Ready or Not? How to actually embed legal AI

Okay. We are live. Hello, everyone that is coming in and joining as we go. We are just waiting for Christopher to come in, but thanks very much for joining our first ever Nomio webinar. I’m joined by Chris Fallon, who’s associate GC at Safelite. We’ve got Rob Jarrett, and we’ve got Nathan. We’re hosting this webinar on Zoom, so there should be a q and a section. As we go through the conversation, please keep your questions coming in.

We’ve got five, ten minutes at the end to answer as many of the questions as we possibly can do. And if we can’t get to, we’ll send an email afterwards, with our kind of follow ups from there. So what we’re gonna do for today is we’re gonna start with a quick round of instructions, and then we’re gonna get right into it. This webinar has come on the back of a report we’ve rich... Recently written called ready or not.

This bit of research that Rob and I put together when we interviewed about 25 heads of legal and general counsels, and the results results were pretty interesting. One of the things that really stood out to me is actually Wordsmith came up a lot as a tool people are looking at and a tool people are using. And so that’s why we’ve got Nathan on and we’ve got Christopher joining as well and talking a bit about that experience.

But with that in mind, Rob, I’ll hand to you if you wanna do a quick introduction and then over to Nathan. Yeah. Thanks very much. So I’m Rob Jarrett, founder of size seven. We provide fractional legal support to scaling tech businesses and lean in house legal teams.

Nathan? Hi, everyone. I’m Nathan. I’m a legal engineer here at Wordsmith, a corporate lawyer by background and a parallel past in the legal tech space. Great. And I’m Ed. I’m gonna be facilitating this conversation. While we wait for Christopher just to get on, Rob, I’ll start with with you. We wrote Ready or Not over the summer. And the headline stats for for those that haven’t had a chance to read it, we’ll we’ll send it in the follow-up afterwards where the pretty much every in house team we spoke to is using AI in some capacity, whether that’s Copilot right the way through to some of the bigger end to end tools.

The vast majority of people we spoke to were working longer hours, and actually, three quarters couldn’t even find all of their signed contracts in one place. So there was this theme of our in house teams running before they can walk when it comes to AI adoption. What did you make of the findings, Rob, as you went through it? Were you surprised? Well, in relation to this... The stats first, I think it was 93% are using AI in some capacity.

That itself is not particularly surprising because as you say, it can be anything from... And I’ve I’ve heard this firsthand using Copilot to help write an email through to, you know, fully automated workflows, agents, etcetera. And whether it’s because people want to or whether they’re told to or whether they just simply can’t avoid it, like, people are are using an AI. So that really high usage doesn’t surprise me, but as I say, it doesn’t necessarily reflect the the actual adoption or, you know, day to day usage.

30... I think it was 38% are not working longer... Sorry, working longer hours than before or as many hours as before. But, basically, most people are working as many hours. Again, that doesn’t... I actually thought it would be more. From what I’ve seen, what I’ve what I’ve experienced, what I’ve read. It’s a bit like... I’m gonna sound really old, but, like, when you’ve gone from, like, post to email or desktop to laptop to mobile, things just become a lot quicker, a lot more efficient, and a lot more gets done.

And just taking contracts as an example, like, if someone’s using AI in a contract negotiation, their bit gets done quicker. It goes to the other side quicker. If they’re using AI, it comes back quicker, and that’s certainly what I’m experiencing. And then businesses just do more and more deals. More and more gets done. Or you’re you’re taking away some of the lower level work, and that’s just being replaced by more strategic, more complex work or prod...

Certain projects that people didn’t have the capacity for before. So it doesn’t surprise me that people aren’t working fewer hours. And then in terms of whether people can find their contracts or not, again, that... Think it’s around 25% know... Have all their contracts in one place. That, based on my experience, is probably quite high, to be honest. And it may be that they know where they are, but they’re not particularly well organized or structured.

But, you know, I I come across a lot of legal teams and a lot of businesses. Their contracts are all over the place.

And, yeah, and in terms of whether teams are are really ready for AI or not, we’ll we’ll come on to it. But with the tools as they are these days, you can get some immediate value. But in terms of using legal AI to their absolute potential, I think most teams are probably quite far away from that. You know, a lot of a lot of teams that we come across haven’t really thought about their legal strategy, what they do, what they don’t do, what they’re supposed to focus on.

They haven’t really thought about the problem that they’re trying to solve or the key problem they’re trying to solve. And none of this is particularly new, but they haven’t got their processes in order, their documentation in order, templates, playbooks, etcetera. You know, there... There’s a lot of foundational work, I think, to be done there before you can really see the full potential of legal AI.

