Latest / Fractional Frequency / AI Should Make People Better, Not Replace Them | Ronan Wall

Transcript

AI Should Make People Better, Not Replace Them | Ronan Wall · Fractional Frequency

• Billie Tienda: Hi everyone. This conversation explores Ronan's journey into technology and learning and how that led him to the intersection of AI and workplace development. We'll discuss why traditional corporate training often fails to translate into real world behavior, how AI can help people access knowledge when and where they actually need it, and what the future of learning looks like as organizations become increasingly AI driven. Welcome, Ronan. I'm so excited to have you on the podcast. I know that you have some great stories to share. So I can't wait for the rest of the world and all of our listeners to hear it. So thank you for being on. Thank you. Thank you guys. Very nice to be here. So with that, why don't we start with the origin story of Dosin? Yeah, of course. And To get there, I've got to go back a little further if that's okay. To ⁓ to when I was fourteen ⁓ and I ⁓ started gambling in Ireland, which is a perfectly acceptable thing to do. I I discovered probability, I loved sports, I got into gambling at a young age and kind of just ⁓ make some pocket money and it kind of followed me through the rest of my life. My first job is actually in a founding team of an online gambling company in Ireland. And we went from like pre product to just north and one point two billion a year and turnover in four years before we were acquired. And then I joined another small online gambling company as like online COO. ⁓ and when I joined, we had about a hundred people. When I left seven years later, we had five thousand and we were the biggest online gambling company in the world. So I think in those twelve years between the two companies, I kind of saw every stage of growth of an organization. I became very passionate about organizational development, people development. Ultimately that's kind of what led me into the space that I'm in ⁓ right now. And this brings me to answer your question, Jenna, which is how Dozen came about. It was actually in that second company. It's a company called Patty Power, who everyone in in the UK and Ireland knows very well for US audiences since I left acquired fan jewel poker stars, both of which are enormous organizations. That should give you just some sense of ⁓ of the size of the company that we built. But when I when I joined Patty Power, 30 of those hundred people reported to me and we were amazing, our team. We were award winning, numbers were green, all the graphs went up and ⁓ to the right. You were amazing. Specifically. Well, what we did it was an operations team, and so we were probably mostly gamblers who kind of got into the industry. It's just how it is in Ireland. But we were so good at our our actual work. So it was customer support and payments and frauds and product and ⁓ lots of different kind of groups within the team. And we were so good. And then within probably about two and a half years, my team had grown to two hundred and fifty people and we were not so good. We were not winning any awards. Our numbers are now red. The graph did not go up and to the right. And it was really ⁓ it was a real challenge genuine thought I was gonna get fired. I probably should have been fired, but I think I had enough credit in the bank that I was allowed to dig my way out of the hole that I I'd created. And the reason for it, you know, a few years later, with hindsight, ⁓ was because I was trying to manage a team of two hundred and fifty people with the same structures that I'd had in place when I had a team of thirty and with a little bit of reading and things like tribal leadership and ⁓ and other books, I realized that that's just not sustainable. And so the group were misaligned, their standards were all over the place. I was churning through really good people, trying to f solve problems by hiring more. Like it's just a mess. And that was really the genesis for dosing, because I never wanted to be in that position again. It felt very avoidable. You know, I'd clearly not ⁓ kind of been experienced enough to avoid it on the way in, but I don't want anyone to have that experience that I had back then. So fast forward a few years I moved to LA ⁓ and set up Dozen with ⁓ two other founders. And the premise of it when we started was really just like let's not let that happen to anybody. There's something here. And as we kind of got into the learning world and knowledge sharing and alignment and all those things within companies, we realized just how I guess draconian the whole thing was and still is to a certain extent. ⁓ the goal when it comes to if you've ever done corporate training, like your goal is almost to learn as little as you can, get through it as quickly as you possibly can, and not get caught. Like that's isn't it? You know, it's like, Dear Ronan, this is your HR team, please complete your training. And that's that's what you try and do when you try and gain the system and you play it and you leave the room and you guess the multiple choice and that just seems like a absolute waste of time for an industry that's a multi billion dollar industry. So and we just thought there's a better way of doing this. ⁓ we were all kind of lifelong learners that founded the the platform so we wanted more personally from from learning. So that's really where it began and we played around with a few models for for a couple of years before we really landed on the one that we felt made most sense. ⁓ and that's kind of what brought Dolson to life and and just to kind of finish off the full circle story. We think of ourselves as a learning platform, but very different to any other LMS in in market. And probably the best