Greg Arthur (02:08.0)
Welcome to the Product Design for Learning podcast. This week we have, something I haven't spoken to you for a little while, but is still a lovely, lovely person. It's Carlo Jose, who is now at GSK. And I'm going to shut up and let Carlo introduce himself.
Carlo José (02:35.951)
That's a lot of questions. Thank you, Greg. Real pleasure to be here and really great to stay connected with you over the years. In terms of where I am now, as you've said, I'm Global Head of Learning and Development at GSK. I joined GSK about a year and a couple of months ago and I'm currently on a mission to
Greg Arthur (02:40.782)
Tell us who you are, why you're here, what you do, where you've been, where you're going. Hit us.
Carlo José (03:06.03)
help the organization modernize and deliver personalized learning to our people. And that's all in line with our ambition to bring together science, technology and talent. That's kind of our mantra at GSK. But before that, and I think this is where our first kind of contact and engagement would have occurred. I spent closer to two decades at HSBC and half of that career was in the HR learning talent space.
Greg Arthur (03:25.804)
Hmm.
Carlo José (03:34.35)
The other half of that was in service delivery and operations. I'm a bit of a, I say still new to the HR practice and the L &D practice? Maybe not quite, because it's been more than a decade, but I'm still straddling in between not quite an HR person and not quite an operations person anymore.
Greg Arthur (03:47.18)
Yeah.
Greg Arthur (03:54.333)
I think that's a nice mix to have, especially with our, this is a killer segue by the way for anyone listening, that is a nice mix of career to have for our topic around data. As in previously, as in like if we take like 10 years ago, data was kind of maybe something that people maybe touched on and it was maybe just something like, give me one reason to do this and the reason I don't even have to be that good, but just give me something to go on.
Carlo José (04:19.888)
you
Greg Arthur (04:22.796)
like here's the budget or yes, you can go ahead of it or whatever it's going to be. Whereas now, I guess in service delivery, product design, operations, HR, learning, all these kinds of things, data is almost killer whether people go, yes, you can go ahead with this or yes, that's a good idea or whether they go, this was a horrible failure. We should do something about this because of the data we've got. So first question, usually ask everyone the same question at the start.
How would you summarize this phase? So when we talk about our product design phase and we talk about data and all these kinds of things, how do you feel about it? How do you feel about its importance? What does it mean to you in as close to 60 seconds as you can? But we're never to a clock on or anything.
Carlo José (04:54.997)
Yeah.
Carlo José (05:01.329)
Yeah, I'll try to summarize what I think about it. So you're quite right. think back in the days, our kind of appreciation and usage of data in any practice, but particularly in the practice of learning, development and design was, I think, very rudimentary. And I think it was kind of very limited. We have a kind of old school practice that
kind of help you to somewhat get approval for projects to get done, but then have some rudimentary measurement of it. I think in the design process, it deserves to have a phase so that you can have a kind of a proper conversation about what data is involved in scope available, missing, what would you use to measure certain parts of your design process or your solution. But I also think that apart from being just a phase with a start and a finish, it kind of is something that threads through across everything because
Greg Arthur (05:32.939)
Mm.
Carlo José (05:59.274)
leading up into identifying a problem statement or a challenge or an opportunity, there has to be a form of it that isn't just feeling that is data driven, but then all the way to the end of it to know whether you've achieved success or you've accomplished anything, you also have to have a little bit of a data, a quantitative, along with a qualitative kind of aspect to kind of consider, right? So I think standalone phase well deserved because it is important. And if it wasn't there before, very much putting that at the forefront of your design process.
Greg Arthur (06:07.914)
Mm-hmm.
Carlo José (06:27.237)
but also something that threads all the way throughout.
Greg Arthur (06:30.284)
100%. And we're working on revamping our process at the moment. And pretty much as you just said, it shouldn't be something that is you only touch on it here within like a confined space. It should be something that in your scoping phase when someone says, I would like, or I have this problem. And first if someone comes and says, I would like, we'll have a different conversation about that. But if it's, have this problem.
Carlo José (06:54.65)
Yeah.
Greg Arthur (06:57.777)
or we would like to have better this, that or the other, whatever it's gonna be. The question always comes back to data really, as in, how do you know it's a problem? And if you want something to be better, faster, whatever it's gonna be, better than what? Like what's your baseline, what's your benchmark? So it's trying to make sure that people understand that these things, sometimes they kind of fall to the background a bit, as in, sure things about visual design, data might take a bit of backseat then, but it's still there. Like...
Carlo José (07:25.647)
Yeah. Yeah. Well, I think it's still quite challenging because
Greg Arthur (07:26.955)
Yeah, it's going away. And we talk about it a lot in L &D about data. think AI is maybe taking over as one of the key topics people talk about now more than anything. But I think storytelling was the one before that. But data was really hot. It was like the hot thing for a couple of years. But where do you see that sitting in the conversation now?
Carlo José (07:53.649)
I remember when I first started working in the kind L and D function as someone who didn't grow up in the HR and L and D space, I was a bit of a kind of anomaly. I came from an operations background with a clearly an energy and a passion for all things learning and development, particularly leadership development space. But I felt like a bit of a fish out of water because the conversations and the kind of questions I was asking around, how do we measure that we've actually done something here?
Greg Arthur (08:10.25)
Mm.
Carlo José (08:21.615)
Apart from the butts and seats, the compliance aspect of it, is very relatable in operations, there is a component of it that's very trackable, measurable, compliance related. There's other aspects of L &D that I thought were kind of lacking in maturity at the beginning in terms of how we manage and operate the learning function, but equally in the design process of identifying what are the opportunities, what do we need to design for, understanding who our customer or learner is.
Greg Arthur (08:31.583)
Mm.
Greg Arthur (08:48.554)
Hmm.
Carlo José (08:48.678)
it wasn't as kind of mature back then. So I think the direction of travel now is, I think that there are still some internal customers, know, stakeholders in the business that are very much, you know, treating you as an order taker. Going back to your point about if the conversation starts with, want this, we need to have a different conversation is what are we trying to do? Because I often, my favorite example is, you know, it's a bit of an off tangent and nothing wrong with masterclass, but masterclass is usually,
Greg Arthur (09:00.253)
Mm-hmm.
Carlo José (09:16.724)
really good at promoting their product on social media. So it's not uncommon that a senior exec is gonna see that on their Instagram or their Facebook feed and go, oh, I really think we should have this. And they approach learning and say, we should buy this for our employees. And I said, well, what are we trying to do? Are we trying to be, how to cook a poached egg by Gordon Ramsay? I don't think there's a problem statement that's clear here that helps you to arrive at we should buy a masterclass, right?
Greg Arthur (09:28.074)
Hmm.
Carlo José (09:44.648)
And so I think connecting that back to data, it is about that understanding of what is the opportunity, what is the challenge we're trying to overcome? What is the behavior trying to change? Is there even sentiment? you know, to your point about artificial intelligence and data and AI, we've got such a great power now to be able to deal with unstructured information that might come from surveys and things like that, that we didn't have before. And I think for kind of a long time, you survey people,
Greg Arthur (09:52.349)
Mm.
