Matthew Cloke, CTO at Endava: Building an AI-Native Organisation and Scaling Adoption

0:26Welcome to Data and AI Mastery, the podcast where we bring you cutting-edge insights, practical advice, and inspiring stories from the leaders shaping the future of data and AI across the globe. I'm your host, Raoul Gabriel-Urma, founder of Cambridge Spark, the leader in transformational data and AI upskilling, career development, and progression. In each episode, I will be diving into real-world case studies of companies harnessing the power of AI to drive innovation, reduce costs, and create new business opportunities. So whether you are an aspiring data scientist, AI engineer, or seasoned executive, this show is designed to give you the tools and knowledge to stay ahead in a world where data is transforming every aspect of business.

1:11Stay ahead, stay inspired, stay masterful. Welcome to Data and AI Mastery. Hey Matt, good to see you. How's it going? It's great. Thanks for having me here today, Raoul. Good to meet you.

1:31so hey matt what a fascinating career you've had so you know nokia ubs now over 10 years at endava where you're the cto so that's really exciting so maybe to kick us off can you tell us what the cto role entails at endava that's a great question i mean whenever a cto walks into a room nine times out of 10, someone says something like my laptop doesn't work. And you're kind of like, really? No, but inside of Ndava, what does the CTO role entail? So I'm a member of the executive team, I report directly to John Cottrell, who's the CEO. And effectively, I wear a number of different hats as the CTO.

2:14So I am responsible for our internal IT systems and platforms. So I have a IO who works for me and therefore I'm like every other person who's probably listening to this podcast going what's the impact on AI and technology on our internal platforms and tools etc etc the other part that I'm responsible for is as an IT services company what is the skills and capabilities that we need to have available to be able to provide a service to our clients so that can be everything from you know how many Java developers do we need what skills do they need with python developers etc etc and then the third part of it really is around proposition development and how can we demonstrate to clients from a commercial perspective that we know what we're talking about that we can help them solve the problems that they're trying to challenge so multi-faceted role but as you said i've been here for 10 years i joined as a simple enterprise architect and have kind of worked my way through to today and the last point to make is I also see myself as an ambassador for the 8 ,000 plus engineers that we have inside the organization and really want to kind of be an advocate for them and their voice because we're a very passionate engineering-based organization.

3:32That's wonderful. And you've named two of my favorite programming languages, so that's way off to a great start.

3:42So I guess maybe to kick us off, you know, I was going to ask you the question, who has the ownership of the ai agenda in a large organization like endeavor sounds like it's it's you the cto yes so i i can talk about the probably the the most boring part of it which is the kind of like what's the governance structure of it and then i can talk about the more exciting part of it so we're a publicly traded company so we we have a board as many organizations do there are many committees that surround that board one of which is the control and policy committee or the cpc and they would own things like our cyber security policy and those type of things what we did around 12 months ago 18 months ago was we myself our legal team and a couple of other people wrote an ai policy so what was in darva's position on ai what did we expect from people who were going to use it, threats, risks, those type of things.

4:43So a very comprehensive policy. And within that policy, we suggested to the Control and Policy Committee that an AI committee was created, which brought people from across our organization. So not just me as the CTO, but someone from the legal team, someone from the finance team, someone from our commercial teams. How did you select those people? Like why it sounds like it's cross-functional, which is really interesting we went cross-functional because at that time when we were creating that policy and the general philosophy back to my comment about gpt kind of being the dinosaur moment with the meteor streaking through the air we've decided that we're very much going to lean into everything that ai can do and therefore from our perspective it's like this isn't about just the you know the back-end functions of indava embracing ai it was about the fact that actually it changes the way that we talk to our clients about what our propositions are it changes the way we deliver those propositions to our client and therefore when we looked at that AI committee the decision was well we need to bring those different voices into the room to be able to get their perspective now what that committee does in reality is you know it meets on a regular cadence and what it's doing is it's creating the environment which allows people to experiment with AI technologies, not using client data, being very secure and safe about how they do it.