way to represent that is we think Claude is our biggest competitor, and I'm sure we'll we'll talk a bit more about that as we go. But that was the genesis for for Dosin and and what brought me your Well, what does DOSIN mean? Good question. I guess that probably once every six months or so. ⁓ DOSENT is actually someone who works in a library or museum. It's a German word. And their job is to connect you to the knowledge or the display or whatever within the building that they're in. And so their job is really to connect you with the understanding and the knowledge and the learning you're looking for. And we always initially thought of ourselves as like a digital docent, like how can we perform that role and get you access to the information and the learning that you need. Yeah, that's what a docent is. And docent dot com, the URL was forty thousand dollars ⁓ and docent.io was four dollars. So we made the lean start up choice of going with Dosand.io, yeah. Well. That's a really good story. I mean, you definitely highlighted why you wanted to move into the learning and development space, just given your personal story with it. And I think a lot of people could resonate with that as well. You kind of touched on it, right? But what what problems like did you see or do you see still to this day, like in the traditional corporate learning that makes you believe that they're was an opportunity, there is an opportunity, right, to continue to build on this model. Yeah, there's a ⁓ I think there's a few things. I mean, ⁓ so I play a lot of sports, so I've been in lots of teams over the years and between work and school and sport and everything else. And like the most fun I've ever had has coincided with being the most successful that I've ever been in whatever capacity that is. Just a highly functioning team because we're all on the same page. You know, communication is easy. There's no disagreement about what our goal is, you know, we all play our own part. Like just those components that that all of us are kind of aware of. And then the worst experiences I've had are when it's been the opposite of that and there's been a lack of clarity and different visions from everybody and no one's on the same page and so on. And so for me, I I think I think I think sometimes learning in education is a little bit of an afterthought. It's kind of like a Q five priority. Well once we get more customers and hire more people and fix the product and then we'll focus on learning. Have money, money can go there once everything else is in place. Yeah. Totally. Yeah. Yeah. As if it's a thing that happens afterwards. But it's just not. Like you can't scale your organization without scaling your people. It just doesn't work that your people are the the resources that scale your company. So you've got to start there. It just fundamentally has to be. And so I would say that's that's not necessarily the way of thinking within all organizations right now. It's almost taught as an afterthought or something to kinda keep people happy with. And the data is undeniable. I won't go through all the stats, but like it's insane. Organizations that do this really well have thirty four percent lower churn rates. Their productivity is way higher. Like all of this data just ⁓ you know, layered on top and it's it's an absolute no brainer. The challenge, and again I'm sure we'll get to this in the conversation, it's incredibly hard to prove that or to measure that at a micro level within a company or at an individual level. So when someone leaves, was it specifically because we didn't have a program for learning? So that that's kind of hard to know. But I would say just at that level, I fundamentally believe that people are just thinking ⁓ kind of in the wrong order. ⁓ and I think if you just build a team and where you invest in the people, you avoid carrying what we call people debt. carry tech debt if you just keep working on your platform without ever kind of really aligning your database and everything else. I think of that as people debt in an organization. So if you're not doing a great job, each of us is carrying something. We didn't get great onboarding, so we missed a trick and we don't share the purpose of the company or we were we weren't in the day that you did something, you know, so we're kinda lacking it. You just you're gonna you're gonna you're tying your shoelaces together and at some point your organization's just gonna start to creak and at least all different symptoms like I think quiet quitting was a was one of our enemies early on when we were just like it's terrifying. It used to be people left their job and that was your signal. You know, you finally got a nudge to go or not do something right. Nowadays, since COVID, people are like, Well, I'm just gonna stay and I'm working remote. It's very hard to tell if I'm really putting it in and they work at about sixty percent of their capacity. ⁓ about two thirds of the whole world are like three to six months away from leaving their job or just like thinking about it. So it's like it's terrifying. the the volume. So anyway, symptoms are very real, but they're all very silent. And you really don't know that you're in this challenge until something else creaks and your every new growth starts to slow. But at that stage, you have a hundred people that are kind of disengaged within your company. So anyway, just fundamentally I feel like that's all very, very fixable or or you know, I've seen it work. So I know that that's possible to do with with not a great level of of effort kind of inserted into it. Yeah. And I guess that makes sense of, you know, let's say like generic ⁓ learning content, right? Like it has a difficult time getting adopted just given, I mean, the actual content that it is providing, right? Like the LinkedIn