Inside Addleshaw Goddard’s legal tech strategy: innovation head Kerry Westland on managing a 70-person team, the explosive rise of AG’s internal AI tool, and what the future of law looks like

Today's guest is Kerry Westland, a partner and head of innovation at Addressure Goddard. In our conversation, Kerry shares how experimentation, curiosity and a bit of weekend coding turned into a career driving change inside one of the UK's most forward-thinking law firms. We unpack the real meaning of innovation in Big Law, the tech stack that's powering Addressure Goddard, the explosive rise of the firm's own internal AI tool, how to balance tech adoption with ROI, and what the law firm of 2035 might look like, including the big question, should junior lawyers be worried? Now, you've carved out a role, Kerry, that didn't even exist in most law firms ten years ago, head of innovation. What led you to head up innovation at a big law firm like Addressure Goddard? It's always an interesting story because it wasn't one that was planned, like most of these roles, so I joined AG about 14 and a half years ago now, and I joined as a paralegal. I needed, I wanted to be a lawyer, so I was getting some work experience, and it was just after the recession, so it was like 2010, and there was a focus at that time on almost like the people element of innovation now if you were looking at it, so after the recession, obviously everyone was looking at costs and things, and it was very much, well, who could do the work? Does certain pieces of work need to be done by partners, or have you, how could you look at it? So, Addressure Goddard created what is the transaction services team in 2010, and I was one of the first 11 paralegals that joined that team. It's now 200 to give you an idea of the scale and the size that happened. And when you think about it, obviously the work that paralegals were doing in 2010 and onwards very much is the type of work that lent itself to the use of technology at the time, like as things went on, so we were doing a piece of work for a big client, and it has got to the point where we were like we can't do this work without a platform now, without some technology to manage it, so it was like thousands of contract reviews that needed to be done across multiple people. So, at the time, was that, you know, your sort of traditional sort of manual processes, lots of sort of word documents and just lawyers spending hours drawing through these things, yeah. Yeah, and I always remember once producing contract summaries for a client, and it was a word document for each contract summary, and they were sort of like, well, now I have to go into each word summary. I might as well have gone into the contract, so we were starting to have this, and it sounds quite, not revolutionary anymore, but putting them in spreadsheets and things like that, we'd started to think about, but when we were doing thousands of contracts across 30 or 40 people looking at it, we were like, how do we do it? So we actually used a tool, which I don't even think exists anymore, which was Apple FileMaker. I don't know if anyone even remembers it, but it was basically Apple's version of Access, and we were then obviously recording contract reviews in there. And it came about that you couldn't, you could have two people in a record, and I promise there is a point to the story of how I got here, but you almost, you could have two people in a record, so therefore if someone changed the record, you didn't know which was the last like amendment, and with like the team and like we can't keep doing this, we need to know like the one version of the truth. So I spent the weekend googling how you could lock these records and ended up with some code, and it ended up working and being able to lock these records, and there's suddenly everyone was like, you're the tech person, and I'm like, I don't know if I app. So you had that, and then we had a tool at the time that we'd sort of bought in for something and locked all the other functionality off, and then I started to hear what they were doing, so that was high key. And I started to hear what other firms were doing, and I was like, we have high key, what's happening, and then ended up speaking to IT and saying, I feel like we should switch on all this other functionality, which we did, and then I got a request to say, could you build a case management system for a client for claims management? And we were like, yeah, we could probably do that, and we built something overnight and demoed it to the client, and that's how you become a tech person in a law firm. And then that started with me, and then a couple of other paralegals came along and were helping manage those large projects. And then in 2017, it got to the stage where it was like, well, actually this could be an entity in its own, and we became the innovation and legal technology team, and then that's grown from there to the innovation group now, with 60, 70 people in it, all focused on how we apply innovation to legal technology. So it's quite a long-winded way of getting there, and it was a bit of an accident, but sometimes that's always the best way to do things. So how does a firm like Adolf Shaw got a defined innovation? It's interesting, isn't it? Because innovation over the years gets a bit of a bashing as a word, like has it overused? For us, it's about how do you come up with new ideas and new ways of working, bearing in mind the environment that you're in. So if you took like, in 2010, it was about the people, who obviously then became about the technology, and it's innovation isn't just technology, but it's obviously a big thing that underpins it, and it's the really obvious piece. But it is also how do our lawyers advise on innovative mandates and the first of the first of the things out there and setting precedents, as well as getting our associates and our junior lawyers to be like, well, why are we doing it this way? Could we do it in a different way, whether that's technology or just sort of looking at the output and thinking about things like that? So for us, it's very much new ideas, new ways of doing the things we've always done, but also new service lines that might be coming along as well, and just constantly having that thought as we're thinking about what we do day today. And what does your day-to-day look like then as head of innovation? We're thinking about this, because