Greg Arthur (10:05.81)
Hmm.
Carlo José (10:12.115)
you kind of got some people to really do some manual work to try to do some sentiment analysis. Now it's like a click button and you can get initial sense of even the most kind of basic unstructured data can try to make sense out of, or you can easily rule out and say, there's not enough information in here for us to work with. And sometimes that part of the which is, is this telling us something that's actually useful input or should we make a decision to discard this and
Greg Arthur (10:17.767)
Mm-hmm. Yeah.
Greg Arthur (10:29.214)
Yeah, absolutely.
Carlo José (10:38.792)
you know, do our own primary research, do additional secondary research or collect data in other ways. Does that make sense?
Greg Arthur (10:45.832)
Yeah, 100%. And it reminds me of a conversation I had with a chap called Tom McDowell from the Learning Network. We did a separate episode, but he talked about data and said, he was talking about the fact that he is very aware that data analysis isn't his strong suit. Once he's got the analysis, he can run ahead and go nuts. But just that bit of going through a spreadsheet and going,
another 100 lines of people going, like this and we should do more of that. So he was like, I work with a data analyst who's really good at those kinds of things and that helps me out. But then to your point about AI and the kind of things that are available, there is an easier and quicker way, but I still wonder if, so I don't use it as much as I should maybe, but I do wonder if there is an element of you still need a human being to go, is that right?
Carlo José (11:40.316)
Yeah.
Greg Arthur (11:41.265)
Is that really what the story is telling us? Is that really what the data is saying? So where do you even begin when someone says, before you even got all of your data, where do you even begin to start with someone going, we have this problem, help me? Where's the first point that comes into your mind about using data as a tool rather than something to kind beat people up with?
Carlo José (11:55.154)
Yeah. Yeah, yeah, yeah, absolutely right. I mean, you don't start the conversation with a response of where's the data. I think there is very much in that kind of performance consulting space for HR practitioners, as well as my practitioners where there is a human element. And I think oftentimes business stakeholders are also struggling to articulate what the problem is. And I think part of our job
Greg Arthur (12:09.001)
Mm.
Carlo José (12:24.406)
is to ask the right questions to be able to draw that out. And I do think that, you know, it's easy to jump to judgment and say, these people don't know what they're talking about. They haven't thought it through. But actually a little bit of our own kind of expressing a little bit of empathy and trying to put yourself in their shoes. They're clearly being asked to do certain things or achieve certain things or have goals that they need to achieve. And I think it's that deeper understanding where you can start to draw that connection back to the data again.
Greg Arthur (12:27.666)
Mm-hmm.
Greg Arthur (12:42.728)
Mm.
Carlo José (12:52.176)
And I would also say, exactly to your point, data is not the be all end all. It's just one part of our tool set that helps us to understand problem, understand opportunity, potentially understand impact and outcomes down the road. So I think it deserves a of a place where it's a conversation topic, but it's not the beginning and the end of everything. So there's certainly in that performance consulting space, an opportunity for us as practitioners to say, look, before we get into the details of this,
let's, you tell me about what's going on with you and why this is coming up and why this is a problem. And I think usually that helps to shake out what are the different anxieties and, you know, problems that are actually underlying the ask for, please buy a masterclass from me and get a thousand licenses.
Greg Arthur (13:25.233)
Mm.
Greg Arthur (13:38.601)
I mean, I mean the X amount of thousand licenses, not just for masterclass. I've not used it, but it looks interesting, but I've not used it. But for any like LinkedIn learning masterclass, wherever the kind of tool is going to be, it does feel like the only conversation around data is how much of a discount can we get for how many X amount of thousand licenses? Never. What will it do for us if everyone starts to use these platforms?
Carlo José (13:44.789)
You
Greg Arthur (14:08.488)
what are we gonna get out apart from engagement? Again, to your point before about butts on seats, it's just the same, but it's clicks. There's no impact measurement. So you mentioned about performance consulting in what you were just saying. So for me, that's, when I try and boil it down to somebody to explain what it is, I'm saying it's about having a really good conversation. As long as you don't kind of like close the conversation off as quickly as possible, you basically wanna get the other person.
Carlo José (14:08.616)
Yeah. Yeah. Yeah.
Carlo José (14:29.703)
Yeah.
Greg Arthur (14:36.21)
to continue drip feeding you bits and pieces of information. So if they say something and it kind of feels like it's a bit open-ended or there's probably more to that story, you're go, tell me more about that. I'd like to, know, or does it mean this? Like, can you tell me a bit more? So where do you use that to start to pull out bits of data from people without them even realizing they're doing it?
Carlo José (14:59.279)
Yeah, yeah, and by the way, so first of all, I have nothing against Masterclass. I've actually used the product before. And I like to cook, and actually the Gordon Ramsay stuff was great. And there was some negotiation stuff by a CIA, know, kind of negotiator, really good. But in the corporate sense,
Greg Arthur (15:16.039)
Yeah, I think you're going to say some negotiations by Gordon Ramsey. That's not the bar that I want for master class.
Carlo José (15:18.494)
might not
Carlo José (15:22.264)
No. That aside, mean, think, you know, one of my favorite kind of probing statements, you know, there's also the five whys, right? Like ask why a couple of times to get to the kind of underneath it. But I also, one of the ways for me to stop myself from jumping quick to judgment is tell me more. That's actually one of my ways to open up a conversation. Cause I think that phrase just kind of
is embodying a bit of empathy. It is embodying, you know, I've got an interest. I want to hear what else you've got to say. I'm trying to suspend judgment. I want to help you get there. And I do think that that's certainly a good tool for practitioners so that they don't come too quickly to a conclusion that says this is not a valid learning requirement. That being said, you may still arrive at that conclusion, which is, this seems to be a bit more than just build a learning solution or buy a learning solution, which is another thing I'd probably point out that
Greg Arthur (16:16.198)
Hmm.
Carlo José (16:20.601)
classically learning functions are thought of as sometimes a silver bullet. this is a training problem. Therefore let's do some training. But I think part of that performance consulting conversation is also about understanding the ecosystem around the target audience or the employee that we're trying to help. But sometimes it isn't just about a knowledge or a skill issue. Sometimes there is kind of motivations impacted by environment. Is it the performance and reward system? Is it kind of a local culture thing?
Greg Arthur (16:26.886)
Mm.
Carlo José (16:48.894)
some of this has connections to data as well that you might be able to quantify. But I think those are the kind of conversations, the deeper conversations you need to have to get beyond the initial, I think we need a training solution to solve this problem.
Greg Arthur (17:01.668)
Yeah, absolutely. I think, I think one of the things that I kind of, I guess I struggle with it when I kind of don't, when people don't kind of get it quickly, because it feels so simple is when you're having these conversations and you kind of keep coming back to, as you said, why is it a learning problem? Because it does feel like that's the easy get out for you to go, it's probably a marketing problem. And you go, maybe, but maybe not all of it is a marketing problem.