And I think the main problem is certainly in a lean in house legal team. They don’t have the time that’s necessary to do that work in order to properly benefit from legal AI. So I think that’s... You know, we’ll get into it in more detail, but I think that’s the biggest challenge they’ve got. And what’s your pick on that, Nathan? You know, with the... Obviously, as a legal engineer at Wordsmith, I mean, get more into what a legal engineer actually means for Wordsmith as a company.

But what what are you seeing across the customers and prospects you’re speaking to? Yeah. Definitely the the same that Rob’s describing in terms of just having time and capacity to actually get into some of the stuff and and the sort of line of I’m too busy to spend the time figuring out how to make myself less busy. And that’s really, really difficult thing to deal with because they know that if they spend an hour or two getting into an AI tool and and starting to think about, okay.

What’s something that I keep doing every day or every week that’s really repetitive? I’ve got a sense that AI can help with this, but I’m not quite sure where to start. I know I’m gonna figure it out in an hour or two, or I speak to a legal engineer who can get there a lot faster. But then you’ve got a mountain of work to get to and spending an hour to a start is just not feasible or possible. And so oftentimes when we’re starting out with folks who are just getting wordsmith or using AI tools, it’s figuring out, okay, day one value.

How can you get into this and just start using it to start, you know, time immediately? Because if they don’t get that, then they’re not gonna... You know, unless they’ve got some sort of mandate from the team or something that we’re going AI native or, you know, whatever that means for your team. We’ll be using AI. We’re gonna give you some time to actually figure this out. Unless you’ve got that as an entity, you need to get day one value.

Otherwise, it’s gonna be really, really difficult to actually start adopting. And I wanted to dig in, Nathan, to the legal engineer role, which I know is something that that Wordsmith, you have you have lots of legal engineers. It’s a big part of the offering you have. Something that stood out to me and and Rob was when we spoke to these kind of heads legal, a lot of people were talking about Wordsmith, and the message was resonating with the fact that you’re focusing on this in house market.

What is the legal engineer, and how does that tie in to maybe some of the problems that we’re seeing? Yeah. So legal engineer as a title means very different things depending on which organization is using the title, what the job description is, etcetera. The way we’ve split it up at because I realize there’s many different elements to being legal engineer. We split it into good market legal engineers, solutions legal engineers, and product legal engineers.

So the good market legal engineers are generally tax lawyers and legal folks sitting alongside our salespeople or accounting sacks, etcetera, going out, doing demos, making sure that what we’re presenting as an offering actually meets what people are looking for and isn’t just a generic, this is a feature. This is what it can do, but it’s this is what Workspace can do and how it will benefit your team.

So that’s one standard of lead engineering is to make what we’re presenting to people more applicable to them. Solutions lead engineers, that’s what I do. It’s sitting down with current customers who will go... Will have a discussion about what are your processes, what are your workflows, where do your documents sit, how do you currently do things. I’m a big fan of flow charts and Figma diagrams.

So it’s map out, pick one or two of their workflows and just map out very clearly. We have a discussion around that. And oftentimes, when we do that, we’ve kinda had a chat about here’s our workflow, and we get it on the Figma. And then different team members start pitching in. We’re going like, oh, wait. No. Is that how we’re meant to do it? I’ll do it this way. I just... Whenever I need someone to go and ask Anne.

Anne was everything. And Anne’s in a completely different part of the business, but she’s been there for, like, twenty years, and she just knows how to do stuff. And so, actually, when you get your workflow on a page, then you can start as a team going through, okay. This is the current state where you want to get to. And once we’ve got the sort of view of where you want to get to, then we can start using Wordsmith to start automating things, passing stuff over to agents, etcetera.

And then we’ve got our product people engineers. So these are ex lawyers who sit with our product team, with the developers to make sure that every iteration of Wordsmith, every new feature is stress tested against what lawyers actually want. So it’s not built with lawyers in mind. It’s actually built with lawyers. Yeah. It’s such an interesting role. We have a similar thing with Nomio as a kind of document programmer piece in there.

Can I just say on the the go to market legal engineers? I could be wrong, but that... I think that’s fairly unique. I’ve I’ve sat on no end of demos in the past in previous roles where you don’t have people acting in that capacity. And so it’s a generic sales pitch, often starting with an NDA, which often isn’t a real problem anyway. And so I I think that’s really important to have people in that capacity able to help show prospective customers how it might solve their specific problems.