people ask, and it's a really cliché way of saying it's not the same, and I was thinking about it this morning because I literally jumped from one thing to another, and we might be like talking to a vendor and signing off the contract and thinking about how we're rolling that out and what we're doing, or it might be talking to a client about their own journey on legal technology, and I've got a call in that day about that, and then I'm talking to the marketing team about a client event we might be doing. So it's really, really varied, and I think one of the things I've definitely learned over the years is that switching is expected. You're constantly switching from sort of one thing to another. If we look at what we do, there's sort of four main key areas, which is how our lawyers work day-to-day, how we manage and deliver large, big, unusual projects, the piece around the client's own journey, and then how you might make things a bit more repeatable. So if you've done something once, how could you, almost like tech-driven services, we call it, how do you build something that you can repeat all the time, and then lots of other things in between around management and vendors and things, but that's generally your sort of flitting between those each day. And how much of that particular aspect, then Kerry, is kind of tech-driven, versus kind of process and systems driven? Yeah, obviously technology is a big element of it, and I would say, especially over the last two years, that's where there's been a really big, you can just see the uptick in engagement from clients and internally. But when we're looking at bringing technology in, you don't just go, oh, here's a tool, you do go, well, what does this mean? What, what does it change? How does, how do we resource something differently? How do we price something differently? What, what might the output now be looking like if we're going to do it in those ways? So sometimes the technology drives some of those behaviors, sometimes those behaviors drive the technology that you're looking at. So it's sort of a mix, and we really try not to lead with the technology, but sometimes you see really cool tools. So you're like, actually, that could really solve one of those problems that we had. So there's a lot of technology, there's a lot around GNI, but not everything's GNI. But there's a lot of change management that comes with that as well, and then having those discussions, so it's a real mix. And will you be responsible then for, for example, looking at how best to structure a particular team to just deliver more efficiently, kind of almost irrespective of technology at that point? So we would, for us, innovation, although as much as we have the innovation group, we're not responsible for all innovation in the firm, we're a bit of a conduit, and we try and help it. So I'll be having those conversations with those practice groups to be like, actually, what could we be doing here? We're seeing what other firms are doing, we're seeing what clients are asking for, could we be looking at it, and we do that jointly. To sort of change our team does work, you would need the practice groups involved, you would need like HR get involved, finance get involved. So sort of helping be a bit of a conduit to bring that all together, rather than that being my responsibility or my team's responsibility. Okay. And I suppose Adosual Goddard is in the market kind of definitely considered one of the most innovative forward thinking firms. And what does Adosual Goddard's kind of tech stack look like in terms of, you know, some of the perhaps more widely known legal tech and AI tools that you are currently using? Yeah. So it's quite interesting. And I actually did a LinkedIn post on this last year. When I first started this, we sort of had high key. And that was the main tech piece that we had. And I spent quite a bit of time trying to get high key to do everything and become that end-to-end platform panacea. And I think that's where a lot of people were thinking because everyone was like nobody wants lots of log ins and all of this type of stuff. So there was a bit of a focus on really trying to be could you take the tools you already haven't do something with them. So through either Microsoft tools or or high key at the time. That was about a decade ago. And actually my viewpoints changed and we now have. And I think the reason it's changed is there's been some really, really good companies out there that do something really well. And they're that like point to say point solutions that do something really well. And when you think about like the work of a lawyer, I think lots of people when they think we can do process and we can do this, think it's quite linear. And you go, you're going to do this. You're going to do this. It's not. If you take an M&A deal, you've got various specialists all off looking at what's in data rooms. You've got the drafting of the SPA going on. You've got ancillary documents that you're trying to produce. You've got Q&A coming back between different advisors and clients. So you've actually got a number of people working on multiple work streams. And having that in one platform is really, really difficult. And then you started to go, well, actually you've got tools like obviously we start with document automation to produce your SPA. But the document automation tools only really work if you've got a really set precedent that you're using. Actually, you're on cillary documents that come with your sort of set precedent. Sometimes they're negotiated right up until the end of the deal. And then you've got to roll out 40 officers certificates or share transfer forms. So then you have a tool like Office and Dragons, which you can then do that multiple automation with, which you can't really do in some of the other document automation tools. And then within that, you'll be doing project management and you'll be doing issues list tracking or document list tracking. So therefore, you have a tool like Logatics, which helps you manage that with third parties and thinking about those things. But you might then want to, if you're doing the due diligence report, show it in a more visual way. So you use something like StructureFloat, bring that all together as well. You might need to do the reviews of the contracts. So you're using the like AGPT that we've built internally or you might use like Lagora that we're doing