Carlo José (17:28.824)
Yeah.
Greg Arthur (17:31.267)
Some of it might be a problem that you haven't even thought about. It could be to do with your historical leadership or the fact that your company doesn't allow a certain technology in or outside factors. But where are you getting that from rather than the kind of the get out of, this doesn't, this isn't my problem. But then, so we always try to back back, pardon me, try and bring it back to a couple of things around your as is in your 2B state. Where are you now? Where do you want to be? And then we,
Carlo José (17:46.006)
.
Greg Arthur (18:01.049)
we try and split quantitative and qualitative data to kind of textual or numerical, but then we've also started to bring in opinion. So bringing these out as three groups where when we're doing analysis, we kind of say to people, we know our beginning and end, where we'd like to be and where we are right now. Then we start to bring out the data from people around, is it a speed thing? Is it a, I would like kind of thing. But then it's also opinion again, well, this has worked, but not really.
Carlo José (18:11.994)
Yeah.
Greg Arthur (18:29.719)
I really like this. when we were actually working together before, not on the same project, when we were back at HSBC, there was something I'd asked somebody about doing a more experience-based program, for example. I won't say any names because it wasn't a bad comment. It was just something I'd heard in a million other companies as well. And their response was, we've tried that. It didn't work. And so you tried it once and it didn't work. But there was no other context that they were willing to give around who did it.
Carlo José (18:53.345)
Yeah.
Greg Arthur (18:58.936)
What was the audience? What were they measuring? And then the amount of times you hear people saying, we've done that before and it didn't work. It's like, well, if everyone tried everything once and it didn't work, then we wouldn't have evolved as a human race. Like we'd still be sitting in a cave going, that wheel thing, that circular thing didn't work. So no wheels, no fire, just, you know, done. But.
Carlo José (19:09.793)
That would be.
Carlo José (19:20.942)
I think that brings in the kind of connection as well with, you know, data for the sake of data, that doesn't really mean anything. But if you've got a hypothesis that you need to test, or if you're testing approach A versus B, or in some cases it's called A-B testing, then it's a purposeful gathering of information and data to be able to prove or disprove something right. And I think that's one of those things that I think is often overlooked. In organizations, there's often the kind of
Greg Arthur (19:28.407)
huh.
Carlo José (19:50.103)
You needed to have it done yesterday, but you've also got to make it perfect and the quality's got to be really good. And by the way, it's got to be cheap, but it's that, you can't have all three parts of that. You're always going to trade-offs. If you want it fast, it's probably going to be expensive. It might not be great. Or it could be expensive and great and you'll get it fast, but it'll cost you a lot.
Greg Arthur (19:52.644)
Hmm.
Greg Arthur (20:06.168)
Mm-hmm.
Greg Arthur (20:12.312)
Yeah, and it's also those things like, so the whole kind of shiny objects discussion that people kind of use is just because it looks really good and just because it was expensive or it was developed really quickly by a potentially really great team or really great vendor, it doesn't mean it's right. It doesn't mean it's the right thing for your problem. And it needs to have that element of a good data capture at the beginning.
Carlo José (20:33.696)
.
Greg Arthur (20:41.199)
within the product and then afterwards to be able to say something happened or something is happening, we deployed whatever the thing is we've made. And this is how we're capturing data to say that that problem has gone away or that target of doing something faster or whatever better, whatever it's going to be, we've achieved that. So I do kind of see people falling into the trap of speed and cost being, well, if we just put more into both, then surely that means better. It's like,
All it means is you spent more money and done it quicker. Doesn't mean anything other than that. 100%, 100%. So just on that, on that kind of topic of it can be quite overwhelming. How is, or how do you deal with, especially GSK being, how many, how many people are there? Like a hundred thousand, must be more than that. People that work at, I asked that's still a lot. That's a hell of a lot of people. I mean, how do you, how do you kind of approach that with getting enough data to know that you,
Carlo José (21:11.547)
Yeah. Dangerous or something? Definitely.
Carlo José (21:28.827)
Seventy thousand globally. Yeah
Greg Arthur (21:40.994)
you can make a kind of informed decision, maybe not a perfect decision, but an informed decision. And how do you kind of stop yourself and maybe your team from getting overwhelmed with just a wave of data? soon as you ask people for an opinion, they're gonna tell you, but it can be a lot sometimes.
Carlo José (21:53.211)
Yeah, Yeah, but I mean, back to your earlier point about do we need to be data scientists and data specialists as learning practitioners? And my view has always been you need to know enough to be dangerous or you need to know enough so that
Greg Arthur (22:10.851)
Mm-hmm.
Carlo José (22:12.817)
people can kind of trick you into believing things that aren't real. So be knowledgeable and interested enough in data and all the kind of fun aspects of it in terms of data quality, data gathering, kind of techniques and experimentation, hypothesis testing, but you don't have to be doing the doing. the companies are still powered by Excel spreadsheets. So you need to know enough to be able to manipulate data, but you don't need to be the kind of data practitioner. But in terms of like, so first of all, it's interesting that
Greg Arthur (22:26.627)
Hmm.
Carlo José (22:41.884)
70,000, 100,000 sounds like a lot, but when you've come from HSBC and you've worked with 200, 300,000, it actually doesn't feel that big anymore. So there is a bit of a practice time experience. I don't get too worried about 80 plus countries, 70,000 people as much as I used to, just actually gone into something that is smaller scale. But that being said, the nuance and the complexity is different in this organization.
Greg Arthur (22:51.255)
Yeah.
Greg Arthur (23:01.036)
Yeah.
Carlo José (23:09.726)
compared to where I was in HSBC before. And so actually the number is becoming a little bit less relevant. The other thing I'll say about numbers is when you reach a company size of more than 5,000, it becomes difficult to wrap your arms around it. And I know this from personal experience, because I've led teams that have a couple of a hundred up until the largest like local entity that I managed that had about 7,000 people in it. And at that point it starts to become quite a lot. you know, it's kind of, you can imagine even the data of
Greg Arthur (23:37.163)
Yeah.
Carlo José (23:40.104)
you know, doing salary reviews for 7,000 people and making sure everything's right. And obviously we have specialists that do that, but it starts to become much, much bigger for one person to handle. So what I would say is that for people not to feel overwhelmed, it is about identifying right level of scope and what is the data that you need, but also partnering with, you know, most companies now have a data analytics team that can help you to do that kind of stuff. And they'll help you all the way from, you know, kind of framing the problem to exploring what data is available and
Greg Arthur (23:49.74)
Mm.
Greg Arthur (23:58.402)
Hmm.
Greg Arthur (24:03.97)
Hmm.
Carlo José (24:10.333)
If this is kind of a recurring data capture approach, they will help you to identify from an employee listening perspective with the right strategy. There's actually something I'm working on now, which is we have a federated organization. Everyone's measuring impact and outcomes for learning in varied ways. How do we bring about some level of consistency in approach, but also some capability in being able to gather the information together? And so that's one way to go about
Greg Arthur (24:16.512)
Mm.