learning, right? Like it's something that I feel like we're the marketing is really, really good for it, but the actual content itself isn't sticking. Do you agree? Yeah, totally. Yeah. I think there's a lot of learning out there. It kind of ⁓ it edges into entertainment sometimes. So like master class you could call learning, but really it's like you're gonna turn it on instead of turning on Netflix of an evening. You know, you're not gonna sit and watch master class for an hour before you start the project to figure out how to get the project done. So and LinkedIn learns a little bit similar and then there's a lot of generic content out there ⁓ and it's much easier if you're in a HR team or an L and D role and you've just got to rock up a program. Like it takes on average two hundred hours to build a custom program. ⁓ and if you're doing your day job as well, that's three or four months of like working extra hard to create something that's probably slightly out of date by the time you finish it and you're probably not gonna be hugely happy with it. ⁓ and so it's much easier to just buy a license to LinkedIn Learning because it looks amazing. But in reality, you'll probably teach people certain things, like the certain things you can absolutely learn, like generally speaking, you can understand more about an organization. You won't know much more about your own organization. after after that. So ⁓ like ultimately you're you're potentially training people for their next role or training them generally on themselves. But what you really want them to be able to do is do their best work in the organization that they're in. So we're firm believers in like take your own data, your own people, you probably have everything you need already to create really effective learning within your company. We've just never had the tool that allows you to turn all of your existing institutional knowledge into your training. So because of that gap, you haven't really had a choice. You can invest a lot of time to build custom programs that are going to be out of date, or you just find something off the shelf. Or you invest in DOCEN and you can create personalized learning, right? Absolutely. And I think ⁓ truthfully, two things have kind of coalesced for us as well. One, we focused on this ⁓ a lot. So we we kind of understand the mechanics of learning ⁓ incredibly well. We think we've built a really nice interface, we've made it easy for people to build learning. We've done lots of really good work. And our engagement rates play that out, our mentorship engagement rates are like ten times industry average. But also something called AI appeared on our radars a few years ago. And that was really the catalyst then to allow you to take 200 hours to build the program and turn it into 20 minutes. And that's kind of where we've gotten to. So we're now massively built around the capabilities of AI. And this is why I mentioned Claude earlier. We kind of think of ourselves as Claude for training because it used to take three or four months to build a website. ⁓ you find someone who could build it, you set your requirements, give it to them, meet with them, do a thing and a thing and first version, second version. And now you go out to Claude and just say, Build me a website and fifteen minutes later you've got one. It should be the same with learning and training. It should be like I don't have to spend three or four months building a product training program for new features. I should be able to just go into Docin and say, Here's our new features. Can you turn this into a program please? And it gives you back a program fifteen minutes later. So yeah, so I think those two things just kind of Combine in a really interesting time for us. And ⁓ yeah, invest in Dozen. That's the that's the third option. That's the right answer. I think the part that's really striking for me is that a tool like Docin allows really a subject matter expert or a leader to become a builder of a training program. I think about you know LD teams that I've interacted with in the past as sort of gatekeepers of such programs. And a lot of times it's you know, a person who's never led building a leadership training or a person who has no technical capability building a technical training. And I think there's a disconnect there that that shines through in the usability and the relevancy of the platform or the interaction for the end user. So that I think to me is sort of a light bulb moment that something like this allows the right people to be ⁓ behind the the program. design and and delivery. And when we talked earlier in the discovery call, Ronan, you mentioned that you feel like AI should make people better at learning. So not just make it easier and faster to build the program, but make people better at it as the end consumer. Can you talk a little bit more about that topic? Yeah, of course. And that's a really interesting point. I hadn't actually thought about that. We just launched some programs this week. And just to kind of reinforce what you're saying, we we allow you to assign programs to different people in the organization and there's actually five leaders across sales and marketing and product who own their program and their job is to sign off on the content and the quality of the content, not the H your partner's job to go through it all and get it. So it's a really, really good point. So I think to your to your second question, ⁓ and I've totally forgotten what it was. We definitely got that making Making end users better at learning. I mean, as as an end user myself, I'm I'm thinking about like sitting in front of the compliance training and I am so guilty of the like, you know, hit play, leave the room, pull up another browser, whatever I have to do to tune out and then just pass the