for bulk contract reviews. And then you might actually be drafting and negotiating the SPA. So you need something like an internal sort of word add in where you're helping you with your drafting or a tool. So we built something internally called Flag where you're like flagging issues within it, not AI, but you're flagging issues within it. So you can start to see why you've got like a varied tech stack. I'm really looking at those point solutions. So I mean, I've given you obviously some examples there of tools that we've got. So it's a mix of point solutions, some enterprise things that we've done with IQ and AGPT and Microsoft co-pilot and then other tools that you sort of build as you go. And when you're doing things like Autologics, so it's a real mix of tools. So is that a kind of firm-wide kind of technology kind of vendor agnostic type approach, is that the one that you adopt? Yeah, so we sort of have like a buy and build. So we don't have buy or build, we will build and develop some things ourselves, we'll buy some things and then you've sort of got like that meld bit as well, which is, oh, you've got these tools. Actually, how do they work together? How you might have bought it for something specific, but actually you're starting to see other practice areas pick it up and think about how they reduce it. I mean, two examples of that is bundle docs, which we bought obviously to do bundles in litigation. But then the trainees using bundle docs went into other seats and they were like, well, actually you could use bundle docs to manage large contracts and loads of schedules and things like that. And if you take Logatics as the other example, their big use case when they first came out like others was managing conditions, precedence in banking transactions, quite like a niche area. And but actually again, once people started using it, it was like, well, you can do issues list, you can do documents list, you can do project management, but what, fire it. So you start to see sort of that build out as well. So that's that mix of buy, build and meld. How do you, how do you kind of square that circle then? Because I suppose if you've got all of these demands from all your different sort of practice areas for all these tools and some of them perhaps aren't quite right at this stage for that particular use case, how do you manage that? What do you say to people? How do you work with the kind of suppliers to solve that? We will be, like we said, we generally when we buy something, we buy a small number of licenses and we roll it out when we see what happens and then you start to see that demand grow. And one of the really big things that we try to do as a group and as a firm is have really close relationships with those suppliers. So we can feed into their road maps and so we'll have quarterly reviews where you all get together and they tell us what they're planning to develop. We feed in sort of feedback and stuff and we've done this from really early on with our companies. So that we can be like, well, it's working really well in banking, but real estate would like it if it did this. And then that goes on to their road map. If other firms want it as well, obviously it goes further up the road map priority list. So it's that it's constant communication and feedback both with our lawyers and our suppliers to make sure that they're bringing in the new technology sort of updates and stuff that you're seeing as well as understanding sort of the little, so it's really interesting because little block, little bits can block use. Like if it just doesn't quite do that. So CP lists as an example, we always struggled with CP lists because depending on the facility agreement that you use, the numbering is different. And if you can't amend the numbering, then people won't use the tool. And that's really, it sounds really little and like a small inconsequential thing, but it actually stops you. So talking to the suppliers about that type of thing and those needs means it those small tweaks help with engagement and adoption. And which then do you see as having made the most kind of important measurable impact for and with your guard art of all of these kind of solutions? So it's quite interesting because there's pretty different reasons why we would say them. So office and dragons when it came in literally just hit the ground running and the minute our junior lawyers found what it could do, absolutely used it. And like I would have been here all weekend and I've done it in 10 minutes like that type of thing. So we were we absolutely started to see that use. And high q has been in has been like a fundamental platform for us for years. And so that has a huge impact. We've been able to do things on when pieces of work using high q that we would just never have been able to do before. So it's like really different impacts of like a really long term platform that you've got versus a one off thing. And then I think we I when I do these I can't not talk about AGPT, which is our internal tool that we've developed. Again, because we almost made it frictionless there's because it's internal, it's secure, all of that type of stuff. We could go out to everyone and say you can use this tool. The adoption of that has just in some ways been mind blowing like every time I look at the stats of the usage. I'm like, wow, how have we got like I think it's like 85% of the firm have used it 2000 prompts a day when we came back after Christmas that first week. It just went back up to normal usage numbers, which shows that it's embedded in the firm. So so it's quite hard to pick one that's had the biggest impact because they all sort of have in their own way. So your own internal tool then how was that developed and what's powering it? Yeah, so obviously when all the GPT technology came out, it became much easier to build things around that internally because it was open sourced. And we knew we didn't ban chat GPT from the firm, but we were very clear about what you could and couldn't put into it. And we've absolutely seen over the years if you put restrictions on tools for various reasons, it makes it a little bit harder and people either don't use it or sort of shy away from it. So we were like, well, how do we do this in a secure way? So like most people who built their internal tool, it's built on our as your instance using open AI models. And we sort of built it to test and try and