LH132 Your People Are Fine: It's The System That's Broken with Robin Hoyle

This episode is supported by World of Learning Conference and Exhibition. Don't miss the leading L&D event returning to the NEC Birmingham on the 6th and 7th of October. Learn, discover, interact, network. Entry to the exhibition is free or upgrade for full conference access. Register now at learnevent.com.

One of the managers turned up at that coaching programme having asked their personal assistant to complete the learning on their behalf so that they got a tick in the box on the learning management system. And you kind of go, what?

Really?

Welcome to The Learning Hack, a podcast about the people and technologies that are creating the future of learning. I'm John Helmer.

Now guess what? Learning is cool. Learning is cool. Learning is cool. Learning is cool. I'm learning. Learning is fun. Knowledge is power.

Knowledge.

Education. Learning.

Welcome to The Learning Hack. Something's rotten in the state of workplace learning. At least that's the impression you get when you listen to some of the thought leaders in the space. And my guest today, a highly respected figure with a long career in learning, is not one to pull his punches in pointing out what's wrong. Luckily for us, he's got answers about what needs to be done to stop the rot. Are they good ones? Are they actually anything new?

You decide. Hack Facts.

Robin Hoyle is Head of Learning Innovation at Heathwaite International and has chaired the World of Learning Conference for some 15 years. He's written 2 books, Complete Training and Informal Learning in Organizations. He's a Fellow of the Learning and Performance Institute, as am I, and he has been arguing with the profession for 4 decades. He came into it following a spell in the early 1980s on a community program scheme running a team of graduates in a council museum's department. With his degree in drama and behavioural science, it's not an obvious background for a thought leader in learning and development, but how many people do I interview from L&D who have obvious backgrounds? Practically none, which is why I enjoy it. In July, Robin published a piece which opens with a statement that gives this episode its title: Your people are fine, it's the system that's broken.

If you pit a good performer against a bad system, the system invariably wins.

It's not just him saying that. He's drawing on a finding from Rummler and Brack published in 1990.

This is 30 years old, this research, and we haven't really fixed that yet. So something needs to change.

In a survey of UK workers published in May this year from Ipsos, Carrion and Box, more than a third of the 5,000 surveyed said their own organisation actively makes it harder to do a good job. And 4 in 10 managers would rather not be managers at all. So what needs to change? Can it change? Robin's answer is to focus on transfer, not completion. And he has this view about the learning management system.

And the LMS is one of the biggest, you know, crimes in learning and development that we've ever invented, frankly, because what it does is it creates a checkbox compliance-related approach to the business of learning and development, which is Completely 180 degrees away from what it should be.

Not a fantastically controversial view, I imagine, about that much-derided software application, but in 2026 every organization still got one. We also talk about why leadership development keeps failing to produce better leaders, about what agency means once the machine is doing the work for you, and I put it to him that a man who works for a company selling training has an awkward conclusion to face if the system really is the problem. His answer to that one surprised me. Here's Robin Hoyle.

So hi Robin, welcome to The Learning Hack.

Thank you very much for inviting me.

You wrote a piece this summer that begins, your people are fine, it's the system that's broken. What's the evidence that it's the system that's broken and not the people? After all, these things tend to go back and forth a lot. Employees blame management, management blames the employees. So what proofs would you give that don't seem like one of those sides trying to pin it on the other? And also, what made you write it? What prompted you? Was there— we have to be careful with the word prompt nowadays. Was there anything in particular that— came up that made you think, oh God, I've got to write this?