6:11But when we then turn around and say, this is a tool that we believe everybody in the organization should use, or this is an approach that we should use with our clients, the AI committee is responsible for approving that technology once it's gone through a security review, once it's gone through a kind of like legal review, et cetera, et cetera. So that committee is very much the governor within the Endava organization. It's great. What it does mean is when people come to us and go, hey, I've got the latest tool or endless stream of people coming with new technology, there's a very direct place that we can go to to get that approval.

6:50So that's the governance side of things. The other part I would then say is, but how do we then become an AI native organization? Because you're not going to do it by having a committee at the top going, go and be AI native, you actually need to have it. And underneath the covers, we have a program internally, which we refer to Keystone, which is looking at every single part of our business, saying, what does it mean to become AI native? So where are we today? How do we see AI changing that particular role or activity? And what do we need to do to be able to move our people, our propositions, everything towards that AI native vision?

7:28And that's done underneath the Keystone program. And to round it all off and bring it back to your first point, I'm the chair of the AI committee at the delegation of John, the CEO, and I'm the overall exec sponsor for the Keystone program within the organization. fantastic you beat me to it i was gonna ask who chairs that committee right ultimately so that's really interesting because some organizations you know they've appointed a chief ai officer you know that has the mandate about the ai strategy make it happen but we're seeing challenges of this model too right one can argue that ai is everyone's concern it's not the one person one team it's cross-functional like like you point out so i find it really interesting as a model that you have a committee set up cross-functional stakeholder and you as the cto you're ultimately the chair of the committee to kind of like bring all the voices together so if that's the case matt could you talk us through a little bit how is that committee setting uh targets for you know adoption or you know even better value creation from ai how does that work out so the committee doesn't take that responsibility it's a lot there to ensure the policies are in place that everybody understands the risks and the benefits of doing things and that's the responsibility of the committee the keystone program of which i'm the exact sponsor is then when we get down to the kind of like nuts and bolts of what you're talking about which is you know what's the adoption how are we doing it how's it rolling out etc etc now on that point it's all well and good having someone like myself sitting at the top of an organization going well we should be adopting AI technology it's going to be hugely impactful on our business it's going to do all of these efficiencies it's then a good old-fashioned change process so how do you make that happen so what we did was again about 15 16 months ago now we decided to invest in chat GPT enterprise to everybody inside of the organization and the way that we rolled that out was using something called the champions network so what you would go and do is you go and find the person in the finance department who was passionate about using ai could see the benefits of doing it and then training them along you know using our people using our skills and our expertise but also working very closely with open ai and so we built out this champions network in all of the different parts of the organization but again even that wasn't enough what we then discovered was the fact that you know you'll go and find that you know Jill in accounts has used chat GPT in a way that no one was thinking about beforehand and then everyone's kind of like staring over their shoulder going how did you do that that's really clever and what we did is we purposefully went to go and find those people within our organization and we call those people the AI heroes and they're really kind of like if you like the influencers the people within the organization that other people listen to so you need that kind of like groundswell from the bottom up in terms of people wanting to use it being able to see the benefit of it plus the top levels support but all of that together and where we are right now is i think we probably have about 85 to 90 percent utilization of ChatGPT on a daily basis across the organization.

10:59We have several hundred GPTs written. And for most people, they can't actually think about doing their job without first going into ChatGPT and using everything that's available to them on a daily basis. Wow, so many nuggets to unpack, Matt. So I love this concept of AI Champions Network because, you know, speaking of your clients, it seems to be the model that's adopted, is identify those change agents in the organization, those people that are more keen than others, train them up so they can kind of cascade the sort of culture internally. But you've described something unique here, which I hadn't heard before, which is the AI hero.

