling you bankruptcy.

I may tell you that bankruptcy is not a great option for you, but isn't that valuable to know as well? And then within a couple of months, I was so slammed with work that I went up to $150 and then I went up to $200 for my consults, which is where I ended it as I transitioned to consulting. But what I found was the more emphasis I put on the value of educating the client, the more money I can save them in the long run. And so that's a lot of what the value of a not free consult does.

I also tell clients that if someone is giving you a free consult, you are the product. It's a sales pitch. That's all it is. And so if you want to go to free consult or free consult, they're just going to sell you on bankruptcy because that's what they're selling.

So if bankruptcy attorneys are going to roll their eyes at me and like, no, no, no, I do whatever's best for my client. But that person doesn't become a client unless you can convince them to file for bankruptcy that you handle with your firm. And so I found over the years that more clients came back and said, the $200 or the $400 I spent on your services saved me hundreds of thousands of dollars in the long run. It was the best $400 I ever spent.

So that's fantastic. I mean, that's the kind of feedback we all need when we are trying to do something new or different or unusual. But you just said something that tickled my ear when you said transitioning to consulting, do you still maintain an active bankruptcy practice? I do not.

Um, I found that I really like the education and helping lawyers and teaching bankruptcy as a total nerd that I don't take on individual clients anymore. My clients are all law firms these days. Well, all right. So how many law firms do you have working with you?

If you're if if that's something you want to talk about? Yeah, so it it depends. So I have some that are like retainer clients where I work with them every month on something. I have some that were in a project base where we're doing specific projects for them.

I have some that just do one-off consulting. They want a bankruptcy nerd in their pocket. So they schedule appointment with me whenever they have some strategic things to look at. So there are about 30 bankruptcy firms in my the different stages of different things.

I run a membership community called Bankruptcy Toolbox, which has a lot of FAQs they can use on their website, um, forms, templates, daily marketing articles for them to use. So it's just a different group of people in each of those buckets. I have to confess that when I was much more active in promoting complete bankruptcy and team accelerator, I would follow you and I would say, What's what's what's Jen doing? What's Jen doing?

So that's a compliment. So it's quite glad to hear that. Some people don't. Some people don't like to follow me because I'm always putting out crazy ideas.

But yeah. I love your crazy ideas. The disruptor. You you talked about AI in a recent blog post.

Uh, what is it? AIDR? What was that blog post all about? I'm gonna put a link to that in our show notes so that listeners can check it out.

What were you trying to say during that blog post? So it was kind of a play on the TLDR too long, didn't read uh term out there. And what I get frustrated by is when I see content that is obviously AI generated. Like I get an email and it's the one-line sentences with the emojis in front that looks with the, you know, to tell the truth or to be honest with you, kind of language in there that, you know, chat likes to throw in.

And so whenever I get those, I'm like, I'm not even gonna read this. You've already lost me with your copy pasta that you've created out of this. And so it was kind of getting to the fact that you can use AI and you can, it's a huge time saver. It saves thousands of hours a year for me.

But don't let it think or be you. Like you have to create your own authentic way of doing things and not what chat spits out the first time. But you know, all right, I'm gonna, I'm gonna push back against that. Yeah.

Right. Because the law, the more you use AI, the more it learns about you, the more it learns about the way you think and the way you approach problems and issues. So it's going to adapt what it has learned about your thought process in the in the content that it creates for you. What's the problem with that?

Nothing. That's perfect. That's what you want it to do. You want it to train it to sound like you.

And so if I get something that is like copy-pasted with the typical, you know, LinkedIn one-line sentences all the way down the page, then I'm like, well, is this really you or is this AI? But if I get some that article was written by ChatGPT. The whole article that I that you it was written by ChatGPT. But I went, but it chat knows I've told it, don't use M-dashes, don't make one-line sentences, don't put stuff in there like to tell the truth and the problem really is like those leaders that chat uses.

Yeah. To be honest, like all those giveaways. I'll tell you this, like I'll say this. Like well, my favorite chat GPT thing is that's the most amazing thing you thought of during this entire process.