Yeah. Exactly. And we’ve got, like... For the go to market team, we’ve got x restructuring lawyers, x commercial lawyers, x corporate lawyers, employment lawyers, and some x GCs, x in house counsel. So when we’re putting together those, like, demos and, like, hey. This is what you can use. Whereas before, It’s not that this is what Wordsmith can be used for. It’s this is what you can use Wordsmith for.

And so actually being able to draw on experience with people who are doing or have been doing the role that the people we’re speaking to is doing just makes it, like, add so much more color than, as you say, the generic. Here’s an NDA, and we can show you this feature and da da da. I think especially safe because I know, Rob, we... We’ve talked about you guys, like, legal tech talk. We’re gonna be going to legal geek next week.

It’s extremely overwhelming. I’m in the process of writing a big state of legal report, and I’ve spoken about 70 in house lawyers over the past couple of months. Should be a 100 by the end of October. And the sense I get is the in house teams are feeling very overwhelmed because there are new tools all the time. Like, what’s your sense of that, Rob? How how should they navigate that? It’s a complete minefield.

I I think I went to legal tech talk, well, I’ve been... This year and last year. Last year, I’m sure there was 80 plus AI contract review tools kind of doing the same thing, different name, different slogan. Like, it’s it’s impossible to navigate that. Though, you know, all of these organizations will be doing a lot of outreach, spending a lot on advertising, marketing, etcetera, and, you know, the in house lawyers will be being peppered with different solutions.

One thing to absolutely avoid is just getting straight into pictures and demos because, again, I’ve been on that side of things, and and this was more like CLN days, but everything just becomes a bit of a blur. Like, you forget what was good and what wasn’t good and who you spoke spoke to. You know, once once you’re on your, like, seventh, pitch for a similar tool, and, you know, it’d be the same now for for legal AI tools.

I think the legal team working with stakeholders really needs to sit and have a think about what it is that they that they need help with, what they need to solve first before starting to try to navigate the market. I I think it’s really critical to do that foundational thinking first. What is the biggest problem they’re trying to solve? It may be that legal AI is not the solution at all. It may be that it’s something else entirely.

It may be that it’s a CLM or or whatever else it might be, that’s what they should focus on first, not just do more and more demos. And and I would say, you know, we... We’ve touched on it earlier, but focus on a tool that’s... Like, once you figured out what it is you want, focus on a tool that’s actually designed for the in house market because some of the big players out there are moving into this space, but that’s certainly not where they started.

And so not necessarily best for in house. Christopher, it’s so good to have you here. We’ve just been discussing the busyness of the legal tech market, the new vendors coming in. We’re really curious to hear about your experience on this. Before we do, perhaps you could give a kind of a quick introduction to you and your role. Sorry. Technical difficulties. Zoom was blocking me from joining. So I’m assistant general counsel at Safe Flight, which is a subsidiary of Bellron located in The UK.

I’m responsible for, among relevant things, all things technology at Safe Flight and AI governance as well.

Fantastic. And... Thank you. One of the one of the themes we were talking about earlier was this idea that I think a lot of in house teams are running before they can walk, especially with AI adoption. They’re sort of... Regardless of where their contracts are, regardless of the setup they have, they’re just buying something and trying to implement it. And then there are May six months later when they get a big bill and failed implementation.

You do things the other way around. You started with getting your contract data in shape first. Could you talk a little bit about that and how you approached getting a kind of contract repository together? Sure. So, I mean, probably similar to many other organizations. We had contracts, kind of everywhere. Legal stored some in a network drive. We had account owners, storing in their email or their their, their work computer.

And it was extremely difficult, if not impossible, to actually understand what was the true source of truth. You... Someone asked you a question, what do these contracts contain? You would have to spend so much time digging through files, to actually find the agreement and then to analyze it. So we really look for a solution where we could automate the surfacing of data, out of your contracts in a way that, everyone could access in a way that it would be updated as new contracts are signed.

And so we happen to be listening to a crafty council presentation about Nomio, and it just like, oh my gosh. I... I’ve been searching for this, and there it was right on my lap. So we’ve led a... Or I led a project around maybe two months or so implementations, eight weeks, six weeks, where we provided all of our agreements to know... Tanomio, and they were able to arrange everything by order of precedence.

Kind of... We gave them all the information that we wanted to know about the contract. They surface that data for our review, and we couldn’t be happier with the product. Amazing. And so you’ve got your contract repository together, and then you go to market to look at legal AI. This is the busy space we see. How you navigate that market? How did you approach the adoption? Sure. So I’m... Well, guess I was a mechanical engineer, and so I care a lot about process and data.