quiz to the end. Totally, totally. But I think there's probably two things. I think one is the learning itself. And I think the w the way that we think about it is you wanna build it in a form that's quite compelling and digestible and interesting. And so there's a variety of different ways that you could do that on doses. So we've a series of features of of ways that you can present the content. A kind of written, visual, audio, you can upload content, like whatever you want to do to essentially make the content ⁓ digestible. We try and cap it at a few minutes long. So like every lesson should be no more than a few minutes. But then also the learning features that you can assign to it. So you can apply quizzes, you can apply live scenario tests. You can actually engage with Elliot, who's your AI assistant on DOS and actually have a conversation with them and really just like stress test what you do in a particular scenario on the back of the learning and a few other tools as well. So that's all very easily digestible. It's a good way of assessing an individual's kind of understanding and comprehension. And it's not ⁓ you can't fake it, you can't fake the life scenario. ⁓ you know, there's just certain things you can't get around. So I think that's ⁓ a very effective way of doing it. And it's all built within a solution. And this is where my old gambling kind of hat comes back on, which is all kind of fun. Like it's designed around ⁓ kind of a gamified approach. And so you can see what your own development tier is, you know, everything from rusty all the way through to bronze, silver, gold, platinum, and you can see that where it is that you're kind of, you know, like my woot band you know, tells me my whoop age and I'm ⁓ gotta get my VO two max up because that's costing me some points. Same with learning, you know, and that's that's where we really think we should get to. And again, I'll we'll speak a little about the future of this in in a little while. But essentially that's kind of the engagement component. But then the other piece to remember and not to be offended by is that people forget ninety percent of what they learned in the first thirty days. So you do a beautiful job on the learning. They're not going remember it all. But what we do remember as humans now is where we got the learning from. And so we'll remember that we got it from DOS and we'll go back and revisit when we need an answer to a question. And so what we actually have Elliot doing as well is building the program and serving the content is being available in a generic AI fashion to answer any questions on anything in any of the programs. And so you just might need an answer to a question from something that you learned three months earlier. You just ask Elliot and Elliot gives you the answer back. And so that component is like the reinforcement of the learning. But in a very non-judgmental way and something that doesn't burden people to find answers. So that's kind of a little bit about how we think about it. One crazy stat that we learned recently is that people are spending about an hour and a half to two hours a day currently ⁓ trying to find information or reproduce information that they already learned. Like that's 25% of their day. Because if you can just have that in the form of Elliot and get an answer in a minute them so much time and also saving them from bothering other people to try and get the answers. ⁓ for us it's that, it's non-judgmental, it's kind of make it fun as possible, but don't judge them if it does doesn't land. Just let them come back and get the answer when they want. And also we just released Elliot's now available in Teams. So you can now do the same thing from your teams. ⁓ from teams as well, you can just chat with Elliot and get answers to questions as well. So again, trying to meet people where they are and and not again judging them for not remembering something that they were trained on six months ago. Yeah. And not having to go back into some hour long recorded training and like slide through it and try to find the moment where they took away a piece of information. Another thing that's maybe kind of a zoom out, Ronan, and I love your perspective on there's a ton of anxiety across the workforce now about AI. People are just afraid of it. They're afraid that their skill set will become obsolete. They're afraid that they won't be able to keep up in building their skill set. How would you advise organizations as they consider how to deploy AI tooling into their organization in a way that empowers their workforce rather than scares them, for lack of a better word? Yeah, really good question. And I think one thing before I go into kind of the organizations approach, I would I would just advise people who are nervous about this to kind of look back in history because we've had multiple moments where And new technology has been introduced and it looks like it's going to be the end of a large number of roles. So ⁓ the calculator. A calculator was a job. That's someone who calculated all day long until they brought out the calculator. But all of the people who calculated for a living just figured out a new role and they figured out how to be a new user of the calculator and they c they calculate more numbers. Computer, a computer was a role. Someone who computed all day long and then the computer So, like we've had lots of examples in history of this, and Excel arrived and all these things. So I think I think with AI, we've actually seen an increase in the employment rate for recent grads in the last two years when everyone assumed no one would need grads anymore because AI is the equivalent of the smartest grad you're ever gonna have in your life. And God, it's a burden to train a grad up, right? That's gonna take a couple of years. We don't