11:39So for the audience, can you help understand how did you identify those AI champions in the first place? Like, how did that work out? You know, kind of bootstrapping the network. and then who gets to be an AI hero? Champions kind of emerge from what would happen with a traditional change program where you would go to you know the leader of a department or or a team and say who within your team is interested in doing this you know here's the opportunity you might have heard about getting training this is an opportunity for you to work with us and OpenAI to develop your skills and you will usually get people who will volunteer within that position but invariably the people who get picked tend to be the more senior people inside of the teams and you're kind of like okay that kind of makes sense because they're the people who are going to think about the the nuances or the problems of rolling it out where the AI heroes come into it is really this idea that as much as you are trying to push something downwards and you're trying to you know model the way in which you want people to use AI when you see someone using AI in a truly transformative way you want to celebrate it you want to lift it up you want to show other people the art of the possible and the thing you've never done beforehand so rather than that just turning into a kind of the champions would report on good use cases we actually wanted to lean into it lift those people up and go hey look you've got bogdan in romania or you've got a frederico in bogota who is doing something completely transformative and they sit amongst you they sit in your team you can go and have a coffee with them you can go and sit next to them on the desk and you can go and see how they're using it so for us it was that kind of like different ways of breaking away from a kind of structured way of deploying ai and embracing a more ground-up organic way of rolling out the technology we've got 8 000 engineers inside the organization they tend to be curious they tend to want to know how to use a piece of technology so maybe we're predisposed to being better able to identify who those heroes are but equally and again to have a long monologue the biggest and first mover users within our organization who successfully modeled champions and ai heroes was in fact our legal team and our legal team are literally the vanguards and the leading lights in terms of how AI can be used to transform their jobs.

14:17Love it. That's fascinating. And I like the point you make that clearly Andaba has already a culture of innovation and generic mindset, curiosity. So it sounds like, you know, those attributes have been really helpful to speed up adoption, maybe more so than other organizations right yeah i mean we did some research recently with idc and it kind of thankfully backed up the conversations that i tend to have with kind of like senior leaders and then different people within an organization and it basically said the more senior you are within an organization the more you could see what were the benefits of deploying ai technology within an organization so you know effectively the c-suite of an organization were kind of like bought in to the promise of what could be delivered and we're wanting organizations to adopt it but as soon as you drop below the c-suite and as you go down inside of an organization the trust of the technology and the view that it could be transformative rapidly drops off you're right it's not an easy problem to solve now man i'd love to take you back to the the point you made right so did i get it straight you said there's been sort of 80 to 90 percent sort of active users adoption of ChatGPT since you won Enterprise.

15:32That's great. You invested in this big change management program with champions, heroes, and so on. So can you walk us back maybe today, I assume you had to make the investment case, right? We're going to spend millions on ChatGPT licenses. And I assume the board, your CEO, what's the ROI? Why should I spend all of that money? Can you walk us through back to those days a little bit? And what are you saying today, I guess? It was very interesting because full credit to John. He is an advocate and a visionary in terms of understanding what is the potential impact of AI being deployed inside the organization.

16:16and he was in the same meetings with investors and clients and others being told that Endava was fundamentally going to go out of business because AI was going to come and do our job. I think where he was at the point that the business case for deploying chat GPT enterprise came along was what I would call the traditional clipboard and pen model which was okay so you're telling me that I'm going to spend this amount of money per person per month, etc, etc. How much more productive are they going to be? How are we going to measure people? How are we going to understand that these things are impactful, etc, etc.

16:56And basically, what I did when I wrote the business case is I did two things. I use ChatGPT to help me write the business case. So the model of rollout, the financial, when we're going to pay the different costs, how is this going to go? what was it going to look like over a two-year period you know what was that as a impact on our internal costs etc etc i used chat gpt extensively to look at spreadsheets build the graphs and effectively build the the entire business case for me and how was that experience well i mean we're talking about chat gpt 3.5 and early versions of four so it was rudimentary it was occasionally like wrestling with a snake in a bag but we got there in the end that was kind of like the mechanics of how do I think this is you know what's the impact on it and I think there were two things that came out of it and hopefully the second one was more impactful on John's thinking so number one was I basically went from the perspective of we we already pay for a lot of coffee in all of our offices we've got nice offices nice coffee machines what I'm effectively specifically advocating for is one coffee for everyone in the office once a day.