Greg Arthur (24:25.09)
Hmm.
Greg Arthur (24:34.754)
Mm.
Carlo José (24:36.189)
One thing I really loved about GSK when I joined is that there is already a kind of a mindset of understanding personas, designing for the end user. There's people experience networks, which is a kind of a group that you can tap so that you can test and learn from like, you know, different ideas or solution designs and things like that. So all the kind of tools are in place for you to be able to get something done and kind of, you know, approach it, not just from a data perspective, but also from a user experience and design perspective.
Greg Arthur (25:03.81)
Mm.
Carlo José (25:04.335)
We have established personas, so we kind of understand what is this 70,000 population made of, differentiating people who are very much desk-based versus kind of on the manufacturing floor, who has mobile devices and who doesn't, are there age demographics that make a difference? So I do think that for learning practitioners operating at a global or enterprise level, there is a good level of research that you can do if you don't have it in your organization already.
Greg Arthur (25:10.082)
Mm.
Greg Arthur (25:19.649)
Hmm.
Carlo José (25:32.37)
to understand your kind of your audience. And as that kind of learning functions are supporting specific business units or sites and things like that, the closer you are to the ground, the better your empathy and understanding is of that kind of experience and what people are actually dealing with in real life. And so being able to tap into that network is quite important so that you get kind of a real life example, experience, of description or...
Greg Arthur (25:35.062)
Yeah.
Carlo José (25:58.776)
kind of understanding of what is it that people are really dealing with on the shop floor or in the office spaces or in the science labs and things like that. I did a piece of work in, in when I was in HSBC to do exactly this, because we had all these kind of assumptions of, know, my favorite one was, you know, let's just enable mobile access to our learning content because everyone's got like, you know, they're 30 minutes or 45 minutes on train. Data access is amazing. Even when you're underground in the tube, you can like listen to a podcast, whatever.
Greg Arthur (26:03.297)
Mm.
Carlo José (26:29.009)
It was a very kind of ivory tower opinion of, well, that may be for some of you, but I would say even within London, you've got, the underground doesn't always have data. Not everyone has an unlimited data plan or wifi access everywhere. And frankly, they're probably thinking about other things like watching YouTube videos rather than developing their skills or doing, forbid someone's doing their mandatory learning while they're on a train.
Greg Arthur (26:31.691)
Yeah.
Greg Arthur (26:46.442)
Yeah.
Greg Arthur (26:50.418)
Yeah.
Greg Arthur (26:55.839)
Yeah, it's the thing I hear a lot from other clients where, like you say, it's that kind of ivory tower or at least that kind of best practice, like, but for everyone, like enterprise best practice and wishful thinking where it's like, if the average commute is 30 to 60 minutes, we can monopolize that time. And it's like, no, sometimes they're having a nap on the way to work because
Carlo José (27:21.478)
No.
Greg Arthur (27:24.128)
they've been up all night with their kids or it could be that they are planning their holiday or they're catching up on I don't know traitors or Ozark or whatever it is they're watching on telly or they just want to have some music to zone out and not listen to the guy opposite them who's having a really loud phone call and listening to you know you and me banging on about data is probably something they go yeah that can wait like it's making sure that you're so I don't think the mobile access thing is a bad idea but it's
Carlo José (27:25.919)
Yeah.
Carlo José (27:42.249)
Thank
Greg Arthur (27:53.825)
but it's that wishful thinking of, we'll just capture all of their time. It's like, nope, doesn't work like that. But I wanted to touch on the things you said about personas. So I spent a lot of time, I know I kind of binned off a lot of social media recently, I, if I'm or wrongly, I've kept Twitter or X or whatever it's called now. It's a really wild west of opinions on there. I wouldn't recommend it.
Carlo José (27:57.473)
Yeah. Yeah.
Carlo José (28:18.816)
I stayed awake.
Greg Arthur (28:21.952)
All I'll say is it's a wild west of opinions. I won't go into further detail, but there's a really good UX community on there. And there's some really great practitioners, there's some really great people that talk about design and you if you can kind of find those people, it's a really nice place to be. But when the word persona comes up, among a few other phrases, 50-50 split, there is no middle ground. There is either 50 % of going, have used them for 20 years, will continue to use them for 20 years. This is the best thing I do.
really helps me with clients, helps me with quick projects, makes me understand my audience better, and I can design based on the information I get from it much more effectively. Then there's the other 50 % that kind of go, why is he using personas? This is stupid. You're an idiot. And then whatever else they say on top of that. There's no mediator going, case by case basis, it works for us. Maybe it doesn't work for another client.
Carlo José (29:07.814)
Okay.
Greg Arthur (29:18.688)
there's just literally black and white opinions, night and day. How do you, so my question to you around personas was how do you, I guess, maintain them? How do you make sure that you're consistently using them? So you mentioned about kind of office-based, desk-based or manufacturing, two very different types of roles. And I guess for different types of projects, you might still be designing for two different or those two particular personas. But how do you ensure that
Carlo José (29:18.721)
Yeah.
Greg Arthur (29:48.339)
The way you design for them is the way that somebody else designs them has the right nuances, that kind of stuff.
Carlo José (29:52.991)
Yeah, that's a difficult one. I think there is no hard and fast rule. I'm not a pure when it comes to, know, first of all, I'm not, I'd never claimed to be a design, you know, UX expert, not qualified to be able to call myself a doctor. I'm a, I'm an enthusiast. I'm a practitioner. dabble in it. And I think, you know, for me, personas is part of the toolbox.
Greg Arthur (29:57.437)
Mm.
Greg Arthur (30:04.255)
Mm.
Carlo José (30:17.183)
And, you know, for various reasons, sometimes you don't have the time, the budget, the resource, the appetite or the mood to do it. And sometimes organizationally, it's just not, you know, some appealing and people don't want to invest the time in it. So I feel quite lucky when the information is available and it's something I can reuse to just help me kind of, you know, now I understand a bit more than I did yesterday before I saw these. And is perfect answer? Not necessarily, but I'm still more well informed than yesterday. And so.
Greg Arthur (30:24.692)
Mm-hmm.
Greg Arthur (30:45.032)
Yes.
Carlo José (30:45.964)
Do I need to kind of invest more time to build brand new learning specific archetypes or personas of employees? Possibly not. And there are actually other tools, you and we have teams that are much more into process design and experience design that might have the capacity to do it. There's a concept called Gemba, which is quite a thing in pharmaceutical and manufacturing, which is when you walk the shop floor with people. And so we've got, you've got people that have the experience and skillset to do that, that will help to enrich our understanding of customer even further.
So think for me, the bottom line is that, which is use it as a tool and as an input for what you can get out of it. And also don't make it more than what it is. It isn't necessarily the silver bullet, the Bible for everything. You can't always kind of adhere to a design principle that says you must kind of strictly follow. If you read this on a persona map or a persona document, this is what you must do. Because I also think that there is still space for
Greg Arthur (31:22.91)
Mm-hmm.