So in my role as leading the legal technology subgroup, I posed the question to our entire global legal team. What type of technology are you interested in? What would be valuable to you? And I listed about, maybe 15 different options. AI was one of them or just traditional CLM was another. And we got very positive response that we’re very interested in using generative AI. So I then took that as a data point and said, well, how would you use it?

So I laid out about, you know, maybe 40 different use cases, everything from contract red line to managing corporate governance, legal research. And I said, would these be valuable to you? And I got back all the survey results, and that told me about... Of that 40, there are about 20 to 25 use cases that would be broadly applicable to the legal team. Well, that became my requirement set. When I went out to look at all the different vendors, I was, evaluating them based on how well they did those 20 to 25 different requirements.

We benchmarked Harvey and Legora, g c, dot a I, Wordsmith, maybe one more. And Wordsmith just blew them out of the water. I mean, the... In terms of requirements, but then also just the usability, the interface, the support and information I was getting from the team. So we... Once we kind of established Wordsmith as this was our go to, because I had the requirements list, I could then build that into, the business case.

Because I could say, well, these users will use it x number of times. They’re saving y minutes, by collecting data before and after a trial, and then backing in kind of approximate figures in terms of salary, you could roll that all into... Well, this amount of time is saved in total across the team, and it’s valued at this amount of money. So kind of building the business case was was rather easy because I collected all the data upfront.

Nathan, what’s your reaction to that? I’m sure you must hear all sorts of stories about selection processes. I’m curious to get your take. Yeah. And when we had this sort of webinar preview and Christopher described this process for selection, I was just out of thinking this is the best selection process I’ve ever heard. It’s so, so good. And did selection processes for different organizations look very different, and it will depend on size of team and resource to actually put into selection.

But too often, it is that case of just you get a top down push to say, we are going AI native. We’re gonna use AI. Go and get a tool. And they just go and buy something, and they’re like, as you said, six months later, like, yeah. That didn’t really work, and it’s not our requirements. No one’s using it, and we’ve wasted a lot of money here. So Christopher’s process of actually leading with the business, finding out what are you actually gonna use it for, what are the requirements, and using that to rank and rate each tool for each of those requirements before making a decision is, like, chef’s kiss perfect for selecting a tool that people will actually use and you’ll get audio out of.

Yeah. I think it’s what’s what’s really interesting as well. I think, Rob, we talked about it. So many people forget about starting with a problem, being able to very clearly articulate what is the problem you are trying to solve rather than... You know, we’re all a bit like magpies. You see these shiny cool AI tools and you get them because they look cool, but you’re maybe not entirely clear on what the problem is.

And so by having that requirements list, it helps so much in in ensuring you get the adoption across the business. I So guess comes to the next point, Christopher. You get... You decide on the tool. How do you manage that change management piece across a sort of 60 odd person legal team?

The change management piece is the hardest of the whole equation. Actually, finding the the right tool is is easier in retrospect.

So there’s going to be a range of skill sets when you deploy an AI tool to a team, large or small. You’re going to have individuals at the lowest rung where they’re kind of scared of any new technology. They’re scared of tech. And there’s the people at the top that are on their own time, you know, vibe coding apps that are managing their whole personal life. So once you kind of recognize that there’s going to be a range of skill sets and interests and comfortability with AI, you start to develop different strategies to how to attack those.

So for example, I mean, Ortsman did a great job of trying to onboard us with here’s some basic training, here’s some basic education. They then kind of just after that, this is what the tool does. They started to understand kind of our pain points. So what are the two... Or what are the other problems we’re trying to solve? Thankfully, I had that requirements list. It was pretty easy to transfer over.

And so they could build out specific modules of... If you wanna learn how to do x, you know, join at this time. If you wanna learn y, join at this time. They also opened up office hours for our team. So they would have someone on call at a specific time that people could just drop into the meeting, and they could ask questions of, I’m I’m having difficulty doing this. I were having... Or I wanna know how to use this feature.

So they can answer kinda live questions. We then kinda brought it out to kind of more of a push model where we would release, emails with, have you ever thought to do x with Wordsmith? Here’s a sample prompt. Here’s how you can just copy paste it and go ahead and do it. We then... We’ve also built out a use case library where we selected kind of all the non obvious uses or ways you can, deploy Wordsmith It made it very easy for people to search it by creating a skill, and then people to contribute to it by also creating another skill that would search their own history and kinda create a standardized template to feed back into a report.