And it will say that same thing to me 10 different ways after 10 different ideas, which is unfortunate, actually, because we become collaborative with the AI environments. Have you found that? Heather named her her Chat GPT. What's the name of your Chat GPT again, Heather?

Bosley. It's Bosley. Bosley. Bosley.

Bosley. Yep. Bosley was a machine that just taught Bosley was the Charlie's Angels, you know, guy on the answering mission, or the the speaker that just sat there, but you never saw it. It was a machine to me in my youth.

So Bosley is my machine that I don't have a face for, but is so helpful, right? But it's really true what you said, Ron, that it just wants it tells you how great things are when it's not great. So you have to train it to say, don't constantly tell me something's a great idea if it's a horrible idea. It just wants to make you happy, which is sounds funny from a robot that it just wants to make you happy.

But that's basically what chat has been designed to do is to affirm whatever you've said. And Ron made a really good point too. He said, What's wrong with that? But you had just said if it starts with something I know is AI generated, then it clearly doesn't sound like that person.

If you can tell in the first sentence that AI wrote it, the AI does not sound like you. You can create it to sound like you at some point. But yeah, mine is true. Mine doesn't use any of those phrases because I've told it specifically to never use any of those phrases.

It is completely, you know, it's very specific. I've got tons of rules for mine, and that's one of them. That's a long list of them, actually. I had a proposed title for uh an episode, and uh brainstorm it with ChatGPT, and we came up with a bunch of different variations, and I wasn't happy with it.

I went over to Claude, asked the same questions. Claude smacked me. Claude said, What are you, what are you doing that for? You're just gonna you're just gonna create a conflict within your title, and you're it's not gonna serve either of the two ideas that you want.

And I said, Wow, that's really interesting. I'm going back to ChatGPT. But they're fun to play off each other, though. I always use both together.

I use both together every day. I do it all the time. And sometimes you really do get some interesting things about it. So, Jen, what levels, what tiers of AI tools do you have that you're using?

I mostly use the pro versions of both Claude and ChatGBT. I use the paid versions of them. I don't have like developer grade, I don't have enterprise grade. I'm just using the pro versions of them for both Claude and ChatGPT.

In your consulting work, are you helping firms adopt AI protocols? Yes. My biggest complaint, though, about AI is everyone wants to just plug it in and not think about it. And really the the process should be you should think about what your workflow, what you want it to be and design it, and then figure out what tools fit in there.

There's so much franken stacked technology and subscriptions out there that it's like this hodgepodge of mess when it comes to AI. So we step back, look at what we want their system to actually do, and then pick the tools that automate within the firm. So yeah. I I have to say that's becoming uh a widely accepted approach to bringing AI into the law firm, which is don't think about the tool, think about the pain points, and we'll talk about how to create uh how to locate the tool that's right for you.

Yeah. Do you ever recommend any verticals for for your uh your consulting law firms, or do you mainly stick with the frontier models? I mostly stick with the frontier models. I like, I don't like a lot of subscriptions.

Practice management sophomore drives me crazy because none of them do what I want them to do. And so there is a lot. I've created a couple of chat of chapter 13 plan creation tools using Claude Code that live locally. And so things like that, I've gotten away from like the, you know, branded legal management tools out there and showed firms how to create their own internal ones that work better and are customized to them.

So yeah, do you share those that you've created with your clients that you consult with ever? I do a lot of times I'm showing them how to do it because, as you know, there's so many different districts for bankruptcy, and I want them to be able to update. Like when the Northern District of California changes their chapter 13 plan, I want them to be able to go in and update on their local one. And I'm not going to run 110 different districts.

Yeah. So I show them how to do it so that they don't have to keep paying a subscription fee. Even better. It's like, you know, feed them or teach them to fish.

Yeah. Yeah. We're seeing this all over the place. We're seeing all these different, you know, learn Claude in four hours and come do 32 agents.

What do you what do you think about those uh offerings? I mean, they're they're fine if people want to like have an overview and that and learning Claude. I think people should at least have a basic, you know, how to do prompts. I don't do the whole prompt engineering thing.