I'd say there's a couple of things really. The first is that just seeing what happens when we work with clients and where things go less well than we would've wanted them to or have less impact than we would've wanted to. Clearly, something's going on and we think the training is pretty good. I had an interview recently with Rob Brinkerhoff, and he tells a great story about how you can kill any learning and development. And because he's made his living out of doing an evaluation of standard training programs. So the training program's the same, the trainers are the same, or there's very similar standard. And yet in some places it works, some places it doesn't. What's the difference? And so that's been the kind of theme which has been running through a lot of the work that I've been doing for a good number of years now. And then along came a post from Ipsos, Carrion and Box. So Ipsos is the polling company many people will know about. And Hassan Karayan, who was the founder of Karayan Box, is now part of the Ipsos team. He was on the World of Learning a couple of, 3 years ago. And I've been following them since. And this email came and one of the features that was in there was that 40% of managers want to stop being people managers because it's too hard. The job is too difficult. So it was a combination of those 2 things that led me to do the research and to consult where I could about, well, why is that happening? So there's a number of various bits of research which says that managers are finding it really, really difficult to do this. And the systems that are in place and some of the research which is in place would suggest that it's about not just the quantity of work that managers are expected to do, but the idea that the organization has paid to a certain extent lip service to some of that cascade information coming down the organization to the idea of setting targets at the top end, which then cascades through strategies at function level, business level, country level, whatever that makes a difference to the organization. And the reality is what that research shows is that there's a big blockage there. And the Corporate Research Foundation and Rob Bryner, Professor Rob Bryner, who posts a lot about the work that he's doing under that heading. He's been very clear about how there's big gaps where people are not getting the kind of information that they need to do the job, where things are being— he talks about the idea of set and forget goals and targets where somebody somewhere says, here's a target, go and achieve that target. And then that's it. Nobody measures whether or not it's been implemented. Next year's strategy is dreamt up on the basis that, well, we've done that already. We don't need to worry about that anymore. Let's get on with the next thing. And then the challenge escalates and continues. So not a lot happens further down the line. And on a practical day-to-day basis, what I see is a lot of managers in organizations have been promoted into the role because they're really good at the job, the job that they are now managing people to do.

Yes.

What they've not been given is enough support and development about how they can then change their role. And in the absence of anything more significant than that, they carry on doing the job that they've always done. Oh, and they're also responsible for the team.

Yeah.

And that stuff. So there's a lack of focus around giving people new opportunities, new learning opportunities, which should, one hopes, enable them to do different things within their role. And then creating the space to enable them to do it. And there's a number of pieces of evidence there which comes up, but I think the one, the Rummler and Bracher stuff, which I've known about for a wee while and it caused me to go back to it, which is that if you pit a good performer against a bad system, the system invariably wins. And I think that really stuck out for me as one of the things which says, well, this is 30 years old, this research, and we haven't really fixed that yet. So something needs to change.

Thanks for saying Rummler and Bracher, because I had no idea how to pronounce that second surname.

And I may well not have pronounced it right either, John. So I take responsibility and apologies if I have got it wrong.

Yes, me too. I think it varies. Is this new though? I mean, that 40% figure, would that have been constant perhaps? Or is there something about the way organizations have changed that has made that more so? Are we requiring more managers? I mean, possibly you could look at that and say, well, 40% of people possibly aren't really suited to management.

Well, I think that's right. And it kind of speaks to one of the things that I've been conscious of for a while and I've been speaking to people about for a while, which is that the only way up the organizational ladder is to take on responsibility for teams of people. And that's the gateway to getting a pay rise in many, many organizations. And in actual fact, that's nonsense really, isn't it? I mean, it's going to create an incentive for people to move up into a role which we've all heard of the Peter Principle, people being promoted to the level of their incompetence. That's what it promotes. So people feel that the only way they have to develop their career, and career progression is still something that people talk a lot about, is to move into a supervisory role, management role, leadership role, however you wish to describe it. And they're not either equipped by the organization nor possibly equipped from their own personality or from their own skillset or their own competence or their own understanding of what the role is to be able to perform that role particularly well. And I see a lot of managers who are struggling.

Right.

And that's not— and everybody says, oh, I don't have time to do things. And what they don't have is enough agency over their work or enough autonomy within their work to be able to prioritize in a more sensible way than the way that they do it in the moment. So they're on this treadmill, not able to break off it, and therefore constantly firefighting, which means that they're on a different treadmill, but it's even worse. And that's not good.

So let's turn to narrowing on what Learning has to do with all that. So you've got a phrase, transfer, not completion.

Yeah.