We we wanted to pair the people that are kind of the lowest quartile users with the highest quartile. Kind of kind of thinking that through more clearly, we decided it’s probably not the best approach. But as a kind of a next step, we were thinking, well, can people opt into kind of like a buddy program or a training program where someone knows that they have a particular, you know, deficiency and skill sets but they can get paired with someone that’s a high volume user, and they can kind of bridge that skill gap.

There’s there’s one thing that I had thought about yesterday is kind of, like, phonal enablement. And kind of the idea there is that it’s really incumbent on the management team to be making it okay to use AI in creating your work product. So we have a fantastic CLO here where when we’re talking in department meetings, or even just, you know, regular meetings, she’ll... We’ll identify a problem, and she’ll say, oh, well, you could use Wordsmith to solve a problem with those x, y, and z steps.

So here we here we are kind of having a senior leader telling the rest of the team, oh, you you wanna solve this problem? Wordsmith can do it. And by the way, here’s kind of high level steps of how to do it. So kind of bringing that all back around is people will respond differently to different messages, to different interventions, and you kinda need to build a variety of them to attack that same problem from different angles.

Great. And as you were talking, Chris, we’ve had a we’ve had a couple of really good questions come in that I’d like to open up to the group. The first is, how do you manage individuals who are super skeptical about AI and see it as a threat? How do you manage that? See it as a threat, maybe, like, to... Well, I guess I can’t ask a clarifying question. So there are people that are not gonna necessarily trust AI.

You’re gonna have to build that trust over time with any type of technology or really any type of person or resource. So I would I would say start small. Start with things that are verifiable. So redlining a contract. If you’re a contract expert and and you put your contracts through, through a playbook in Wordsmith, you can very instantly see the quality of the output. If you’re doing legal research, Wordsmith very helpfully provides you direct links to the subject matter material.

Click into those. Review the statutes. I mean, you should be doing that anyway as kind of your own due diligence verification step. And then if I’m understanding the threat portion of that con... Of that request, I view this as being a force multiplier. There are... There’s no shortage of opportunities to pair, or to to partner with the business to solve problems. Wordsmith enables you to do that better, more efficiently, provide broader or deeper set of analysis.

So the need is always going to be there for, you know, for attorneys, for, you know, knowledge workers. It’s how you leverage that technology to make you better at what you’re already probably very good at. Yeah. I don’t think that human layer is going away ever. Certainly not anytime soon. I think that’s... That is probably the threat they’re talking about. You know, no matter how powerful the tool is, you still need that judgment, that risk awareness, commercial awareness to be applied.

And it just elevates how you operate as a lawyer, I think. But I do understand the concern because I... Probably a few months ago now, there was a lot of noise about how AI was gonna take all of our jobs, and, obviously, that’s not the case. It it will just make us better, more efficient, more powerful as always. I just wanted to touch on a couple of things that, Christopher was talking about previously in relation to selection process and change management.

Christopher benefits from working in quite a big legal function.

Lean legal teams or sole councils will have a very different experience with all of this. So I think first in relation to selection process, they’re already drowning in work as we talked about earlier. They need to somehow offload some of that business as usual work and potentially bring in some sort of legal ops, legal legal engineer consultant to help them with all of this. But then on the flip side, I think the change management process with tools nowadays might be a lot simpler.

Obviously, the legal function itself is a lot smaller, so you’ve got fewer people to convince that it’s a good thing. But, also, lots of the tools, league... Legal AI tools and even CLMs now are integrating with the existing tech stack. Slack being an example. You know, if your organization works in Slack and everyone’s used to dealing in Slack, they don’t have to leave Slack to benefit from the likes of Wordsmith.

So I think the change management piece can be a lot easier than than it used to be, particularly with the lean and legal team. But I think I think also what I’ve learned in speaking to lots of lean and legal teams is don’t underestimate the power of communities. There are some fantastic in house legal communities out there, and you can drop a WhatsApp or a Teams message or a Slack, or you can ask a question at a group of what’s working for you because you get that word-of-mouth, that referral.

You can see and say, hey. I’ve got this problem. And I think as long as you can articulate the problem you have well and ask your peers, lean on those communities. When there’s so much noise in the market, I’d really, really recommend doing that. I’m curious, Nathan, on your take.

In in relation to most of my adoption or the... Will AI take over everything? In in relation to adoption, I guess. Adoption. Yeah. Yeah. For for adoption as well, it’s it’s like we’re saying earlier. Like, off the bat, you need that, like, day one value where you can open a tool and just start using it and get value from it. Because if you don’t, you’re gonna open it once, try it. Doesn’t work for me.