I have created my own over the years of like how I just talk to Claude and Chad GPT. I think that firms would be better off working one-to-one with someone who's actually going to identify the things they need to know. There's so much overwhelm out there and so many things you don't really need, but you want to sign up for everything. And so I really like the style that I have, of course, because I do it, of working one-to-one and figuring out what they need to know and finding the right either tool, course, whatever they need, instead of here's a huge four-hour overview that first of all, you don't have time for.

You have too many cases that you're working on. And second, three hours of it is stuff that you don't need to know anyway. So do you also? I'm curious because this is something that people often ask me about these AI courses, Ron, because we see them all the time.

People talk to us that are listening to the podcast. I tell everyone, start within your own AI. Claude has Skilljar, which you can learn probably the most from anthropic about its product. When you go to someone else, you're just, they're just giving you different, they're just, they're just basically giving it to you in a different context or in a different order, or they're picking out what they think might be the most important parts.

But you can go into Claude's skill jar and you can you can look at everything or anything you want. You can search. It's it's very helpful. Same with same with chat GPT.

Open AI has its own courses you can take for free. And that's why I tell like everyone I asked it, like, well, can chat do this or Claude do that? I'm like, ask it. That's all you have to do is ask it, can you do this?

And it'll tell you exactly how to do it. It'll hyperlink it for you. It'll say, here, go to Joe, go here, and this will take you to where you learn about this. Exactly.

You know, before we started recording today, Heather reported feedback from a podcast listener that enough with the Flintstone Simpsons Jetsons, you know, you don't have to do it every time. That's the first time we've heard any kind of feedback uh in a negative way about it. And I certainly want that feedback because we want this podcast to add as much value as we can. But Jen, do you feel like we need to adjust our presentation style and content based upon where lawyers are when we meet them?

I do, to some extent, because it's hard to introduce something that's too advanced to someone who's still a Flintstone era. And I struggle with that a lot because a lot of law firm owners have been doing it for many, many years. They're very resistant to any kind of change. But then you have this newer generation of lawyers who are wanting to be Detson style.

And it's this conflict within firms. And so I do find that going back to basics sometimes and showing the owners of the firms how much money they're losing by not looking at some of these efficiencies and automations is really the way to go. So I do think speaking to different levels is important. I think podcast hosts these days kind of have to pick their niche.

Like where do they want to be and who are they talking to as a wide or narrow segment of the population? So yeah. There's another theme that we've talked about from time to time, which is the concept that should Flintstone's lawyers only try to work with Flintstone's clients, and should Jetson's lawyers only try to work with Jetson's clients? Or can we have that kind of like cross-breeding, for lack of a better word, in tech levels between attorney and so one thing that I often tell bankruptcy lawyers is you don't have to sell to everyone.

You need to sell to the type of client you want. And so I do have some Flintstone lawyers who don't want to learn the new tech and they don't need to because they want to help the people who aren't using technology. I also have the other side of it. If you can't make an online appointment and you can't upload stuff to a portal, I don't want to work with you, as how some lawyers look at it.

So it's a really good tool for getting clients that you want into your firm to focus on the people that you want to help with their technology, where they're at. So I do think that Flintstone lawyers don't have to work with Jetsing clients because they're they're at odds and it's not your ideal. You'll be miserable trying to help people that aren't on your same wavelength. So everybody everybody will be miserable.

But don't you think it happens all the time inadvertently? Yeah. Inadvertently because so often it happens because you don't get you're not going to figure that out right away often. So that's where you have to build up.

So I do a lot of client experience building. And so part of your client experience. And onboarding and lead generation needs to be speaking the language that the ideal client you want is using. And so if your website says we work virtually, we only do things via online appointments, we pay online, you are intentionally narrowing that people who are going to work with you.

Lawyers get scared of that. They're like, no, no, no. I don't want anyone, I don't want to narrow anything. Right.

I'm like, you're thinking about it wrong because you're going to make yourself miserable with the clients that come into your funnel. So yeah. And they do. I experience this all the time.

I work with attorneys that do that and all the time. All the time. You know, one of the things with marketing that so many business people get wrong is they feel like they do need to appeal to everyone. And there is a there is a certain anxiety that business owners have that the phone is going to stop ringing or the email is going to go dark.