18:10Because when you break down the cost of what is ChatGPT Enterprise over a month, and you look at it, it's the cost of a coffee in one of our offices every day. So it's kind of like, think about this as like one more coffee, which, you know, was probably a terrible mistake, but I'm a CTO, I'm not a chief salesperson. But the second thing, which was in the business case, when I spoke through it with John was around the part of the challenge we have is AI is going to enable something new that we've never measured in the past. So yes, I could give you a how much quicker can I produce a PowerPoint if I use AI, but that's not really the story.

18:49What the story is, is actually, it can help me research my clients, it can help me give a level and depth of response, which I can't do or would take me a lot more time because I've got to go off and I've got to do all of this research or I'd give the job to someone else to do etc etc so I focus less on the individual productivity improvements and more on how can you enable someone to do something that they couldn't do beforehand and and to take that kind of like story arc narrative to where we are today recently chat gpt enterprise has an enabled connectors and connectors can connect to sharepoint and they can connect to teams and they can connect your inbox and do all of these type of things so those connectors they all go through our ai committee so we make sure that everybody's comfortable with us using them etc etc but now i can go into chat gpt and i can ask a question about my inbox i can ask a question about my team's conversations and guess what i get a a message from john a couple of weeks ago which was basically these connectors are completely game-changing i've just done a job that has taken me 15 minutes that would have taken me days to complete and i'm kind of like i'm kind of like okay my my worth here is this yeah that's great when your boss tells you you've just changed my life that's always good i love i love that story matt do you mind if i unpack a little bit because it's quite rich of really good nuggets i guess what i'm hearing typically when it comes to measuring roi there's sort of the horizon tall productivity so everybody in every role is now using a tool like chat gpt and you can't like infer that you know they'll be more productive because meeting notes emails connection but it tends to be really difficult to measure bottom line impact because it's just so widespread across the organization it's you know how to attribute i guess real dollar now what i'm also hearing now is actually there's maybe more of a focus on verticalized ai you know how do you reimagine the legal function the sales function or how do you think about new customer delight through you know new products how important is that you think in in a business case you know let's say you're a CTO in another organization and they have to convince the CEO, what advice would you give?

21:24And how do you think about this idea of maybe more verticalized AI, you know, still within York? So a lot of the examples that I've talked about are us doing AI to ourselves, because I have the responsibility for all of the internal platforms and enabling people to do their jobs what you're talking about fundamentally mirrors the conversations we are having with our clients which is you know you're going to talk to an insurance company and they'll say we know that we have lots of manual processes in fact we're not even entirely sure we understand what our business process is but what we do know is it takes four days to pay out a insurance claim in this particular market etc etc and within there when we're looking at okay so how can you deploy AI technology for that client to enable them to do precisely as you said which is to you know reimagine an entire department or a business process within the organization then then you're you're kind of in this world where you're kind of like it's your original point what are you measuring okay so then guess what we used to have these conversations when we used to talk about agile delivery what is the value to the client what is the value stream analysis what's the input to the process at what point is a value if that's in air quotes released to people etc etc and then it's looking at and guess what we use ai to help us do this to our clients so we're using all of the tools that i've provided through internal technology to enable people to do it but what we're then looking at going is well what's the architecture that supports you is there you know is this a do you need an agentic solution to this problem do you need an agent space or an equivalent to go and be able to do that particular workflow?

23:15Could you reimagine a business process entirely using agents and with a human only really providing supervision over a process? What's interesting about that conversation is when you start going down that route, you find fewer and fewer clients who are really ready to embrace that level of change. So they understand the potential and they understand the hey yes i get that we could get there and then there's a lot of backpedaling very quickly in terms of but we're not there yet we haven't adopted enough other kind of like use of ai our systems are too immature we don't know where our data is etc etc so then you kind of like take one step back and you're like right how are we going to eat this elephant well step one let's sort out data governance let's look at where your data is inside your organization.