That'll be pretty understandable to most people. Regular listeners to this podcast will be familiar with the importance people like Paul Matthews, who we've had on about 3 times, I think, and others place on learning transfer. He's written a book about it, I think. It's harder to measure than completions and therefore more expensive, which may be one of the reasons why people tend not to do it. I don't really want to litigate that particular why does L&D not measure issue again, but if we look at the kind of beginning of that process, what difference should this distinction between transfer and completion make to the way we design and deliver training?

Well, I think there's, first of all, there's this idea, and I've spoken about this a lot, and people who've heard me speak will be bored of me saying these things, but basically we measure the things which are easy to measure and accord importance to those numbers rather than measure the things that are really important. And one of the challenges that we've got with transfer is, you're absolutely right, it is difficult to measure. But because we are measuring the wrong things, we're measuring completion rates, we're measuring smile sheets, we're measuring whatever it is that's simple for us to measure, course attendance, et cetera. And we forget, therefore, to design into that learning process for people The workplace transfer component. We don't have a set of tasks, activities, processes that people need to go through to move from, I'm now aware of what I should be doing, to, I am now doing what I should be doing. And that's a gap within the learning design process. So from an L&D perspective, actually getting to grips with, well, what are people going to do differently at the end of this? And how do we know when they've done it? What support mechanisms are going to be in place when they're In work. And people talk about workflow learning and all of those sorts of things. And yes, it's important, but actually, it's a discipline, it's a structure. And getting people to have clarity about what's expected of them the day after the training intervention is really important. And then following through on that, ensuring that people are monitoring that, are giving support when it's required, that there are other inputs happening to just keep people on track or to enable them to ask questions. questions and to clarify what they should be doing next if they've got any kind of level of misunderstanding or misapprehension about that. Those sorts of things are kind of pushed into the too difficult box because we're focused on delivering an event or we're focused on delivering some digital learning content or we're focused on ensuring that people have completed things in the LMS. And the LMS is one of the biggest crimes in learning and development that we've ever invented, frankly, because what it does is it creates checkbox compliance-related approach to the business of learning and development, which is completely 180 degrees away from what it should be. So we've got a real issue there, which is, yeah, I'll give you a case in point. We've got a program which is entirely digital. We supplied it to an organization. The organization sent their managers through it first because we were then going to do a program around coaching for their managers, how they were going to support their team members to use the same content. And to help them apply that content to learning. One of the managers turned up at that coaching program having asked their personal assistant to complete the learning on their behalf so that they got a tick in the box on the learning management system. And you kind of go, what?

Really?

But that is a very extreme example of what happens, but it is this kind of tick box. And we do quite a lot, everybody, all of us, we all We all at times get to do some compliance training and all the rest of it. And you think at the end of it, well, none of that stuck. I just know it hasn't, but at least it's off my to-do list and I'm not going to get hassled by the team who are requiring me to finish this piece of compliance learning that I've got to do to show that I have been trained in how to spot financial fraud or something. It's just, ah. We're going about it The wrong way from a design perspective. Let's focus on what do we want people to do? How will we know when it's been done? What does good look like? Does everybody know what good looks like and can they support the individuals to achieve that? And then we work back from there. Is there an intervention required? Sometimes there is, sometimes there isn't. Is it a digital intervention? What's the media that we're going to use? What's the methodology that we're going to use to enable people to get their head around this? Does the team within which we are working need a common language so that everybody is working in a similar way and understands when we're talking about stuff what we mean? And then we fill those gaps, having first of all defined what work performance will look like when people are doing things in the way which we consider to be good practice. And that's the model.

So to come back to the very beginning of your answer there, something you said reminds me of the well-known cliché, what we can't measure, we can't manage, which always makes me think, how are you going to manage the stuff that you can't measure? What you seem to be saying there is perhaps what we don't measure, we can't manage, and we are measuring the wrong things.