And where am I going to be? Have you seen like the opposite as true? That the more you narrow, the more success you find. Yes.

It started with charging for consults. I'm was the only one in my area doing it. I was way busier than the attorneys who were doing free consults. Um, so again, being different and people who were willing to pay were willing to hire me then.

And then when it comes to your ideal client, I loved as a practicing attorney working with small business owner, primarily women business-owned businesses. We did a lot of chapter 13 sole proprietor work. And as soon as I put that out there that that was my ideal client, I had the floodgates because people identified then with my exact ideal client. If you say, hey, I need a bankruptcy attorney, most people don't even know what bankruptcy consists of and they don't know what to ask for it.

And so when you start putting out the specific things, the floodgates open versus shutting. But it's hard to convince rational business owners that that's how it works. I I wrote a book, a little, a little marketing book once upon a time called File Bankruptcy and Get Rich. Which was which was a great lead magnet.

It you know, everybody was interested in it and it was very successful. But then I shifted. I did another little book called Financial Recovery for Single Moms. And that that blew up.

That blew up. Because I went back and I said, well, wait a second, who are my main clients? And at that time I had more single moms as clients than any other particular category. So I decided to kind of lean into that, and it really did work.

Yep, it sure does. Because people then think of you. They, you're not a lumped in a mass of 20 other attorneys in their mind. You are very specific and they know exactly when to send people to you.

And everybody wants to feel seen. Everybody wants to feel heard. And the more we can do that as practicing lawyers and business people with our clients and prospective clients, the more success we're going to have. All right.

So, Jen, what are you working on these days? I mean, I know you're always working on something. Yes. Um, I'm always working on educational uh courses.

So I do have a course for lawyers who want to add bankruptcy called building a bankruptcy practice from start to finish. It takes them from like very first steps through chapter seven, chapter 13, initial consults, marketing, uh, all the fun things that go into building a bankruptcy practice. So I'm always adding onto that and creating more content as things change. We're adding an AI and automation module to that.

Um, and so those are big things I'm working on. And then I did launch bankruptcy toolbox earlier this year, which is a community membership site with free courses included. And um there are some weekly and monthly events in there. Stump the bankruptcy nerd is in there.

We do hypo hours where people bring their crazy bankruptcy questions and we talk about them. Um, so things like that. I'm always looking for what the questions are that bankruptcy attorneys have and how I can help fill that need. And then also, is there an AI tool that would improve the client experience and the firm experience in the meantime?

So yeah. How do you expect to add the AI function into your educational materials? So a lot of the AI functions already built into my platform that I'm in. It it creates my funnels.

It does a lot of the one-off things that I used to hire a virtual assistant for. Sorry, virtual assistants out there, but now I have my AI assistant that can go through and build things out for me. Um, but also just keeping abreast of things that are, it changes so fast. And state bars are coming out with competency requirements for AI.

So building that into everything we do is kind of like, okay, here's what you probably should know right now and not overwhelm people at the same time. It's really hard to balance that. So that's how I'm building into things. Did you read about the uh there's a judge in, I think in New York, but don't quote me on that, who issued a uh three-part order for AI.

You have to disclose if you used AI, you have to certify that you you checked all the citations, but then he must have a typo in in section three. He says, You've got to certify that this document was not produced by AI. Yeah. Have you seen that?

And how do you make sense of that? So I think a lot of courts, like so in San Diego, uh, Southern District of California also came out with an interesting order. What I think courts are doing is overreacting slightly. It's exposing bad lawyering.

It's not exposing bad AI. Like the tool itself isn't the problem. It's lawyers who aren't checking the like who files a brief and doesn't know the underlying cases they cited, bad lawyers. And so I think that these rules were already there.

Yeah, right. Like if you had done that. Right. Well, over a thousand to date, over a thousand to date.

I follow all of them. Yeah. But what I've seen is some of these lawyers who are now getting sanctioned for this, if you go back and look at their history, they've had problems all along in their careers with these types of issues. It's just AI is now the, you know, buzzword of the day for it.