I’m gonna go back to actually doing the job I need to do, and so that’s really important. And then as Rob is saying as well, just having whatever tool you select meet you where you work. So, like, Rob mentioned example of being able to use and interact with Boris Smith directly from Slack. Super powerful. If that’s where you’re working, if that’s where you’re receiving queries, if that’s where you’re operating, being able to put tool that just links into that, is super powerful and super important.

And our own change management in general and some of the stuff that Chris has been doing and and Rob as well around Wordsmith being able to connect into other tools like Nomio or, like, your SharePoint or Google Drive or, like, wherever your documents are stored. Having a tool that can just link straight in and not having to go through a big exercise of rounding the herds of documents and then getting them all into the one system and then having a process in place to ensuring that is continued going forward and it’s not just a one on one project.

And, you know, some of that’s the the hangover of the CLN piece where there is such a massive change management exercise that, you know, you’re two years in to a CLM rollout, and you’re still not complete yet. Like, the the world has moved on some options. Now you do just need a tool that... Very quickly that you can get up to speed to integrate and get value from very fast. And then just on the...

Set below on the... Will AI take our jobs? And I see there’s another question on how do you deal with the... Will worst best deal my job reluctance? I’m very much in agreement with Rob and Christopher. If anything, I see it going the opposite way. There will be a greater demand for legal, especially within a business. You’ve got so many questions within the business, so many business people that are sitting there going, I should send this to legal, but it’s gonna take, like, a week for them to come back to me, and maybe it’s not that big of a risk.

And that’s fine. I’ll just push it out. And then four months later, they’ve pushed it out. It’s... Something’s gone wrong. They didn’t consult legal. The risk was better. Then And they’re coming to legal teams like, hey. So I did this a few months ago, and everything’s gone bad. Can you fix it? And so, actually, within companies, if legal can operate faster and more aligned to businesses’ needs and kind of show that they’re on the front fruits of being proactive and mitigating risk before it arises.

The business wants more of that, not less. So the whole question about will AI come and take jobs, it’s AI will enable you to be even more valuable to the rest of the business. Completely agree. Question for you, Christopher, we’ve had in, which is I’m currently looking at solutions and considering Nomios repository and Wordsmith for pre signature. Two questions. You’ve mentioned the Wordsmith onboarding is good.

What was it like with Nomio? And two, how do you actually use the two of them together? I think, Nathan, you mentioned this connection piece. Curious to dig into that and what that’s looking like as you’ve been through that adoption journey. So with respect to the Nomio onboarding, it was white glove. The the the team, particularly the the the sales team, kind of met us exactly where they were. They said, we don’t care how you get your documents to us.

Just... We’ll find them for you. We’ll go into your DocuSign. We’ll pull everything down. They would... They said, well, most clients wanna know these data points about your contracts, but we can also provide you all of these other data points. So they they really did a fantastic job of holding our hand and doing all of the lifting. It was really more of us just to make the decisions. So I can’t be happier with that.

With respect to the integration of Nomio and Wordsmith, it’s... I kind of use Wordsmith as almost like my command center. It is a place that is connected to Outlook, Teams, SharePoint, NetDocs, which is our document management system for all contracts, and then, of course, Nomio for post signature. When... I’m working on a large project where we’re we’re selling, a new b to b product. And so we have hundreds of existing contracts that are stored in Nomio that need to get amendments done on them.

And so we can, you know, actually sell this product to our clients. And so what I have done is create a pair of agents that will, all I have to do is launch the agent, Just give it the name of the client, and it will go into Nomio. It’ll find all of the agreements associated with that client. It’ll then go into a into a knowledge base or repository, pull in the templates for the... For that amendment.

It’ll then build out the the the order of precedence. It’ll build all... Or kind of create all of the, the terminology so that the the amendment matches what the... What the client refers to itself as and what it refers to us as. So it’s kind of the... You use Nomio as the place to store things, but it’s also the place to fetch it. And so the the contracts that are stored there are gonna be used subsequently, like, across the the life cycle of your clients.

That is so cool. I think I love hearing some of the stuff that I’m hearing in house teams building now that you have tools that can connect together. I was speaking to some of the legal operations person at the Financial Times, Erin, and she was saying that as part of their rollout of Wordsmith, they run weekly show and tells where it’s someone in the legal team’s opportunity to show something they’ve built like an agent, and you create that fear of missing out, that FOMO.