experiment and see what might be possible. And I'm built it quite quickly and then got some people trying it and then suddenly it became really clear. It'd be useful sort of across the firm. So when it first came out, it was like ask questions to it. I think we could upload one document of sort of like a small length document. And then people have gone, well, could we have this a bit like we do with our suppliers? Could we have this? Could we have that? So we have an internal development team who take those and work through and sort of update to LGBT. So it was updated yesterday with some new bits, so sort of slowly iterating it. And what we said probably had a much bigger impact than we imagined it would. And what are the main use cases then that lawyers are using it for? Yeah, so we can't see exactly because we don't see what the actual prompts are going in, but we can classify the prompts. So it's non-legal drafting. It's like one thing that we're seeing and then legal drafting, but non-legal drafting is more translation is one of the other bits. And then document review, whether that's like reviewing a one-off contract or a legal document or putting articles in or case law in and asking it to do summarisation and various things. There are other some weird and wonderful things that people are doing, so I don't know coming up with poems for leaving cards and things like that, but that obviously wasn't the main thing for it. But yeah, it seems to be that like helping tight and up drafting review summarisation and translation of the bigger ones. And as you said there then, so why wouldn't lawyers just use ChatGPT? Are they finding that obviously there are some sensitivities around what you can ask it and that blade, etc. But presumably your in-house tool is actually performing better for those non-legal, for example, drafting tasks. Yeah, so one, it's because of the client and confidentiality piece and we can, everyone can put whatever they want into into AGPT. And we've got some prompts in the system prompts which make it more specific to AG and what we're doing. And then within that we've built prompt libraries that people can use as well. So where teams have worked together and gone, these are the best prompts to get the best output of this particular task. Let's put it in the prompt library so every person in a team can access it and that doing those developments have helped sort of people using it. And it sounds then like adoption hasn't really been a problem for you here. No, not for AGPT. And I think so it's really interesting because I think in some ways it's turned technology purchasing on its head. So obviously we've spent many years saying you don't just buy technology for the sake of it, you'd have to do what's the problem you're trying to solve and things like that. And we still have those, we still have some really big, naughty things that we want to solve. And we're still working like really closely with our lawyers and how you do those things. But you've got other areas where the volume and scale would never have been enough for them to come with a business case to say I'd like to buy this piece of technology. Well, that technology doesn't exist because it was the market wasn't ever there for it. So you'll have like the pensions guys and the tax guys. Some of the stuff they would want to do just was never at that volume and scale that it would make sense. But with something like AGPT, they can create their prompt libraries or the GPT that they want to use for it for that tool in a way that you would never be able to do before. And you're almost going, think about what it can do. People go, could AI do this? And we're like, I don't know, let's see. Let's ask the question. And because if you think about old AI and machine learning where you had to teach it, what to look for, it obviously took a lot to get somewhere to be able to use it with these tools. If you ask the question and it's not quite right, you amend the question again or you amend the prompt that you're putting in or how you're asking for it. And that ability, I think, especially for how we do legal works sometimes is you might ask a question. And then you go, actually, I want to know a little bit more about that. So you deep dive into it. You weren't able to do that with the tools before. So that's really driven different uses. And we had some use cases in mind and we produced a use case wheel of what we thought might be possible. But it's not until it's in the hands of our colleagues where they're like, I've done this with it. And they're sharing that that you go, didn't even think about that as a use case. So keep going and sort of share those ideas. So it's sort of from an adoption perspective, just slightly turned how you would normally do it on its head. And it sounds like then you almost you can do all of this work yourself in house. Yeah. Does that mean that you were sort of shy away from using those sort of broader AI tools like a path or whatever it might be or? Yeah, we obviously cost comes into that. Which I think is a really big element of this. And as we all probably know in the conversations you've had, there's a big focus on what's the ROI of these tools and how do you show the value and all of that type of stuff. And what's been really interesting when we've done surveys and when we've looked at things is some people are saying, it's not really saving me time, but it's adding value or it's better quality or it's doing this type of thing. And so for us, it was always that sort of element of how much do you spend and how much are you getting sort of back for it. And it's not always cost savings. And I think the really interesting thing for me is some of the stuff that we seem to be saving time on isn't stuff. We would normally charge our clients for it. It's sort of all of the other elements around sort of how lawyers work and how law firms work. So when you're having those conversations, you're really looking and saying, if a license for that tool is X, what is it worth us rolling that out? You have some concerns with third party tools that do your clients allow you to do it. Have you got consent and all of that type of stuff? Whereas when it's within our own instance, we have that sign off. And so again, it's removing some of those barriers to adoption by having it internally. So those are some of the balances. We probably can't develop as quickly as those