get a lot of value from them. Why would I do that? Let's just use AI. So it's not quite playing out as people thought. So I would say to that extent, I think for individuals, it's just about accept that this is the new norm in the same way that the internet became the new norm. Get on board with it and learn it and just make it part of your day job. And like when I'm hiring now, we the thing I love talking to people about is how they do their job with AI. So I still have to have them do their job. Like they've got a thing to do and a problem to solve. But I what I'm asking them is like, how do you solve that problem with AI? How does it help you solve with higher quality or more quickly or like and that's fascinating. So I I think it's it's it kind of upon all of us to do that. And then from an organizational perspective, I would say ⁓ I think you've got to make it clear that whether this is true or not, I think you should certainly start at this place, which is you want to empower your people with AI because if you can two X, three X, four X your productivity, your revenue line with your existing Team, you're gonna take that opportunity. Your first choice isn't gonna be to remove people to save cost if you can increase revenue. That's always preferable for companies. So that should always be the starting point. And then empower your people, like give them some of the tools, pay for their licenses, and like give them direction, give them training on AI, like help them to understand it and consume it. And then when you do that, you just watch it explode. Like we introduced it. kind of almost in a mandatory sense last year for our engineering team. I just said, look, you just have to figure it out to two, three, five X your workload. And I d that could be quality or volume or pace or whatever. You know, and and they did once they got into it and we invested in them. Like they literally didn't three weeks. You know, so once you're removed, they're like, I don't want it, I'm not moving you, I'm empowering you. You tell me what you need and I'll I'll pay for it. So I think that's a good starting point for organizations. And then when you've done that you'll now start to see where this is settling out. Like what are the tools that people are using? How are they utilizing them? What's the impact and the value of that and who's rejecting it? And if you're hard rejecting AI, it's like hard rejecting the internet twenty years ago. It's just like you're just not gonna be around in that role and it's entirely back on you, I think. But for the organization I would say invest in the tools and the training to try and upskill your people and you should start to see lots of value from that. So ⁓ Yeah, I think as a starting point, that that's probably the best approach. Yeah. I I like the message that sends too. When people are invested in, they can feel that. And it's a much stronger message than just ⁓ you know, go figure it out on your own. If you give them some some guidance and suggestions and tools to go check out, it's it's a better place to start. So back to the L and D space. Yeah, one thing I've never seen go well, honestly, is measurement of that function. It is always boiled down to and hopefully I don't have past peers who are rolling their eyes at me, but it's always boiled down to like how many people completed the course and like checked the box or you know, how many people like turned up to the webinar ⁓ and and that sort of thing and taking attendance, but really nothing that that you could translate to real business impact. And so I would love your thoughts on that. As companies or leaders are evaluating their learning and development strategy. What metrics should they be paying attention to? I love this question because this goes back into data. ⁓ so there's something called the Kirpatic model that I'm just I'm an I'm a nerd on. I really, really love it. I think it's an excellent way of measuring the impact and the value of learning. And it's what we've based everything we've developed in terms of insights and value around. And you're you're right, Aaron, in that it's been incredibly difficult to do this for a long time. And so people have just been left with no alternative other than to say we trained twenty people, twenty is more than eighteen, and we did eighteen last year, so we're doing well, right? ⁓ and so the carpatrick model is is kind of like four stages within it. I won't go through all the detail, but it it's kind of goes from kind of qualitative to very quantitative. It goes from I really liked that training all the way down to here the dollar amount that we increased our revenue by on the basis of the learning that was applied. Like so top to bottom. And so we're there's kind of four levels. We're around about level three right now. We can figure out how to get you to there, which is like the impact on your work of the learning and the training that you received. But we're not there yet either. I haven't seen anyone get there. I don't think and I haven't seen anyone be beyond us in terms of their ability to measure ⁓ the impact or the value of learning. So we're on a path where we know exactly what it is we need. But the only caveat I would say is organizations are not good at measuring impact generally and that's the thing that we need. Like with someone who's in a sales role, that metric is much easier to measure. Like how many sales did they make? What was the value of the sales that they made? How many customers did they retain? There's some hard data. And then when you apply learning and training, did those numbers improve or disimprove? You can do cohort, did they get to those numbers quicker or in the same amount of time? So you have methods of getting there because that data is is gathered. How do you measure s a