So I think the courts are slightly overreacting. We all are supposed to check our sites. We're all supposed to, you know, certify that what we filed is truthful and accurate, that kind of thing. Um, plus, they don't define what AI is.

So if I used Grammarly or if I used spell check, is that considered AI? Or are they talking about large language models like or something like that? It was the development of large language models that really turned the AI world on its head. And and that's what's created that vocabulary confusion.

Uh, it's a great though. She put puts up a great uh point though, that it is human error, even in the AI. It's lack of human oversight and judgment. And that's interesting that if you look up some of those same attorneys that have been sanctioned, that they might have been sanctioned for other ethics or irresponsible filings in the past that had nothing to do with AI.

See, Heather, this gets us, this gets us back to our ongoing debate, which frankly I think is really valuable for listeners. I hope they find it valuable about the extent to which AI and using it saves time. Of course it saves time. Of course it does.

But my feeling is that it's primarily going to be used to elevate your work product as opposed to saving time, because you do have to go back and look at what it did and double check it. What do you where where are you on this uh uh spectrum, Jen? Save save time or elevate practice, or a little bit of both? I think it's a little bit of both.

I estimated last year in 2025 that it saved me about a thousand hours of time. And that is because I'm a power user where I'm using it as thought, like a collaboration partner. I use it while I drive. So I talk to Chat GPT while I drive and come up with like outlines for things and things like that.

I'm not using it necessarily for final drafts. I often use it for like white page syndrome where I'm staring at a blank page and I need something like in front of me. So I will say it saves me a lot of time when it comes to like getting words on a paper that then I can then edit. So yes, you have to oversee it, but I do think it saves a lot of time used appropriately.

It shouldn't be used for final drafts. It shouldn't be used for citing cases unless you've checked them all, that kind of thing. Right. The the safest, I told uh a lot of attorneys in that that's always the most asked question when you're at a panel is about hallucinated, you know, hallucinated citations.

And a very easy fix for that is to get whatever AI you're using to not add the citations. Give it the citations you're going to use with the correct language and tell it not to change this. This is what you include. Then do the rest of it, right?

Draft it around my site, the citations I've already found that are absolutely relevant and exist for this case. And then make sure it doesn't change it because I have seen it try to change. Tell it to change it. Tell it not to change it and not to change those things.

And then you still review it. Sadly enough, you could do all of your legal research the old-fashioned way, you know, create PDFs of every case, put it into your case file, uh, and then tell whatever AI tool you're using, only use cases and authorities from within that folder. And it may do that, but it will still hallucinate from within those cases. So there's no getting around it.

It's just, it's just the way it is. And I think that uh as we go forward, maybe someday they'll come up with uh a large language model that doesn't hallucinate. I don't think it's gonna be in 2026. I totally agree with you on that.

All right. Yeah. All right. So so let's now that we're talking about not having to have employees do certain functions because we have AI to do it, uh, let's talk about a practice signal for today.

Uh and this one, most of them come from Reddit because people are willing to because they're anonymous. They're willing, they're willing to show how crazy they are. So today's practice signal comes from a lawyer trying to hire a receptionist, uh a legal assistant. Uh the candidate interviewed beautifully, 15 years at a law firm, great interview.

Within a week, she was fired. Now the lawyer is asking, how do I better test applicants before hiring? What do you think, Jen? How did if if one of your lawyer clients came to you with this problem, how would you advise them?

So my first question would be is what kind of training did you provide? Because a lot of times what I find is law firms lack systems and procedures, and you are going to get failure and you hire somebody and then you don't give them systems and procedures to follow. But then also, what did they test for? Did they, you know, check phone skills, whatever they're looking for, whatever's most important, a law firm needs to actually check those skills, whether that's how they addressed in an interview, how they are, you know, on the phone, if typing is important, that kind of thing.

I think it's kind of gone by the wayside. There used to be a lot of these. When I applied for a job, you know, back in 1998 or whatever it was with the 10th agency at the time, I had to take a typing test right there when I applied for it. So I think people have gotten away from that.

And it's really easy to fake some of those things these days. It's really easy to fake. I've seen hackers that are faking like interviews, AI is showing up for things. So I do think you have to be careful.