tools and there's probably some functionality in those tools that's really interesting. And then might come a point where you get that tipping point and you go actually it now makes sense for us to do that. And with all of our learnings that we've had over the last couple of years, we can really then monster like a rollout of a tool. But right now, we're still still doing that balance. We do have, that's not to say we haven't bought tools, so we have bought LaGora. And for us, that's our bulk document review tool. And so we have AGPT more as your assistant and then we have LaGora as your, where we can upload thousands of contracts and do that review. So we have that and we also have tools that help drafting in word because I think the other element to this is buying an assistant doesn't answer everything. So if you're drafting a live document in word, you need something a bit different. If you're reviewing a thousand contracts, you need a slightly different UI. And if you're asking questions or reviewing a one-off contract, that again is different. So you're sort of making that piece. So it's all about ROI that isn't always money and cost savings that there's other bits to it, but getting that balance right. So what do you think in a kind of perfect world? We would all agree that the ROI on these tools is. It's honestly, I think it's really hard. You could probably go, well, how many people, how many times have people used it and put a cost to that? Because time savings and that. And you go, well, on average, is it saving a person who has access to this 20 minutes a day or not? Like, you could look at it that way. I think there probably comes an element that it's the cost of doing business as well at the moment. Like, there is an expectation you will have these tools that your lawyers will want to use them, your colleagues across the business will want to use them. And if you don't have that tool, what's that saying for staff retention and what you're doing? The other element to me is that we've absolutely seen it be able to drive almost new things that you hadn't had done before. So, and almost technology has been like this from the beginning, which is always quite hard when everyone goes, oh, what's the cost saving to clients? Is we would never have been able to do some of the work without technology? Like clients would never have paid for it manually. So it's quite hard to really calculate an ROI on it because you're like, well, we would never have done this work. Or you could say that it would have cost X manually, but no one would ever have paid the X number. It maybe wouldn't have got done or you would do a sample. So it's sort of lots of different bits that all add up together to show whether you're adding value in that ROI. Do you see clients pushing for lower fees then because of this technology? They're asking. So I don't think they're saying it needs to be completely lower, but they're asking what the impact is on their fees. I think there is probably an expectation that it is saving lots and lots of time. And I think on some things it is. But we also have very many clients and also the conversations we have with our insurers and things like that saying we still have humans in the loop. We still have the cost of the technology on top of that, the cost of managing that technology and getting those prompts right. So if you come to us to do a large contract review, we have some stuff built in absolutely. And we have some prompts that we've done for our standard contract reviews. But for that particular client on that particular deal, there might be some specific things they need to look for. So you need to make sure that it's working for it. So all of those elements also go into that cost and what you're getting for your fees. So we've had a number of conversations around pricing and how you value legal work and just if you get it quicker is it more valuable and all of that type of stuff. So I think people are still coming out in the wash. I think there's an expectation it will mean lower fees. But I'm not always sure it's going to translate to that. And how are you dealing with the impact on the billable hour model? Like I said, it's still I think we're still figuring it out. And I think most law firms are we are having conversations around we've got a couple of specific things that we can do with clients like workstreams where you can do them on a fixed fee basis and you can be quite open about that cost the cost of the technology as well as the cost of the time to do it and what that would look like. The other conversations were saying, well, what is the impact on this? And as much as quite interesting because I've been doing this for over a decade. So it still feels like GPT is quite new, but everyone wants everyone to have the answers right now on it. But I think some people are any still just rolling out their first Gen.A. I tools this year. So I think it's going to be a slightly longer conversation about pricing and fees. And I think it's going to be a two way conversations with clients on that. And like if you how do you come up with and no one's ever been able to solve this even before sort of Gen.A.I. and technology, how do you value legal work? How do I decide whether I want to pay £50,000 for that thing as an in-house lawyer? And almost like when you get your decorators in and you ask them how much it's going to be to plaster your room, you have a bit of an idea of how much you want to pay. So sort of those conversations really need to happen, I think, over the next sort of year to get that right. And I suppose you absorb the cost of all this technology, you're not passing it on to clients directly, let's say, there's no premium to the fee, right? There's no so it depends. So there will be certain tools that I use by lawyers on a day-to-day basis that will become just part of how we do that delivery. There are other matters that we do where the client, where we're in conversations with our client and we are specifically building something for that. And that's, and I don't mean developing from scratch, but it's like you're, you're sort of knitting all your tools together to be able to do a particular task. And actually, if you're put in 10,000 contracts through an AI tool, there's a cost to put in 10,000 contracts through an AI tool. So there is also, so we have elements of those where there is a tech cost, and that's the actual