product person? You know, like did they do a better or worse job? Was quality better or worse? Like it's very difficult to really measure that 'cause there's so many inputs and so as you apply training, it's hard like there are no metrics for success in terms of that. So so I'd say there's also challenges in terms of just generally understanding the measures of success of specific rules, and I would say the majority of rules, ⁓ which then makes it virtually impossible to apply training and see the training improve or disimprove that data because it just doesn't exist yet. So what we're trying to do with organizations is integrate into their existing systems where they gather some of this data and where we want to get to is the ability to say, you apply this training and improve these numbers or disimprove them by X. tweet the training, apply again, and improve by why, and you can now start to get some really kind of material or why. So that that's kind of where we're getting to the current status for us. There are some things, the good news there are some things that you can measure today to be able to feel comfortable and confident. And the good other good news from it is that they're still delivering enough measurable value that it warrants putting the effort in. ⁓ and the biggest one for us right now is level of effort to produce training. by individuals. It used to take 200 hours. You can do it in two hours on DOS and now a fully fledged program. You've saved yourself a bunch of time. There's an actual cost to that that you can actually measure and bank, which is very easy to do. And also for people that are reinforcing your learning, if they spend an hour and a half to two hours currently assessing thirty things a day, there's a ⁓ time associated with each search. If you can get them the answer in thirty seconds or a minute through Elliot, we now understand that that saved them amount of time. So you can get efficiency gain today. That's just access. That's the building of learning and the access to information. And so that's what we represent as Oral White today on the basis of kind of efficiency gains. And then there's just science that says organizations that do this well have lower turn rates, higher productivity rates. And again, we display those without saying these are your numbers. And again, the closer we can get to that data, the Quicker we'll be able to say for your organization to specifically improve these numbers by by X. Yeah. So there's a way we're doing it today. That's probably only about 10 or 20% of the total value being added. ⁓ but as I said, there's enough value there that it warrants being displayed for organizations. Yeah. Onwards towards Carpatrick level four. That's the that's the dream. Yes. I think that's that's hugely beneficial to at least be able to talk about. the behavior of the users that are interacting with the platform and the time savings that's material and and more impactful certainly than attended scorecards and and that sort of thing. And I imagine maybe the future horizon sort of hinges on businesses being able to even articulate their own kind of core business and performance metrics and and you know, to say that that's figured out ⁓ in most cases is maybe a stretch ⁓ today. So totally Excited to follow along with ⁓ your progress as you track towards the fourth level. Yes. Yes. The other thing, this was sort of an aha moment as as I was listening to you and and Jenna chat chat a little bit earlier, was that you know, L and D, I think, in some contexts, in some companies, has been seen as more of like an employee benefit than a critical business accelerator. And I think what's clear in the conversation today is that in in order to be successful as the pace of business continues to change and move faster and faster, you have to see it as the latter of the two. And so sort of curious, Ronan, for your perspective on how companies can connect learning outcomes and learning quality to real business outcomes, and maybe specifically outside of the research that you've already mentioned. Have you seen a really high performing company do learning really well? And and if so, can you comment a little bit on the relationship anecdotally there? Yeah, I'd love that you asked that question. And again, I'll start kind of in future stage and I'll come back to where we are and then I'll I'll ⁓ give you an example of a company that's doing that well. I fundamentally believe that learning should be a marker that is used when key decisions are made about individuals. So if you are about to promote someone on a someone Determine their annual salary, if you unfortunately have to get rid of people who you're gonna get rid of. They're key management decisions. And when you're making those decisions, you have a series of inputs. And one of the key ones is how good are they at their job. So I'm making this up. You're one of the companies who actually has goals, and I can see that Ronan only completed three out of his five goals, but Jenna completed four out of her five goals. So Jenna's better, right? Better employee. There's a number of other inputs as well, but they're all highly qualitative. And unfortunately, you might have very little time that you've engaged with Ronan, you've spent more time with Jenna. So some this is a little hard to kind of utilize. There's no input for learning, curiosity, self-development. And at the end of the day, they're the people that you're going to want to put into more senior positions and keep within your organization. ⁓ because you have to have people that are comfortable with change and have the ability to advance and be curious and learn and adva and ⁓ and work with change. I would love for us to be able to provide that marker. Within DOS and today, we can tell if someone is rusty