Greg Arthur (31:37.662)
Hmm.
Carlo José (31:45.154)
test and learn and experimentation. I personally like the idea of borrowing from software the beta and the alpha before you release your gold version. Because those are the two iterations where you think you're nearly there, but you can make final tweaks before you put something out to market, which is your final product or whatever. yeah, so I think I don't necessarily sit in one camp or the other. I suppose in the other camp of what are personas for and why do we need them,
Greg Arthur (31:47.167)
Hmm.
Greg Arthur (32:03.485)
Hmm.
Carlo José (32:13.804)
Well, they must have other tools, right? Where they're doing some kind of user-centric design and using other kinds of inputs. So I don't think that it's the sole source of the truth when it comes to understanding customer. One of the things you can use.
Greg Arthur (32:24.476)
Yeah, yeah. think, I think I, yeah, I don't tend to disagree with that. think hard and fast, but if you, you only relied on them or as to say the night and day opinions, if you only did or didn't do them, I kind of feel like it's too, it's too restrictive as in, so my wife, you know, not every time, as often as she can when we're talking about Apple will mug me off. So years and years ago,
Carlo José (32:52.003)
Okay.
Greg Arthur (32:52.233)
the iPhone came out and I was like, God, who's gonna buy that? It's rubbish, out of Blackberry. And I was like, this is it forever, guys. I've got emails on my phone and a little keypad. And I think it got to iPhone 4. I was like, hey, got an iPhone now. It's pretty cool and there's no buttons and very exciting. So completed a 180 on that. And then same with the iPad. Saw the, watched the whole presentation. I was just like, it's just a big iPhone. Who's gonna walk around with one?
I have an iPad Pro just over there. Use it all the time, love it, it's incredible. But then the reasoning that they were giving, not necessarily for the iPhone, the iPad, at the time, this doesn't mean anything to me. I don't need it. And then when I started seeing people, it was mainly on flights. I mainly started seeing people with the editions of the iPad, especially the really massive one that came out at the start, just looks bonkers.
Carlo José (33:21.507)
Yeah.
Carlo José (33:37.443)
Yeah. Yeah.
Greg Arthur (33:50.088)
they were just watching telly on it or watching films and I was thinking you've paid 600 quid, 700 quid for a just a flat screen telly that you can put in your bag which just sounded mental. Whereas now if I was that persona or if I had that feature request or that benefit need I know how I'm using it in my personal life and I'm using it for work. So my data has shifted. My opinion has shifted. So I kind of feel like
Carlo José (34:00.165)
Yeah, yeah.
Greg Arthur (34:19.655)
when we're talking about performance consulting and personas and that kind of stuff, exactly as you said, it's one of the tools in the toolbox. then I do find people kind of trying not to get overwhelmed, so limiting tools they use, but then also, if they kind of go too broad, they're going to get a bit overwhelmed. What kind of other things are what other things, guess, of data-ish related, so not always.
Carlo José (34:27.328)
Yeah.
Carlo José (34:36.571)
Hmm.
Greg Arthur (34:47.517)
kind of hard and fast data, but like opinion-based or even things like poll surveys, like do you bring in just to give you a steer?
Carlo José (34:53.306)
Yeah. Yeah. Yeah. Yeah. So I'm a big fan of, so back to your point first about limiting your tools. I really think we should always be exploring and it's not solely for the practice of L and D. Corporate like life in general, we should always be around and testing different tools.
Greg Arthur (35:08.509)
Mmm.
Yeah, yeah.
Carlo José (35:14.308)
you know, it's a bit like saying, chat, chat, DPD is part of my toolkit now, because sometimes I just need a quick kind of, you know, inspirations, get me my first kind of 50 % of an idea, and then I'll build off of that, right? And the kind of feed that you get out of that. Now, why would you deny yourself access to a tool like that to be able to help become more productive? So I think when it comes to like, you know, things that are data centered or insight gathering centered, I use all sorts. I'm super kind of...
skunkworks to the extent that I can get away with it. Sometimes there is something at work that I want to test more broadly outside in the outside world. And I'll use like a poll in LinkedIn to start a conversation. And it's going to be small sample or sometimes larger. you know, sometimes you're posting LinkedIn perform well. Sometimes they don't, who knows what the out. But you get enough kind of interaction and engagement with people to see, A, is it a hot topic that people are interested in? And B, people have like fully formed opinions that they can share with you.
Greg Arthur (36:07.132)
Hmm.
Carlo José (36:10.585)
And then you can kind of tap your own networks. I'm sure we all have our different WhatsApp groups and things like that of practitioners that you can tap into. You've got your ex-cine group that you can tap into and kind of push out a poll or push out a comment and just kind of harvest reactors out of it. I think the challenge before was if you had a popular post, you'd have to read everything and possibly respond to everything. Now you can dump all of that stuff into chat GPT and go, okay, tell me what the group said. And, you know, what's the kind of sentiment leaning in one way or the other? What does the data say? Right.
Greg Arthur (36:29.347)
Yeah. Yeah.
Greg Arthur (36:35.057)
Yes.
Carlo José (36:39.783)
that corporate setting, I've been known to run like behind the scenes little polls or just check with kind of informal networks and things like that. At the very least to give it, you know, a bit more credibility that I've tested an idea out. And if you then need to run something much more formal with a stratified sample, then you get the experts to do that for you and use the right corporate tools to do that for you. You just can't do that recklessly and just be surveying all the time, right? But it's a bit of a balance. It's a bit of a balance, I think, in terms of
Greg Arthur (36:52.188)
Mm.
Greg Arthur (37:02.256)
Hmm.
Greg Arthur (37:05.892)
Yeah, yeah.
Carlo José (37:09.604)
Where do you get the intel about real life? know, of annual surveys are quite static and one time, and it's kind of a snapshot in time. You can get something out of that as a kind of general direction, but then you got to drill in a little deeper and understand the problem a little deeper in my opinion.
Greg Arthur (37:26.214)
Yeah, yeah, absolutely. And there was a couple of things that kind of spring to mind. So I can't find, I've looked for, can't find the exact year. I think it was around late 2000s, early 2010s. Google publicly announced they were going to become a data first organization. So they'd said, we're going to still have people working for us and they can still make decisions, but we're going to be basing a lot of those decisions or the conversations to make those decisions around data.
And it kind of didn't feel like a weird move because they're Google and they hear and see everything because of who they are and what they do. But then loads of people were like, oh my God, this is revolutionary. And I kind of just thought, it? Kind of just feels like the sensible thing to do. And now a lot of companies are kind of going with that approach of everything has to be three or four steps.
Carlo José (38:03.05)
Yeah.
Carlo José (38:19.876)
.