24:04Let's look at your business processes. So yeah, that verticalization route that you highlight, those are the conversations that we're having with our clients. And again, just to make this go full circle, guess what? We are having to reimagine how you deliver software systems because traditionally, we've really lent into distributed agile at scale. A team working on solving a problem would be seven plus or minus two people using an agile methodology being jars the developers or testers or whatever it might be well guess what we are now having to think about that what verticalization of what is sdlc when you have agents when you have all of these different things available to you yeah great i'd love to talk about agent shortly but if i play this back it's really some good nuggets right because successful organizations tend to practice what they preach and in this case you know you've invested in in ai internally into the platform, equipping your workforce with the right skills, using AI as a thought partner, and you're able to support your clients accordingly.

25:09So that's really powerful. So I think that's great to hear that. And I'm also hearing that when it comes to verticalized sort of use case with AI, there's a notion of you just need to start working before you can run. So it sounds like, you know, there's still quite a lot of work needed in fundamental infrastructure governance data infrastructure your pipeline have that already before the business can go and like you know integrate ai to it you know we're not the first people to use the phrase by any stretch of the imagination so the nation of crawl walk run and it's kind of like what do you have to do to start crawling then walking and then run and then a couple of months ago i was in a workshop where someone put their hand up and said i think we need to wake up before we even get to the stage of crawling you're kind of like okay so there's a step before you even start crawling which is someone somewhere inside the organization has to have that aha moment so then again we're back to those senior leaders within the organization who you know someone at the top has to go aha i think i think that this is a thing that we should be looking at and then that's the kind of like waking up moment before you go on the crawl walk run i love that i'll reuse that mac so how do you get that exec to have this aha wake up moment especially maybe traditional industry what's the is there a trick to get people to realize the importance of what's happening so interestingly for me about well just over a month ago i was invited to go and speak at a ceo conference for a pe firm so their portfolio uh ceos were at an event and went with john to go and talk about AI in general and the the single biggest thing that came out of that conversation both when we were doing the presentation and then afterwards when talking to them one-on-one was people want to understand what's the practicalness of what is being told to them on a daily basis so you know there is a hose pipe of information that is being directed at senior leaders across the industry you know if you're a retail CEO someone's telling you how AI is going to disrupt your business if you're an insurer someone's telling you how ai is going to solve your business what people are interested in is what's the practical next step and and their aha moment is going i can see other people have gone on this journey so it's not a completely fruitless exercise where you know it's like the nft revolution that lasted for all of six months it's it's kind of this is something that is real I can see and I've heard a story I'm speaking to someone who knows that this has had an impact within a business and then it's kind of focusing on what's the real thing that's like where's the itch that needs to be scratched and again another conversation last week with the CEO and it started off with a conversation about well can you come and talk to our developers about how they can use AI to be better and we may need to use Indava to give us burst capacity etc etc so a very traditional come and talk to us about that 15 minutes later we're talking about how a lot of their supplier contracts are shared on sharepoint and how many people are involved in pulling those documents down reading the contract terms then putting that data into excel then trying to mine it for better you know supplier deals etc etc and you're like well it's unstructured data you could point ai at that you can consume it ingest it and you could use natural language to say you know how much do and do i pay to supply or exit for you and then that was that ceo's aha moment because they're like oh i was thinking about it over here as this particular problem but you've just told me that this problem which i thought was fairly intractable or would require a big program of work to solve how there might be a quicker way to arrive at that point so it's the same thing as ever what's the business problem what problem are you trying to solve forget ai right now how can we use ai to help you solve that business problem super super so you know the aha movement tends to happen from learning from peers and use cases that you know you can relate back to your business but probably proven use cases exactly the worst the worst thing in the world and it happens to me because people can see my name on linkedin and elsewhere where someone is trying to sell you a product or a service and you're like this this just sounds like hot air i've got no way of knowing whether it's true false etc etc and guess what you're the fifth person today who sent me an email or tried to ring my phone to sell me something that that's how ceos feel they're overwhelmed what they want to know is this person's done it and that sounds like they know what they're talking about absolutely but wouldn't you say there's a balance between you know wait and see can be punishing blockbuster i guess is a famous story right like you wait and see and then game over so how do you balance out i just want to wait you know to see what the ROI is versus i really need to get on it yeah so i use a terrible analogy for this so you can tell me it's a terrible analogy uh about this i love analogies so imagine that we were we've invented a space a starship that could leave our solar system and go and explore a planet and we've identified that that planet is the place that we want to go to but it's going to take us a thousand years to get there so we invent this cryogenic technology we freeze everybody we put them on the spaceship and we send them off and it's a thousand years until they're going to arrive at that planet but ultimately we know that they're going to arrive at that planet there are some people who go oh i think we can improve that engine and we can do x and da da da da and 100 years later someone invents a rocket that can get there in 500 years time and you're kind of like so so do i do i now send out the rocket that can get there in half the time to arrive at the destination or do i just let this one carry on going etc and and for me that's the problem that ceos and cc executives are facing all of the time is i can wait and i can see whether someone is going to invent that better rocket that's going to halve the time to enable me to get to where I'm going.