on silver, gold, platinum in terms of their learning journey, how many programs they've completed, mentoring other employees, you know, what were their scores within these programs. And so even though I got three out of five, let's say I'm mentoring two other employees. I've completed four programs. I'm platinum in my ⁓ my tier. And Jenna, who had four out of five on her goals, actually hasn't done her programs, isn't mentoring anyone, she's in the bronze side of the tier. That's a huge input, right? That should potentially change the decision that a manager would make as to who to promote or what to do in a particular situation. I think that's crucial and that's where we need to be getting to to give us a much better steer on individuals, but also really importantly, make learning matter because if I knew that that was going to be an input when it it was decided at the end of the year who got promoted, you bet your ass I'm going to get into my learning and I'm going to want more of it. And I'm going to start pushing my manager because there's not enough of it for me to maintain my platinum level. So now the recognition of it ultimately will drive a kind of accountability with individuals. So I think that's where we want to get to, where we are today is we now have tiers in place. whereby we you can see the level that an individual is at, you could see their progress, their scoring, you know, all of that is now available for you as a manager. We're piping it back into HRIS systems and through our integrations, which means that it's now going to be available for managers where they already are as well. ⁓ and then I think in terms of who does does this well, we just launch a bunch of programs, a kind of a whole new cohort with one client of ours. ⁓ and they've done a brilliant job ⁓ of really embracing all of this. They've leveraged their own people. The CEO is like interviewed and is the video in a bunch of different lessons. And you know, so you're actually hearing from Wendy and now you're being asked about what Wendy's talking about, the context for the company, ⁓ and all of that. And so they're now leveraging that, they're using the tiering system to essentially determine success for individuals within the within the organization. So that's about as good as we're seeing right now, but ultimately that's where we want to get to is that's a core input because I think chances are I'm gonna help someone bridge the technical gap that they have quicker than I'm going to encourage someone from being disengaged with their learning to being a highly engaged learner. I just I don't know how to do that, but I for sure can teach someone to be slightly better and deliver more of their goals. So yeah, that that's a long answer to a short question Erin, but ⁓ yeah, that's kind of how we think about it and how it looks today. What's sort of interesting is that what stands out in that particular customer is just the executive involvement. And it seems like so simple, but I've never really seen an executive team or even a leadership team lean into learning like that. It's always like the off the shelf paid actor experience, which as a consumer, it's like, I don't care. And and to your point, I'm not rewarded for interacting with this. And so it's I'm I'm disengaged immediately. And I think it changes the ⁓ the perception from users and consumers of the learning when they see their superiors and and company leaders interacting and and leading the way. So huge. Okay. I'm gonna shift gears on us. We've got two quick hit ⁓ kind of advice type questions to wrap us up, but I have just sort of gotten lost in this conversation. I think this is such an interesting topic and industry. So I'm I'm so pumped to share this episode. But before we go, Ronan, if a company wants to start using AI to improve learning today, where should they begin? With Dozen, obviously. I would say not to not overthink this too much. You know, I think like the problems are still the same when it comes to learning and training and development. You know, they haven't changed a whole pile. People still want to be developed. They're always going to be eager for more learning. You've got to train people up in compliance and standards, or you put your company at risk. You have to on board, you have to help people. Like it's the same stuff. AI is really there to just aid you to do that. More efficiently, more effectively, with higher quality, with greater scale. So I would say just look for tools that are best leveraging AI and can deliver back the solution to the problem in the most effective way, or certainly look like they're at the cutting edge, because what's gonna happen is in six months' time, you're gonna look at the way we used AI today and laugh at how basic it was the way that we used it. So if you find a partner Who's really at the edge of this and will continue to develop as AI develops, I think you're in a really good place. And that's where I would certainly invest my time. And I say invest because you should like put the time in now. And I don't need a lot of time here. I'm talking about a few weeks where you really just think about what's the portfolio of learning that you'd want to have in your company. Like set the framework for yourself and then start to chip away at it. And it's it's not difficult to do. We'd even help you do that. It's the first thing we do about companies. based on their stage and industry and all that. So like set yourself a little bit of a map and then work against it as quickly as you want to. so that's what I would recommend. ⁓ and then there's always and and this is where I mentioned earlier how we kind of think of Claude as almost our competitor right now. Question we're getting right now is like can I not just build this myself? Because you can kind of