But also, I think some of it falls back on the law firm because I've seen too many firms that have this is how we've always done it. And we have the last person train the next person, and nobody actually has documented anything for how things are done. So, Heather, between the three of us, you're the one that's probably doing the most hiring and firing. Have you updated your systems and protocols regarding that?

No, but I think that that story in itself has a lot of red flags. For instance, if you fired someone within a week, there was something far more severe than their skills usually happening. You couldn't have even gotten started in a week. You know, you in a law firm, they they probably they should have been just shadowing someone for the first week, especially if they were a new legal secretary.

So they would have just been learning the firm systems and not probably even using them that quickly. I'm surprised and would say within a week, you know, for me, that sounds like the person was maybe not showing up. Uh, and and I do. I hire every quarter.

I'm in a hiring round right now. I'm about to hire two people we've already made a decision for. And it it is always very stressful at first two or three months when you hire someone, because you have to watch them more and sure that they're doing the job, their the job and that they are capable of doing the job that they say that they can. Because even if they did, and I vet people so much.

I talk to every law firm they've ever worked for, you know, not just the two glowing references they give me. So good. Once I do that, I often still it doesn't work out, you know, not often, but sometimes it's still not gonna work out. It's a different environment, it's a different company, it's different, you know, attorneys that you're working with.

It's so many things at play. But but this attorney, if if they fired someone within a week, I mean, first of all, it does happen. You can hire someone that seems perfect on paper and in the interview. And once you try to get them, yes, it happens to all of us.

I've done it. But it within a week at a firm specifically does seem like there was something more going on than the skill set. So the the question ultimate was how do I better test applicants before hiring? And of course, ChatGPT had a recommendation.

ChatGPT says, don't ask, do you know Microsoft Word? Instead, say, here's an unhappy client email. Use whatever AI tool you'd like, draft a response, then watch. Do they improve the AI or do they simply paste it?

You know, in the way that Jen was complaining about in her blog post. That will tell you far more about the future employee. What do you think, Jen? Is that a legitimate approach?

I mean, I think yes, because I think experiential, like seeing how they actually handle it. And we do interviews for like management and things like that. We're always asking, okay, how did you handle a specific situation or can you tell me how you would handle it? I think that tells more about people than yes, I'm experienced with Microsoft Word.

And I do want them using AI tools. And so I think that's a good plug to put in there and say, I don't care what tools you use, but create something. I would also check to see, I do like the comment that you had in there about uh if they updated or they changed. Like, do they pay attention?

I've seen some horrifically generated letters recently that were not AI generated. That I'm like, does your firm know you're sending these letters out like this? And so I think it's a good way to test is to ask them to actually do something that you would want them to do in practice. Um, in within reason, I have seen some crazy firms like try to give lawyers like full-blown, you know, tests for briefs and things like that.

And it's like free work, basically. I wouldn't go that far. But I do think it's okay to ask them to, hey, how would you do this or walk me through your process? Heather, what do you think about that?

Would would you give that kind of AI test to a prospective employee? No, but that's I have such a different model and needs. It's very different. You know, we don't we don't have bankruptcy paralegals utilizing AI for anything.

It's, you know, a different monster. I've been wanting to ask you this question because I know that you tell your team of paralegals not to use AI, but what if the law firm said, this is our system, we need you to use this AI. Well, that's different. That's different.

Of course, we accommodate what the firm's doing, and then that's under the attorney's supervision. And to be clear, we do I do have them utilize some AI. They utilize Gemini for their spreadsheet help and their weekly status reports to attorneys and their weekly follow-up emails to debtors who, you know, are lacking things. Those are just very easy, time-saving tasks that they can do that we have processes around.

But we do have firms, for instance, utilizing Glade AI for bankruptcy. So we have paralegals working in that system. Best Case Now has an AI document collector. So the paralegals working with best case firms that are utilizing that are using it.

Uh, and and that is absolutely going to happen. But what I want to make sure that they're not doing is ever, of course, uploading any sort of client information into an open source AI tool. So that's why our policy is as stringent as it is. Sure.