31:30My perspective on this is the longer you wait, the harder it is to catch up with your peers who are already on that particular journey. So you've got this, you know, I've forgotten who came up with it recently, but they're saying, is AI going to replace you in your role? And the answer is no, AI isn't going to replace you in your role, but a person using AI may replace you in your role that's how people should be thinking about their organizations which is if you wait because you're waiting for something to happen to go now is the time for me to push the button and start moving on this journey you are taking the risk that you can catch up with your peers who've already begun on that journey so it's it's a it was a terrible analogy but it gives you this idea of i understand why people want to wait but i don't think it's the best advice i love your analogy matt i'm a sci-fi fan the three-body problem and so on great great great ones

32:35so matt i'll move you to a quick fire round of questions if you're up for it i am up for it go for it you know what's my favorite so what's a concerned view you have in the industry? Okay, so I'm going to be quite controversial with this one. I don't believe that any organization can finish a data warehousing project. So I've been in many, many clients and many, many industries over my 30 years, where someone at one point or another will stick up their hand and go, the solution to this problem is to build a warehouse where we combine these data sources so we have a single source of truth and you know all of these wonderful benefits will emerge from having this kind of like golden warehouse that sits inside the organization and the project starts and the project will deliver you know the first milestone and maybe even the second milestone and then it precipitously slows down and no one is interested in building that warehouse anymore or maintaining that warehouse anymore and then many years later someone else pops up and goes, well, what we need is that we need a data warehouse.

33:44So my contrarian view is whilst I am an advocate of data and structure and order and governance, projects where people say we will just solve this problem by building the one true data warehouse to solve all problems, I've never seen it work in my career. I love how specific you view it. That's really interesting. So data warehouse is more of a journey than a destination. that's cool next question is in the environment you're seeing if you had a magic wand what's like one issue you you wish you could uh immediately solve so i'm gonna start with the environmental impact of ai because it's something that is at the front of at least some people's minds in terms of what's the consumption of water what's the consumption of electricity i read a lot of papers where people go but ah but you're not measuring the environmental positiveness of deploying AI technology because if someone can do a job in one hour as opposed to five hours then that saved that amount of electricity blah blah blah but if I could wave a magic wand it would be to solve an energy problem that relates to using data and AI as a as this thing that's seen as a very very negative thing to it and and you know it probably comes up in every new conversation I have with someone about AI technologies what's the environmental impact so if I could fix that I would and if I could have one other thing that I would like to fix is clearly the foundation models got built at a time when no one really understood the value of the things that they had placed on the internet so whether it was books whether it was art whether it was music whatever it was so if I had a way of being able to recognize the individuals who contributed the copyrighted material to those foundation models through a micropayments network or something like that.