cost and some of the licensing, it's some of the time to set up those things and manage them going forward and things like that. And that's what we're starting to see on those specific and bespoke things. The other day-to-day would more get factored into sort of how you're pricing. Sure, so I suppose where a project has a sort of clearly defined scope of work that is not expected to kind of materially change, then I suppose you can factor all of that in and kind of quantify that development cost. Yeah, right. Yeah, yeah. It makes me think so. So recently, you know, you've probably seen this as well, but it seems like some firms are moving to a stage where they're actively looking to, you know, productize their technology and actually sell it to clients on a SaaS basis. I think, you know, A&O and McFarlane have been talking about this type of thing this year. What do you think about that? So we've had a couple of those over the years where you sort of build your legal knowledge in once and then you sort of, almost like as if they were buying it from a technology company, but I think it's also really clear to say we're law firms and not technology companies. So actually, that product management piece that comes with that, the management of a SaaS contract and all of that type of stuff is a bit harder to do as law firm. I think we've got something quite niche, it makes sense to do. We've sort of pivoted away from it and we're doing it more on a legal services underpinned by technology. So we've got a couple of products that we have a legal front door and we have like a property management tool that people can buy on a subscription basis. But on the whole, it's more how do we sell our legal services that are is powered by technology. So and sort of that element for us is when regulatory change happens. So if you take LIBOR as an example or Dora recently and there's other things coming down the track, can you sort of make it less bespoke than you may be some of the bigger client projects that we've been talking about, but you can sort of build in a process that you can then be really comfortable with what the fee is, what the process is and how the client works for it. And that's probably where we're putting more of our focus than building a tool and selling it on a subscription basis. And Kerry, where do you think that the most disruption will be from this new technology, particularly AI? So there seems to be sort of two takes as I see it at the moment. One is saying that in-house teams might be most impacted because businesses see those in-house legal teams as a cost center and they want to reduce cost as much as possible, use technology instead of lawyers. And another one seems to be that sort of mid-tier firms that do a lot more sort of project-based work that perhaps considered a little lower value might be impacted. You know, with the kind of higher-tier firms focusing on that real big ticket kind of, you know, big transactional stuff. What do you see there? Is it a bit of a cop-up to say all of them? Because in some ways, some of the conversations I've been having over the last two and a half years obviously it just accelerated from what we've been doing in the last decade is this technology is affecting all work. People have already fed up of me saying I sit here and I really struggle about how to advise my kids on like what they should do at university and what jobs they should do because I just think we're a place where every sector has been affected by this technology in a way legal technology sort of sat there on its own and we were a little bit slow to like everything that was going on in the market. So in some ways I can absolutely see in and we speak to lots of in-house teams about how they're going to use technology and in some ways they're in a really good place for everyone talks about structured data and how you need to use that for AI. If you are an in-house team and you have your own playbook you're in a really good place to be able to use these tools at the moment. That's a bit harder for us as a law firm because we deal with multiple clients with multiple things. We don't have a playbook as a law firm that you might have as an in-house team so I can absolutely see. If you think about how you might do intake and triage of the work coming into an in-house team even just managing that piece thinking about playbooks that you might do and stuff. But it takes quite a bit of the moment to get that up and running. You need to have your playbook simulation. You need to have people to understand how you would build that into a technology platform and things like that. So I think there's going to be an impact. I don't know how long it's going to take to get there and what that would look like. And in-house teams say to us all the time they would like to do the really strategic work as well. So you always have conversations with them about you could use technology to do some of the work. Should that still be done in-house or do you give it to you outsource it and you do some other bits around it. So I can see that the interesting bit for me on the transactions for law firms. I sometimes think people put things in boxes and buckets that it's not that black and white. So even if we are doing really sort of strategic emanade deals there are still aspects of those big transactions that technology could help with. So even if you're doing like a precedent setting matter and you've got your lawyers doing that really valuable piece that they that's what their brains are for. That doesn't mean that the cheat diligence aspect can be done by technology or you might not like use logatics to do your issues list whilst you're doing it. So I think there are some elements that are probably more commoditized or standardized that you could do. And I don't think that always is low value either. I think there are certain aspects of what lawyers do that could be in that box. But it doesn't it's not to like sort of the extent that there's nothing in the other end of that. I think sometimes it's always been is it commoditized or is it more like rocket science but even if it's rocket science as a bit you could do. So that's why I'm saying it's a bit of a cop out to say I think there's impact everywhere. Do you think junior associate should be worried about the impact of this technology? I think if you keep your mind open you're curious and you embrace this I think