Greg Arthur (38:24.443)
from the bottom line, it has to be related three or four steps. So if a certain product is performing well, we kind of go back three or four steps as to what are the things that make it perform well? Can we replicate that in another area? But then when we're creating learning products, obviously, unless you are literally creating another Google, I don't think we're going to be creating anything as mad and as widespread as that. But at what point do you look in kind of the weird and unusual places to give you
Like how many steps back do you go? Or is it just a case of I need to know my problem, where I want to go to, in the middle is where I play, that's it. Like how restrictive do you kind of put barriers around yourself just so that you don't kind of spend far too long or kind of lose yourself in opinion and stuff like that.
Carlo José (39:11.435)
Yeah.
I'd like to be as unrestricted as I can be as a default setting, but I think where constraint, time, capacity, access, reality are probably the real life things that will keep you from going that much deeper, right? So, take time, because you can go deep into a rabbit hole in investigating a problem. But for me, it also goes back to this very elusive impact measurement in terms of an actual financial outcome or return on investment.
Greg Arthur (39:19.919)
Mm-hmm.
Greg Arthur (39:40.378)
Mmm.
Carlo José (39:43.975)
That classic old level four, you know, and above Kirkpatrick that, yeah, think LND still kind of struggles with. I, I, as a kind of formerly non LND practitioner are, I'm quite comfortable with that. The fact that it's quite difficult to get to because it just isn't always possible to do so. I mean, if I think back to one of the best things I was involved with in HSBC was developing a program for aspiring CEO, CEOs.
Greg Arthur (40:12.569)
Hmm.
Carlo José (40:12.616)
And it was to help us create this pipeline, you know, because in HSBC, there's hundreds of CEOs, know, there's market specific and business specific CEOs, slightly different jobs, lots of commonalities, and then a few kind of variations in terms of, from a theme perspective, this is what this person needs to be focusing on versus this person who's CEO for another business. But one of the things that we did there was to interrogate what are the kind of actual enablers and blockers for them to deliver on what they need to do.
Greg Arthur (40:31.747)
Mm.
Greg Arthur (40:41.015)
Hmm.
Carlo José (40:41.254)
And also what is it that they A, held accountable to, but B, actually delivering? Because sometimes they're slightly different. And then there's also the kind of intangible, indirect things that influence that, which it's not always easy to draw a causal relationship. It's a classic, if my employees are engaged, my results are going to be good. Well, yeah, kind of logically think that's true, but you can't always kind of, you can pay people a lot, but you can run a unprofitable business. And so that's kind of...
Greg Arthur (40:46.775)
Yeah. Yeah.
Greg Arthur (41:09.027)
Yes.
Carlo José (41:10.78)
Interesting exploration into well what makes a great CEO? and so I think you know for me we went as deep as we could in that the approach that we took was that I Interview and kind of do a bit of ethnographic research if I can call it that Not in strictest sense, but understand the life of a CEO and when it with a sample of high-performing and again subjectively identified these are high-performing CEOs that we know have
Greg Arthur (41:28.954)
Hmm.
Carlo José (41:36.681)
done the job in multiple markets and they're a good source of kind of insight and information. So we learned a lot from them. We tried to codify that into what are the different things, the capabilities and skills that are important for this job? What are they delivering? What gets in the way and what enables them to do it? And so we were able to understand that complete picture and then be able to say, okay, well, I don't know if this is a silver bullet, but it feels good. Let's test it out. And this was a program that was pre-pandemic. So was like five plus years ago. And I, you know,
Greg Arthur (41:41.207)
Mm.
Greg Arthur (41:51.907)
Hmm.
Greg Arthur (41:58.42)
Mmm, yeah.
Carlo José (42:06.179)
not that it's any kind of structured data gathering. I look at my LinkedIn feed now and I look back at the alumni from that program that I ran for a couple of years and go, so he's now a CEO and he's now CEO, he seems to be doing well. And so sometimes there is also this kind of immediacy of impact that people think, you you're to get it right away after you've done, you know, kind of learning intervention. And I'm like, well, sometimes there's a long game that's to be played here. Like not going to know, like, you know, some people
Greg Arthur (42:25.687)
Yeah.
Greg Arthur (42:31.203)
Yeah.
Carlo José (42:34.111)
Part of our program for Aspire CDOs was helping people to decide, this the right job for you? it's the right thing to Because we could be banging our heads against the wall and you're not interested or not really the right profile. It's not a good fit for you. you know, pursue something that's okay, you know, no harm, no foul. And we'll find something else that might be better suited, more interested, et cetera, right? So there was that kind of risk of also saying, well, out of a hundred people that we put through this,
Greg Arthur (42:39.748)
huh, yeah, it's a big move to make, yeah.
Greg Arthur (42:49.207)
Hmm. Yeah. Absolutely.
Carlo José (43:02.258)
We think that about 10 % might actually fall out because we've kind of selected the wrong people and this is not actually quite right for them. But that's okay, that's a win for them and that's a win for the organization, right? But anyway, I always kind of, you your question sparked the memory about that program, which I don't know whether I've answered your question really, but I went into that rather.
Greg Arthur (43:12.332)
Yeah. No, no.
Greg Arthur (43:22.296)
No, it was a great answer. It kind of got me thinking about like, there is definitely, so when we talk about, so slightly side note on that was when people talk about vehicles or formats for their learning, people tend to dump on face to face as an example. And it is an easy one to dump on because it's been done so badly for so long, largely. But I think it also comes back to your data. And as you were saying, when you expect to see that return on investment.
So if you're, oh, there's my little thing popping up. So like if you invest in a stock market, unless you know there is about to be a massive global event happening, you can see in the future, and you know that I'm gonna buy this particular stock A because it's really low right now, but I know on Thursday next week, something's gonna happen that stock A is gonna quadruple in value, great, go for it. But you can't predict the future. So I think with things like face-to-face selling,
Carlo José (44:00.266)
Yeah.
Greg Arthur (44:20.322)
people tend to take that approach of, they turned up. It's like, okay. The only data point you can give is they turned up, they stayed for the whole time, and then they left. But it's the same with the digital experience, the same with an experience designer, any kind of vehicle you put in learning through. The only data you can gather is they completed it or not, or they spent X amount of time on it. It's what you do afterwards, and it's that how long do you...
Carlo José (44:41.772)
Yeah.
Greg Arthur (44:48.6)
How long do you give them to start to change a behavior or how they think about things or the way they put it into practice? And do they need to practice it a number of times? Do they need to maybe go through the motions of getting it wrong? know, it's not everything is straightforward. And I think that's where data can be. It can be something very hard and fast at the beginning, as in this is happening. We can measure it with this. But then you talk about impact and that kind of a level four, five, I have a bit of a
Carlo José (45:11.436)
Yeah, yeah, yeah.
Greg Arthur (45:18.791)
hit and miss with Kirkpatrick's model but I think it's not a bad starting point but like I don't think it's as hard and fast as you can measure now but
How do you measure impact and behavior change, especially on a massive scale? Like, if everyone does or everyone doesn't, otherwise you need to go and figure out some more metrics. Like you can't just go, yeah, doesn't, you know, it's a, I, I know completely understand, cause I don't know if I even, don't, I don't know if anyone has a proper answer to it, but how do you look at things like impact or behavior measurement? Do you, do you break that down into something very specific or do you?