35:40I wish that that could happen so that truly creative originators of content can be recognized for their contribution in something that is so transformative within our industry. That's great reflection. Absolutely. All right, maybe a couple more personal quick quick questions favorite programming language oh it changes over time so i am going to say python i'm not going to bore you with the answer so i started as a c developer actually i started with basic then i did c then i did objective c then i did swift then i did jathon then i did java then i've done quite a few different programming languages but yeah i quite like python python is i mean it is the lingua franca of the air well so that's a great choice i call it an accidental language because it's not like anyone set out with the intention of making python the de facto ai language it just so happens to be the people who did the ai used it as their language and it's a similar story with objective c i was an Objective-C developer because I used to program web objects and Next Step computers.

36:54And then the iPhone came along and everybody wanted mobile apps and everything else. And Objective-C happened to be the language that Apple used. So I was accidentally, for a short period of time, someone who actually knew the language that all of the mobile apps were written in. But again, it was accidental. No one ever said that's the best language for programming mobile laps. Definitely. I love that story. What was your favorite subject at school? My favorite subject at school? Rather bizarrely, my degree is in computer science and psychology. So I actually quite enjoy psychology and the study of individuals as part of that thing.

37:37So I'm going to say psychology more than computer science, because computer science was just something I could do and I didn't actually have to try very hard to do well at it so therefore I think I enjoyed the challenge of psychology that's really cool delivering technology transformation at scale is people processing technology so I can see the blend here that makes sense Matt and my last question for you what is your favorite music genre well I'm going to go with a music genre that is derided and not very popular so I quite like progressive rock so I'm a big fan of the band called Marillion that have been around for like since the early 80s and I've sent them in concert many many times really enjoy their music but to be honest I'm an equal opportunities employer when it comes to music and I'm just as happy getting to a Taylor Swift concert or listening to the latest release by Lorde.

38:37I love it. I love it. Well, Matt, it's been such a pleasure to have you on the show today. I really generally love the conversation. Thank you. It's been awesome. And I hope I've provided some nuggets for the people in the listen to this.

38:58What a fantastic discussion with Matt. Generally, that was such a treat. there's so many nuggets. So there's three that comes to mind. One, we talked a bit about the business case for AI. Two, we talked about adoption in the organization, how you support that. And then three, how do you make it a wake up call for senior leaders? So let's start with the business case. What's really interesting about what Matt described is that they've set up an AI committee to support the governance and review some of the investment cases. And when it came to the investment case, Matt looked at that from a horizontal productivity point of view, so incremental improvements on an individual level, but also on a vertical level.

39:39So the sort of really impactful use cases that benefit Endeavor internally and for the clients. So building that capability in-house at scale was really a vital part of the business case. The second nugget was around how do you get your wider workforce, right? Like thousands and thousands of people to adopt AI tools like chat gpt and he described the importance of setting up a champions network that's cross-functional so identifying those individuals that are already excited and are willing to be trained and support their colleagues so this way you've got like a sustainable organic adoption and then he described this idea of some of them are going to become ai heroes and celebrate it because they are the influence in the organization so that was really really cool and finally the wake-up call for executives.

40:28Really, it's about educating them through use cases that are proven where the ROI is established. And even more so when it comes from your peers in similar industries. You know, if a CEO knows that the CEO of a competitor, they got this massive ROI out of AI, obviously you want to get on it. So that's really powerful to have that wake up call. So thank you everybody. It's been such a treat and see you on the next episode of the Data and Mastery show and do not forget to follow and see you later.

41:03Thank you for tuning into this episode of Data and AI Mastery. If you found value in today's discussion, make sure to subscribe so you never miss an insight from the leaders driving the future of data and AI. And if you're a data and AI leader looking to upskill your workforce with the fundamental data and AI skills to transform your business, Cambridge Spark is here to guide you every step of the way. Be sure to reach out to us on LinkedIn or on our website, cambridgespark.com. Until then, be sure to keep pushing the boundaries of what's possible with data. And remember, mastery comes with continued learning and action.

41:40Until next time, stay ahead, stay inspired and stay masterful.

41:48Thank you.