that puts you in a really good place as a lawyer to see what's happening. And actually I think the role of a junior lawyer could look different and that should be something we embrace. I think if you think about technology and what we've been saying for the last number of years and what is it going to do and everyone always asks me what's the effect on junior lawyers before gen AI? What's the effect of document automation on junior lawyers? Are we going to have less? I think you're just opening up more demand and we're like we're bigger than ever. You've got more people coming in than ever. You've got different roles coming in. So I think actually people should see it as a real opportunity to go what does legal practice look like now? I think when gen AI first came out and GBT sort of started to be in talking about people likened it to when electricity was invented and how we all sit here 100 years later and go that must have been amazing suddenly you had light and it's like well actually they had gas lamps so it replaced stuff that was already there to start with and it wasn't until there were developments of television radio computers that knew things started so it's almost like are we in that element now? You're using gen AI tools to do the stuff you did before but what is it allowing that will be coming that we haven't even thought about yet? I think if you can be part of that wave as a junior lawyer that sounds really exciting to me. So I think there's lots of opportunity. So what does the law firm of 2035 look like for you? It's quite interesting. When summer of 23 we wrote a board paper on what the short medium and long term impact would be and I think our long term impact was five to ten years. Some of that's happening now. The pace of change is so much so in some ways I can't even imagine what 20 to 10 years would be because of how much it's changing but I think it will be we won't be talking about gen AI tools in the way that we are now because there will just be there. Everyone will have Microsoft co-pilot, everyone will be used to doing that type of thing and there's almost in the same way that word and email became sort of like the core part of how you do things. I think there will be a there will be more fixed pricing I think than there would and there will be more open conversations about what that means. I think if we're being really honest fixed pricing over the years has still always been a function of the hourly rate. You'd guess how many hours it's going to take, you charge your hourly rate and that's how you get your fixed fee. I think there will be more conversations about value and if I can get if you're a tenant and I'm negotiating your lease and by using these tools, not just gen AI but other tools, I can get you in two months earlier. You're going to get two months more takings and profits so actually where's the value and how do you do that? So I think I think you'll start to see that I think some aspects of what law firms do will be done by technology and no one will question it. If you think about like e-discovery and bundling and data rooms and how they changed, I think you'll start to see some of those more admin tasks go there. Yeah, so that's probably as far as I've got. So you don't think then that the sort of people pyramid will get a lot kind of narrower. So you'd have your partners at the top and then basically supported by a kind of handful of associates relying on AI. It's not going to transform that. No, I think there is a huge amount of untapped work out there because it was so the number of conversations we've had over the years with clients even if and it depends if you want to do this type of work or not as a sort of as a firm but loads of our clients have not digitalised their contracts. Like they are on share drives or they're in draws or they're in safes and they're not digitalised and no one's ever done it because the manual effort to digitalise it wasn't worth it. The manual sort of the manual effort and cost wasn't worth it but actually now we're are we able to digitalise those contracts to give you more of that insight to lead to that type of thing. So and we're already seeing pieces of work that I just don't even think would have been possible to think about that we're doing. So in some ways I think there'll be more demand and more work and we have to think about how we do that. So at the moment I don't know and I'm very happy to sit here and have this conversation in 2035 when it all looked very different but I think there's more than enough work out there if we can do it and using technology you can grab some of that that is just sitting there at the moment. I would agree with that. That seems to me like the best take actually an increase in demand actually for legal services. Yeah. Okay great Kerry so just to just to finish up then I'm going to ask you a few sort of quick fiery type questions if I may. The first one is just to ask you you know do you have one innovation in legal tech that you're most excited about at the moment. I think it has to be how we're going to use AI to do document review and then what the output's going to be on the bit like at the end of that and the reason why I say that it sounds quite probably sounds quite boring because we've been doing sort of AI on contract reviews and document reviews for a long time but before it I could always I always got this sentence from like people we demoed it to like it was when you first looked at old AI I shouldn't call it old AI should we but like machine learning it was always there was always that sense of disappointment well it didn't find the weird stuff and I'm like no because you can teach it you can't teach it the weird stuff like that it's that type of thing so there was always like it got you so far and same with playbooks and things like that got you so far but it didn't really get you where lawyers wanted it to be like it think like lawyers and I absolutely since we've seen some of the tools and what we've been doing around GPT technology the ability of this to review documents at the output and exactly what lawyers want there's so much opportunity there to really speed that piece up and then it really gives you that option to go well what if I was doing a GD report for an M&A deal what does that look like what does what am I giving my client at the end of that piece am I giving them data in a platform because actually