build anything on Claude and the answer is you probably can. But I don't know if you want to go and build an enterprise level SaaS platform. ⁓ you know It's just a lot of effort and you gotta maintain it and you're probably gonna miss something. And some of the secret sauce is not obvious. You know, like it's taken us a few years to hit ten times engagement rates versus industry average. Like that didn't happen overnight. It certainly wasn't in our first programme. So it's it's some of the more subtle things that that can help you get there. So but I would say that's the other thing is like choose between builder buy because you can get a DOSIM license for less than you can get a cloud license. And if you're gonna start buying cloud licenses And expect people to just be able to go and build this. You're just gonna be sorely disappointed. They can do very cool things with Claude, which probably still have a Claude license, but if you just expect them to build versus buy, I think you're you're in for kind of a tough learning curve. So so I would just say like pick whatever partner looks like it's gonna be the one for you for the next little while and kind of staying abreast of everything that's AI ⁓ related, and then choose between builder buy. ⁓ and I don't really don't mind which one it is, but like pick a lane and go for it. Don't leave people in limbo and certainly don't buy them clawed. And then tell them that's all that they're getting. Go fix learning because you're kind of giving people a rake and ask them to dig a hole. Like it's just it's kinda get there a bit, but it just won't be as effective. So ⁓ that would be my my recommendation. And I've yet to see a negative ROI on an organization investing in in learning and training. Just have it. So ⁓ it's not a difficult decision to make, I don't think. I think the great news is For mid market or enterprise companies, there's probably a lot of dollars ⁓ sitting in this budget for these off-the-shelf kind of content library type tools that exist today. So you'd probably find to your point that there's substantial cost savings in rethinking that model. ⁓ yeah. We see like six to eight programs and platforms and tools that organizations use. Like they'll get a Google Sheets and then a little bit of an LMS and then some cloud and a thing and a thing and a thing. And we did one comparison recently where they were like, we're choosing between you and these other two. And I was like, okay, cool. But you can't have that one on its own. That doesn't work. You gotta have a second tool to plug into that one. ⁓ okay, well, what about that one? Well that one only gets you this bit. You've actually got to have this one as well. And really quickly they were like, ⁓ God, you know, like it's just a lot and it's multiple tools and it's so don't through it. Just just use to us. Yeah, there's a lot of d a lot of zeros on those contracts. Yes. And so totally appreciate that. Last question for you, Ronan, kind of zooming out. What are you most excited about with respect to the future of learning and human potential? So where where we're going because we're excited about it is everything I've described today is very much about the idea, mostly about the idea of the organization being able to build quicker, higher quality, custom training models for people to engage with, all of that. So it's still being led by the organization mostly. I really want to get to a world where the organization merely trains the model and sets the kind of boundaries, if that makes sense. Like this is our culture, this is our strategy, this is our no and lets the individual be accountable within that and starts to move the responsibility for learning back to the individual. And so instead of saying we've got five programs leadership, management, communication, the individual should be able to manage their own career and their own development and own it. And then from there, it's up to them to say, hey, this is where I'm at today. This is where I want to get to. This is where I'm stuck. This is what I've been asked to do. And educate, self-educate. And then the organization knows that they're going to be educated within the boundaries of what they've shared. So they're not going to be making things up or getting inaccurate information. And when you get to that point, now there's even less responsibility on the organization and really you start to see who in your organization are the hungry curious learners who can manage change effectively. It's just it's an absolute no-brainer like who's active and who's not, ⁓ is really just the metric. But also you're controlling your own career. And at the end of the day, again, you see people that are kind of ⁓ motivated enough to be able to to do that. And then building, just like you can with Claude today, we're all getting very creative with how we use it in our work. Let's create get creative with how we learn as well. So that's the bit that really gets me excited and as much as we can, we're really starting to lean into that at the tail end of this year, ⁓ to start to kind of pull that accountability back towards the learner and out of the hands of the organization. Yeah. It's a very cool North Star. I think that people would be really empowered to ⁓ roll into development conversations and that sort of thing, having architected themselves rather than than otherwise. So thank you so much for letting us pick your brain. This was so much fun. really enjoyed the conversation. I know others will as well. Ronan, if anybody would like to find you or learn more about Docin, where can they find you? So dosan.io is our website. You can reach me at Ronan or O N A N at dosan.io and ⁓ I'm all over LinkedIn. The first time, Ronan, that a guest has shared their personal email address. So bravo. Totally for