I I think it has to be until you know you can control the configuration of the system. Yeah. All right. So, like I said at the beginning of our of our show, this is season two, uh, in large part because when we started this, I knew really nothing about AI.

I was a user, I was excited about it, and I've learned a lot about AI. In this is our fifth-month anniversary of starting the podcast. And now I'm really seeing that the Flintstone Simpsons Jetsons framework needs to evolve. And so we're not going to be offering tips on every episode, perhaps to the happiness of the lawyer that complains to Heather at the uh Maryland Legal Summit.

But what we are gonna be doing, we're gonna maintain the framework and we're gonna start doing a multi-episode, not training's too strong a word, but kind of a rethinking about AI and the Flintstone Simpsons Jetsons framework. So we're gonna try to build. We're not asking you to buy anything, but we want you to start organizing the knowledge you already are. So all right, on the Flintstones level, whatever your primary practice area is bankruptcy, family law, estate planning, immigration, whatever it is, find your 10 best forms.

The forms you're using a lot, the forms you've created, you've worked through, find them, put them in a folder somewhere. We'll get back to that next time. But that's your that's your mission for this week, is just to identify your 10 best forms. Now, if you're uh if you're on the Simpsons level, and don't forget, we're we're still self-sorting.

You decide where you are, and if you decide you want to be on Flintstone's level, that's where you are. Simpsons, make a list of 10 questions your clients ask over and over. Uh and this isn't gonna be legal trivia, but these are the real client questions. Uh, will I lose my house?

You know, uh how much child support am I gonna have to pay? Am I gonna be deported uh because of this problem? Uh, what's the likelihood that I'm going to be sentenced for this crime? Et cetera, et cetera.

All just find the 10 questions your clients ask over and over. Put it in a Microsoft Word document, I put it aside. If you're on the Jetsons level, I want you to identify the 10 most important firm systems. Is it intake?

Is it onboarding? Is it discovery if you're a litigator, you know, going through a discovery process? Uh maybe these are HR systems. Uh, maybe we're talking about systems for getting clients in to sign their wills, whatever the 10 most important firm systems are.

And then we'll get back to that and we'll move forward next week. We don't want you to redesign anything yet. Uh we'll talk about the next step in turning those systems into AI assets. Jen, help me out.

What what else, what else do I need to tell these lawyers? I think those are good places to start because it's overwhelming to start. It's overwhelming to figure out where to start. And those are really good practical ways that each of those functions can um get started with some of the AI and automation that needs to be done.

I feel like season two is going to be a sort of uh law firm AI stack consulting session built out over time. So it's really useful series though. I think because this is where AI is. You know, two months ago, I was all about, well, the prompt has to be great.

But the more you get through what's going on with AI, you know, Claude for AI came out. It's not for people in our milieu, but you got to start somewhere. And I do think eventually Claude and Harvey and Lagora will uh produce products for smaller and solo law firms. I believe it.

It's too big a market for them to ignore. Before we close, Jen, I'm gonna ask when we have a great guest, I always have to ask, what was a recent wow moment that you had, especially using AI? A recent wow moment was when I was going through and totaling up the number of hours. I had chat help me figure out how many hours I was saving.

And it was super scientific. I mean, it was I was going through all the things I'd asked it in the past year and the documents I put together. But that was a pretty wow moment figuring out that I got a thousand hours more of work, of product done in the last year than I would have gotten had I had to try to work all those hours. So that was a pretty good wow moment for me.

All right. Fantastic. All right. Well, Jen, thank you so much for being here.

Uh it was, I think I learned a lot. It was a great uh episode, and we hope to see you again soon. And best of luck with your um your uh it's like a Patreon. What's it called again?

Your bankruptcy toolbox. Yeah. Your bank membership. Are you still doing Rebel Roundtable?

I do. Friday mornings I have Rebel Round Table. Yeah. Wonderful.

All right, Jen. Thank you so much. And Heather, we'll see you again next time. Bye.

Thanks, Jen. This is AI. This is AI Tools for Practicing Lawyers. See you next time.

That's it for today's episode of AI Tools for Practicing Lawyers. Thanks for listening. We'll see you next time.