they need to integrate that new business that they're buying and am I giving you a really interactive DD report that shows you timelines of integration and things like that so I think for me there's been so much talk around DD for so long and we've attempted it other firms have I think that I think there's a real opportunity there that to get something right and make it really valuable. I always think just on that one you know even if this technology came back with this fantastic summary whatever it is of these particular suite of documents and you thought wow that that is fantastic you would still need always forever more to go and just verify but that is the case right and how do you I don't know how to solve that because if you've got you know hundreds of pages of documents that you're looking through you know has it missed anything you know understand that you can just sort of pull out the relevant provision and cross check it okay fine but how do you ever know if it's missed anything I don't know but then how do you know if humans have missed anything as well and I think the tools are definitely getting better at being able to do that spot checking and going yes and then you go well how many times do I need to spot check it and so then feel happy with it but then that also leads to those conversations with clients about well what risk profile is there how much if you think about M&A at the moment generally what you do is a sample I mean you take a materiality threshold you're looking for particular things and you check a sample so there's a whole load of other documents that aren't really looked at in it so actually where's the risk threshold if I did if I could review all of them and we're about 95% accurate where what does that look like and how our clients prepared to accept that what and all of those types of conversations so I think that's going to be really interesting over the coming years as well because it's not going to change where you know if the law firm gets it wrong they are the ones on the hook to the client you know how do you then mitigate that risk versus the technology you're using I yeah that's a challenging one I think you have the open conversations with clients and you say if we're going to do it this way if it has missed it where does the liability sit and you have those conversations and we are we are having some conversations around work streams where we say if you are prepared to take that output that hasn't been validated then this is the cost this is like the risk within that and then so you can almost like take certain parts of what you're looking at and say for this piece we'll review these elements and spot check these but for this piece you're prepared to take the risk at this level and and we're really having those conversations at the moment what's the best piece of career advice you would give to junior lawyers today Kerry um interestingly I wrote a post on this a few years ago and it still sticks for me it's to be brave and to be curious like I I remember doing a course when I became sort of slightly senior I did a course and they talked about career gates in this course and I'm like I really wish someone had told me this when I was junior which is there are elements in your life or parts of your career that if you go through that gate is going to change something in your career and ultimately that's the thing about being brave and saying yes to things so chair that meeting go on that pitch say yes to that client event that you're going to and actually you didn't you don't sometimes don't know it's a career gate until you've gone through it and you look back and and I think people really need to understand your career is your career and you're the you're the person that cares the most about your career so you need to take sort of the balls by the ball by the horn and and just sort of push yourself forward and part of that as being brave and in this world reading being curious understanding how it's going to affect things all comes as part of that as well for me if you could change one thing about the legal industry what what would that be do you think um people have probably heard me say this a few times I still think we're in a place where risk prevents innovation a bit so if you think about um some of the terms that we have with clients around the use of technology there are blockers in the way um so we we would need to get consent for certain things or there's like prohibitions on things or there's notices around like you need to tell us every time you're using AI and I think that stops law firms actually using the technology so if we could have a real really good conversation about what those risks are why clients are worried about it what the impact would be if we could use those tools then I think so at the moment if I had a magic wand there would be something about um sort of the consent from clients to use technology that I would try to sort of smooth out because I I think we've got some tools sitting there that we could absolutely apply to legal work and to have a really really big impact but we can't so that would be my why would change if I could okay last one for you then Kerry um I normally ask this one so so if you could recommend a book legal or non-legal that shaped your thinking what would that be I don't know if it's a bit cliche but the innovator's dilemma probably um I think that I've especially noticed over the years and and I think technology has moved on so much it's making me think differently now I think we've always tried to use technology and innovation and new ways of doing things to almost produce what we've always produced so it's like I want to take I want to get where I was always going but quicker and you're a bit like well now I think there's a real question around that and I think the innovator's dilemma basically says you can spend ages doing that bit but you should be really looking at the disruptive models so I think we should really be going what does this mean for us like how is it going to change because we can't yes we could use technology to you almost have been trying to force technology or force people into technology to still give us the same output I think we just need to completely change that mindset and that the innovator's dilemma is all around that. Very good okay um thanks a lot Kerry for your time that was that was great. Brilliant Thank you very much.