Carlo José (45:43.353)
Yeah. Yeah. Yeah. Yeah. Yeah. I mean, I think part of the challenge is that, you know,
Greg Arthur (45:59.421)
leave it as open-ended as we need to see what happens when it comes to things like impact and return on investment.
Carlo José (46:09.262)
For example, specifically on leadership, leadership programs try to cover so much because there is a temptation to cram as much as you can. practitioners and designers say, that's probably too much because you've now covered 20 topics in 14 hours of work. How is anyone going to retain anything unless you've got a real clever way of synthesizing everything? On the flip side, there's also kind of an aggregation point for
Greg Arthur (46:14.007)
Hmm.
Greg Arthur (46:17.771)
huh.
Carlo José (46:36.493)
capabilities, right? you you could be a classic example from HSBC times was around, we had a program called Leaders as Teachers. And it was a program that was meant to help our senior executives and middle managers be part of the learning and knowledge transfer and help to build that culture. And was to equip them with skills that they can use so that they can facilitate their own interactions better, but also be part of our faculty for leadership programs.
Greg Arthur (46:56.694)
Mm.
Carlo José (47:05.707)
And so the kind of broad capability was to teach, to be able to teach, but then you were teaching people some skills within that, like storytelling, kind of clarity of focus, you know, kind of managing a room and being able to facilitate a conversation, know, avoiding the temptation to be the main conversation and the kind of the source of information. You just facilitate discussion with people, that kind of stuff. And so you could almost aggregate that into...
Greg Arthur (47:27.51)
Mm.
Carlo José (47:32.367)
What is the impact of this program? Well, how many people are actually A, using these skills and running their own team meetings and interactions better, but also B, how many people are signing up to actually help us to deliver our leadership programs? Because our program approach then was reliant on we had people from the inside partnering with professional facilitators. It has to be one plus one. If you didn't have enough uptake, we've got a problem. So there was kind of a tangible hard measure of have we got enough faculty or
Greg Arthur (47:45.238)
Mm.
Carlo José (48:00.94)
have the very few enthusiasts, myself included, are overburdened. They've now got a side hustle that isn't that much of a side hustle anymore. It's becoming a main job because there's so much demand that they're getting tapped out and they're spending half their time in the training room rather than in their actual job. So I think it's kind of an, depends. And there's also kind of hard and fast measures that you can create depending on the scenario. I do think that...
Greg Arthur (48:06.933)
Hmm.
Greg Arthur (48:18.475)
Yeah.
Carlo José (48:28.43)
I think the last point I'll make is, know, level one, level two stuff is kind of interesting in the short term. But I actually think in the long term, you know, when you're happy with your core solution, switch that stuff off because you don't need to burden people with so much kind of survey anxiety anymore and start measuring the impact, the change, the behavior, the kind of post intervention stuff is much more important than the term, you know, was the air conditioning and the food good at the classroom training or did you like this e-learning? That's okay in the short term.
Greg Arthur (48:33.686)
Mm-hmm.
Greg Arthur (48:43.403)
Mm-hmm.
Greg Arthur (48:49.865)
Absolutely.
Greg Arthur (48:56.875)
Yeah.
Carlo José (48:58.55)
And maybe every now and then you'll do a bit of kind of product feedback when it's time to recycle or refresh it. But I think what's much more important is we spend our time on the level three and level four stuff.
Greg Arthur (49:08.342)
Absolutely and I think it's that kind of like you said the post release stuff is where the rubber hits the road with your audience. Just because they've engaged or paid for or whatever the thing you want them to do is because they've done that once, are they going to come back? Are they going to tell someone it was really good? Are they going to take something from your intervention or your product and is it going to benefit them in some way?
Carlo José (49:25.582)
Yeah, yeah.
Greg Arthur (49:38.934)
And if it isn't, I think it's more of a marketing thing, but they're probably more likely to tell people that it was bad than it was good. So I'm assuming that's the same with learning products as well. But again, I think it's ultimately, and I said this some reason actually was, if you're not measuring the output of your learning products, what would happen if your entire learning team just stopped turning up? Like who's gonna notice? Exactly, like.
Carlo José (50:03.383)
Yeah, yeah nobody
Greg Arthur (50:08.501)
apart from the learning team going, shit, I'm not getting paid anymore. It's like, other than that, who is really gonna care? So like, if Apple stopped making iPhones, people would notice, they'd be like, well, where's my new thing? But they, maybe they're a bad example, the last multiple versions and iterations of iPhone haven't been that dramatically different. But then,
Carlo José (50:13.743)
.
Yeah.
Carlo José (50:36.106)
Yeah.
Greg Arthur (50:38.365)
me and many other people every two years go, my contracts up, I'll get a new phone. And it's pretty much the same phone. But then they've hooked us for different reasons. But then with learning products, especially internal teams, when they're trying to justify their existence, their return on investment, and the benefit they give to an organization, they need to be a lot better at quantifying, firstly, their reason for existence and also
Carlo José (50:45.047)
Yeah.
Greg Arthur (51:06.751)
the reason these products should exist, not just, well, we need to give them something. Because you could just give them a license to masterclass or LinkedIn learning or whatever it's going to be. All are great, by the way, we're not affiliated with them, just in case anyone's going to get me on that. But yeah, if you're creating something or giving them something, give them value and let them know what that value is if it's not immediately clear. But it drives me nuts when people just go...
Carlo José (51:07.217)
Yeah.
Greg Arthur (51:34.356)
I think the Maslades talk about this a lot, go, when the learning teams go, we really enjoyed working on this, we put 300 hours of our team time into it. They're Yeah, so what? Did you waste 200 of those 300 hours? I mean, who cares? What's the thing? What's it gonna do for me?
Carlo José (51:43.98)
Yeah, and then what?
Carlo José (51:50.288)
I'm absolutely with you on this, Greg, because I think one of the kind of observations I had when I started with GSK was that we were still very much in the mindset of kind of making sure butts were in seats and we've got, know, class offerings were being utilized and things like that. said, okay, that's probably good for the short term. That probably answers the question of did we send X amount of our managers through a certain training, what have you, but then, and then what?
What happens next? How do we know that it's making a difference? So apart from being able to say that 80 % of our people went through it, we've got to be able to show something else for that. And we spend X amount. And sometimes people forget until they start looking at the overall spend, well, how much is this costing us? And what percentage of our total investment in learning is actually getting sunk into this? And what are we getting as an organization? What are we getting out of it? And so I think there is certainly...
the intersection of different sets of data that comes from employee sentiment through kind of impact analysis, level one to four through a kind of measurable operational results in behavior change and all that kind of stuff is all very interesting kind of, you know, a space that we need to be much more comfortable in as I think a learning practice. And I think from a design perspective for, you know, the practitioners of learning design and solution design, very much so, because all of this stuff feeds downstream into these things and
and the output of your work needs to be measured. It's very difficult to arrive at the end of a project or at the end of a calendar year and just be able to say, we had X number of customers, don't know what they liked, don't know what we made out of that, but there you go. This is the base level of integration, that's all I've got. And it's like, okay, that's not gonna be good in the long term.