So again, when it comes... You go from feeling maybe nervous about the rollout of this tool to suddenly really excited to think, oh, shoot. Maybe I could actually use this to solve this problem or build something. Which, again, I think it... When it comes to the adoption, it’s something that’s a really interesting thing you can try and use when you’re rolling these tools out is running those weekly or monthly show and tell sessions just to show the kind of stuff you’re building.

Because I think that sounds amazing, Christopher. Are there any other kind of agents that you’ve got for similar use cases across the two?

Across the two. That’s the primary one. The the the amendment builders... I can’t think of another one that uses Nomio, but I I mean, I have a ton of agents that do all kinds of things. I I built one where, when we’re onboarding someone new, you give... Essentially, it creates kind of a, this is what you should be doing on day one. This is what should be doing on day two. They get these emails. It’s built to the 30, 60, 90.

It’s built to, information that’s relevant to their position, pulls in information about SafeLight as a business, tells them, you should go meet with this person, and here’s the topic that you can discuss with them. So there’s... I mean, the the possibilities are endless in terms of agent building within, Wordsmith. I’m curious to hear, Rob, from your perspective. When we spoke to people, the appetite for legal tech and legal air...

AI really varied. You’ve got people who are building agents, and you’ve got others that are terrified to even open up Copilot. Why why do you think that is? And especially across maybe some more of the leaner teams as well where they’re just so busy. They don’t feel like they have the time. Why do you think that is, and how do how do we try and close the gap? I think time is the biggest factor, as you said already.

They’ve they’ve just got too much on to be able to spare any time to look at this at all. But lawyers generally are also pretty risk averse. There’s there’s natural concerns about security, confidentiality, legal privilege, you know, accuracy. I I think Nathan touched on it before. You know? They try it once. Something didn’t work perfectly, and they think, oh, this is rubbish. It’s not for me. I’m just gonna do it myself and then the way I always did.

Obviously, that... That’s not the right answer either. How how do you close the gap? I think... Sorry. Before I go into that, I I don’t know that hurdle is, which, again, we’ve kind of touched on. People have been banned in the past by failed implementations of other legal tech.

I’ve had similar horror stories to the one Nathan was talking about where people have spent $250, and it’s just about automating NDAs, just not not doing anything else. You know? So they’re spending money on rescue missions or bidding something they’ve already spent a fortune on. So, again, natural nervousness. It’s it’s a very different market now. The tools are very as you’ve said, it’s it’s very quick and easy.

Depends on the usability of the tool, of course, but it’s very quick and easy to see immediate value. You can open it up, run a contract through it, run a, you know, a question through it, a policy through it, whatever it might be, and it will help instantly in some capacity. So I think it’s somehow getting people over the nervousness of that. You know, it’s gonna take a huge amount of time or I won’t see immediate value or I’m gonna spend a fortune.

It’s just not the case anymore. The market’s moved on significantly. And then it’s up to the the vendors to be really selling their tools in terms of the uniqueness, the power of their tools, confidentiality, security, etcetera. And some are obviously doing that much better than others. But, yeah, it’s a difficult one. I think the the natural risk averse nature of the lawyers, But also as we’re seeing in the questions here, people worried about it taking over, so there’ll there’ll be that skepticism.

They’ll be concerned about whether they should use it at all because it gets so good that they don’t... They’re not needed anymore. It’s just not gonna be the case. They need to... As we’ve talked about earlier, build the... Do the foundational work, go through the selection process, get the right tool for the right problem, and then just grow from there.

And, you know, they can go from a basic user to a power user pretty quickly. What role they Just to find the accuracy piece. But... So there’s a differentiator at Nomio that isn’t readily apparent until you start going through the onboarding. It’s accuracy. Nomio is obsessive about data quality. And it will not surface information to you if it’s not a 100%, sure that it’s accurate, to the point where you have almost a small conversation about little data points within the data point of, you know, the the the platform where a human is reviewing the output and verifying that it’s correct.

And if there’s any ambiguity, kind of going back out to the client and say, we’re not entirely sure about this data point. Can you verify it for us so we can kind of put a pin on it and say that it’s correct? So that is that is a level of service and a level of comfort that you can’t really get elsewhere in the marketplace. Amazing. If you ever want a job in sales, Christopher, just hit me up. We’ve had a we’ve had a question which is a really common thing we see, and I’m sure, Nathan, you must get this a lot as well with Wordsmith, which is we really struggle to get buy in from senior leadership to spend on legal tech.