Greg Arthur (53:36.456)
Yeah.
Greg Arthur (53:41.587)
No, I mean, if the first data point you get from say those 80 customers is they've all found out where I live and they hate me. Like that's a bad thing to find out later on. But if you can find out they all loved it and they want these little feature requests added or they've got some really great ideas for version two or they think this could be really great if you repurpose this for a different reason in a different market and it will solve another problem.
Carlo José (53:49.563)
Yeah.
Greg Arthur (54:09.949)
These are all really great things to have. But you kind of need to know either way what's good, what's bad, what's upsetting them and what are their ideas. If you're not connected to your audience and you're literally checking off, you know, number of sales or number of bumps on seats, then, man, it's wild that that still happens, but hopefully it doesn't as much.
Carlo José (54:12.505)
you
Carlo José (54:28.165)
Yeah, it's not. Yeah, it's a tough. I think different organizations as well are in different levels of maturity, right? So I think if people listening to this are saying, well, it's all well and good. You're in a multinational pharma company. You've probably got loads of people to do this kind of stuff for you.
I get it, you know, and then when I go to our kind of, you know, communities networking and conferencing and you get a sense of there's a difference between the Coca-Cola's and the pharma companies and the HSBC banks and all that, who are dealing with masses of, you know, kind of budget, but also large audiences of people. And then you've got, you know, kind of a kind of domestic player with 500 employees. I understand that it's very different, right? But I think the kind of principles around this remain the same and you just got to work within.
Greg Arthur (54:49.906)
Mm.
Greg Arthur (55:03.346)
Hmm.
Carlo José (55:14.803)
your means and within the the tools that are accessible to you. I still think from thinking back to my kind of early days in smaller companies, there is an agility that you can get away with in a smaller company that you can't do in a large organization. So you have to use that to your advantage. And you just have to be a bit more cunning in your planning in a small organization versus in a large organization.
Greg Arthur (55:25.842)
Mm-hmm.
Greg Arthur (55:34.396)
Yeah, 100%, 100%. And last thing before we go, any tips, any tools or activities that you do or that you've seen people do that you think other people could adopt? Like if they're kind of starting out on this kind of new approach of trying to make some sort of sense of their data or they want to try and up their game a little bit, what kind of advice, like real practical advice could you give them? Or do you want to give them?
Carlo José (56:01.554)
I mean, this is going to be super cheap, but you know, if you, if you don't even know where to start, get chat GPT or deep seek or whatever equivalent, and just start asking your questions there. Because I think that is the perfect place to go when you're not quite sure where to even start your unstructured thinking. And you can easily get kind of a bit of a starting point and links out to resources on the internet. If you're part of a large organization, I think there is some work to be done.
in terms of you understanding what does this organization do from a data perspective? And is there a connection to my work that can be made? Because I feel quite lucky that my team is attached to our people data analytics team. And so we have partners there that are real specialists. They know what they're doing and they understand the data structure and architecture of the organization when it comes to our people, when it comes to sentiment scores and surveys, when it comes to our performance results. And soon, because of the work that we're doing,
Greg Arthur (56:37.905)
Hmm.
Greg Arthur (56:52.401)
Mm-hmm.
Carlo José (57:00.152)
making the connection with learning activity and skills and capabilities to be able to really see that kind of connection. One of my big personal missions is to bring some standardization on how we measure impact and outcomes for learning solutions. Today we have a federated model where people are kind of doing their own separate things. Some are probably great, some are probably not so great, but we don't have a place to be able to aggregate them together, right? So I think if you're in big organization, it's about networking and understanding how that's working.
If you're in a smaller organization, if you don't have as much budget, get the free tools. It feels to me like ChatGPT paid version is kind of a tool for life that you just gotta swap it for your Netflix account or something. you've got to it's kind of one of those, it'll pay you back in dividends, and it's worthwhile investment for something specific like this or for anything in general.
Greg Arthur (57:45.488)
Yeah.
Greg Arthur (57:49.873)
Yeah, yeah, I don't disagree with that. I don't know if I'm ready to get rid of my Netflix though. I'm little bit sucked in. The only extra bit of advice I would give on that is there's a chap called Roman Pitcher, which is P-I-C-H-L-E-R, who's a product design and many other things guru. One of the, I think it's one of the free tools, I think it's on his website and you can probably just Google it as well. It's product canvases. Loads of people use them. They're really, really good.
Carlo José (57:54.944)
Hahaha
Carlo José (58:11.829)
Yeah.
Greg Arthur (58:19.717)
When I've, and they used to like the entire process, like for the whole thing of creating a program. But when you're talking about data specifically, and it's in our process as well, we kind of adapted it for us, was when you have questions where you go, I don't know the answer to this, or we found something else out and that's formed another question. Or when someone says, what's your audience type? What makes them tick? Or.
Carlo José (58:41.237)
Okay.
Greg Arthur (58:45.701)
Who do you need to work on this? What type of skillsets do you need for the type of thing you're thinking about, even for prototyping? Anything you don't know the answer to, write it down. And if you're looking at things like, if you're trying to visualize it on a product canvas, if you've got big gaps where you haven't written anything down because you don't know the answer, that's probably gonna be where you should focus on your data. So all these little tools are just ways of, it's almost like looking at a puzzle and going, I've got some massive gaps here and I don't know what's gonna go in them.
use ChatGPT, use a product canvas, use a notepad and pen just to write these things down and just go, I need to go and find out the answers to those things. And once you find out the answer, you'll either have a load more questions or you will have enough to, like we were talking about earlier, not to kind of get overwhelmed, but to kind of kick yourself off and kind of move on. But really enjoy the conversation and a lot of useful things for people to...
Carlo José (59:17.554)
Yeah. Yeah.
Greg Arthur (59:42.928)
ponder on and get started on and yeah, my final thought would only be don't assume anyone's gonna do their learning on the way to work. That's when the Netflix account comes into handy. yeah.
Carlo José (59:51.67)
You
I absolutely agree, Greg. And I think anyone that thinks that a lot of corporate learning is happening on the commute is probably sadly mistaken.
Greg Arthur (01:00:06.447)
Unless it's literally tied to your end of year bonus. It ain't Carlo, thank you so much. Absolute pleasure, sir. And where can people get you if they want to find you? If you use anything you want to plug, now's your chance.
Carlo José (01:00:21.652)
Oof, well, nope, not selling anything. Working for GSK. I'm on LinkedIn Learning, Carla Jose, and probably at GSK or HSBC, and you'll find me. That's where I've spent most of my corporate life. Thank you, Greg. This has been fun.
Greg Arthur (01:00:31.01)
Amazing.
Greg Arthur (01:00:34.541)
Mega, thank you sir. See you soon. Bye bye.
Carlo José (01:00:37.204)
Bye.