Often pushback is why can’t we just build something internally? How can you convince leadership of the value of these specialist solutions? You must see this so much, Nathan. I’m sure Rob, Christopher, you’ve had this as well. How do you get by in? Yeah. There... There’s a few different points to that. One is the question, but why can’t we just build this ourselves internally? For something like Wordsmith, we’ve got dozens of developers working on product side of things and then dozens of ex lawyers and legal ops peoples and GCs and and those and legal counsel within Wordsmith who are all building the best product possible.

And on top of that, we’ve got hundreds of customers that are all using it, feeding back, building out different use cases, some of the awesome agents that Christopher has been doing and stuff we see across our customer base. And then as, all of these different use cases push the product to its limits and then beyond, we continue to build out more and more and more to make the best product possible.

So it’s... With vibe coding tools and being able to do certain things, it’s not to say that teams couldn’t build some version of a tool to help them internally, but it’s... Would you rather spend the time and resources doing that or get a tool that has got all of this sort of resource and backing behind it to really pitch what you actually need and then build out from there? And there’s also the maintenance burden that everyone forgets if you build something internally.

Doing a one shot to the five coded solution is great, and it can be really cool to help with a particular thing. And then it becomes part of the team’s workflow and process, then you’re just like, okay. Who’s who’s maintaining this? Who’s managing it? And then you’ve just shifted the burden of we’re too stretched, and we needed something to help us to... Now we’ve got more stuff to manage and maintain ourselves.

So that’s really difficult and something that senior leadership don’t take into consideration. A lot of the time, they’re saying, can we just do this ourselves? And in terms of buy in and showing value, getting on to, like, a trial, something like Wordsmith, where, like, if there’s no commitment you got to stick with it, but just, you know, use it for a couple of weeks, use it on real work, real matters, and show the value very quickly, then gives you a really strong business case that you can go back to senior leadership with.

With statistics, with numbers, measure how much time you’re spending doing things before using a tool like Wordsmith or Nomeo depending on the use case. Do that trial phase and then show, hey. We used to spend four hours doing this task, and now it’s twenty minutes. We used to spend a lot of time dealing with these queries, and now we’ve got an agent that the business can self serve with. So, actually, giving to your leadership numbers, metrics, data allows you to make a much stronger business case than just saying we know what we need and we need something to help us, but we don’t have the numbers for that.

Amazing. We’re we’re pretty much at time, Robert. Just be keen to hear just finally your your perspective on that, certainly from the lean teams. Any advice or any things you’ve seen that have helped get that buy in from senior leadership? I think it’s kind of posing questions back to senior league... Leadership. Would you like our tech folk... Focus building the product that we’re trying to sell, or do you like them focusing on building a tool to help legal?

And, similarly, do you want your legal folk focusing on legal work to, you know, help drive the business forward, avoid risk, whatever it might be, or do you want them to be trying to build a tool and probably not doing it very well when there’s all this other stuff out there that’s been designed specifically for what we need? Having said that, there will be some value that legal teams, lean legal teams can get from the existing tech stack as a short term solution while they try to build a business case.

You know? Things like using, yeah, project management tools for intake. You know? If they’ve already got Asana, Jira, monday.com, that kind of thing, they can use that for legal intake. There there will be some quick wins with the existing tech stack, but they’ll quickly find they get to a point where it’s just not enough and building something for all the reasons that Nathan’s just stated is not the right answer.

And things aren’t... Just aren’t as expensive as they used to be. Again, we’re not talking hundreds of thousands here. We’re talking, you depending on the size of the team, a few thousand a year. So it just doesn’t make sense for many reasons to attempt to build it yourself. I’ve I’ve loved this conversation. I feel like I’ve learned so much. Unfortunately, we’re at time. We’ve had a ton of questions, so we’ll do our best in the follow-up to try and answer those.

I’d say my three kind of key takeaways here are don’t run before you can walk, try and fix your data first, get your contracts in one place, and then build the AI and the agents on top of that, on top of clean, crisp data. When it comes to looking at things, know your problem, pick your own use cases, and make sure you score the demos against those use cases and be really clear. I think the final thing is adoption is not a launch.

It’s a program you’ve gotta run. It’s a program you’ve gotta run over time as well, and you’ve gotta meet people where they are. So a big thank you to to Nathan, to Rob, and to Christopher for joining. A big thanks to everyone for dialing in for all your questions. Sorry. Couldn’t answer all of them, but thanks so much for being so keen. And we’ll have plenty more of these webinars coming over the coming months, so stay tuned.

In the meantime, have a fantastic Thursday wherever you are, and we’ll catch you later. Thank you. Thanks, everyone. Take care, everyone. Bye bye.