I think measuring the wrong things is certainly clear. I think that's part of it. We kind of stop at, did people turn up and were they in possession of a pulse? Well, yeah, they did. So what? What are they going to do differently? Have we enabled them to do something that's different? Have we given them the space, the time? You talked about Paul Matthews earlier. He talks about capability at the point of work. And that's the idea of it doesn't matter how skilled you are if you don't have the tools and the resources to do the job that's required. And that seems to me to be where we fall down is that we don't give people the most important resource, which they need to be able to change what they do and how they behave at work, which is the time to practice, the time to make mistakes, to get things wrong, a supportive environment in which it's okay to try and fail and try better next time. That whole idea of people learning in work by definition is slightly less efficient than continuing to do the stuff that they did before. But if they're doing the wrong things, then efficiency is not effectiveness. And for the strive for effectiveness, sometimes there's a little bit of inefficiency which has to be managed by the organization in order to enable people to grow in their role. And that may require a learning and development intervention, might not. But I think most people recognize that sometimes there is a kind of, let's just get everybody on the same page at least so that then when they're trying things out, We can provide them with that support. And I think that's a perfectly valid way of thinking about the learning and development role.

If you're looking to connect with the top minds in learning and talent, there's no better place than an iVentiv event. Since 2009, iVentiv has hosted over 300 events for global heads of L&D and talent. I've attended multiple iVentiv events myself and found them to be invaluable. Less about sitting and listening and more about real impactful conversations. It's an unmatched opportunity to collaborate with leading Chief Learning Officers on the challenges they're tackling today. Join the conversation. Visitiventiv.com/learninghack. That'siventiv.com/learninghack. This episode is supported by World of Learning Conference and Exhibition. Don't miss the leading L&D event returning to the NEC Birmingham on the 6th and 7th of October. Learn, discover, interact, network. Entry to the exhibition is free or upgrade for full conference access. Register now at learnevent.com.

So to dig into your backstory a bit.

Sure. It's a long one. It's a long one, John. We could be here a while.

Well, it's not as long as mine. You came into this trade in the early '80s, I think, through a community program scheme and a That's right. That's an astonishing place to start from. What was your degree in?

My degree was in drama and behavioral science. So, I mean, frankly, from there on in, I was going to be in training. I didn't think that at the time, but that's where it came from, which is basically I could stand up in front of people and talk, and I kind of understood a little bit about what makes people tick. Not very much, I think it's fair to say, but I knew something. So that was where the background came from. And the community program, younger listeners may not remember any of this, but in the 1980s, we had a series of schemes for people long-term unemployed. And the community program was the one that was targeted at people who were not school leavers. So I had a team of people that reported to me, bizarrely, all of whom had degrees.

Right.

We had a team of 18 people working on this community program scheme, and our job was to create events and literature, interpretive literature for people visiting a 16th century museum and the associated grounds. And so we've got lots of different skills and capabilities there, everything from dry stone wallers to copy editors to marketing people. And built this team. And what I realized was that very quickly is that quite a lot of them needed quite a lot of learning about how to work because they'd come from unemployment having done a degree with an expectation that they would get a degree and get a job and they hadn't. And some of them had been unemployed for quite a long time. In the '80s, graduate unemployment was a real thing.

Yeah.

So we had a lot of work to do helping people to become a team, work together, develop stuff onto a deadline and a timeline which was acceptable to the brief that they'd been given, et cetera, et cetera, et cetera. And so yeah, I ended up kind of doing quite a lot of training and development work within there. Although if you'd have asked me at the time what I was doing, I wouldn't necessarily have known that. But because the community program schemes in that particular local authority were run by the education department, I soon got into the rest of the training and activity within that local authority, which I did for the next 3 or 4 years.

That's interesting. So when you dig into it, it's actually a lot more relevant than a lot of the backgrounds we have going. People come into L&D from some very odd places.

Oh yeah.

I've discovered. But that is pretty cool. I mean, you've got a bit of psychology in there and you're doing a lot of you're managing teams and you're actually helping people learn. And then since then you've done a management buyout, run an e-learning company, written 2 books, chaired one of the biggest L&D conferences in the country for over a decade. In all that time, does it feel as if there's been just one idea that you've been arguing for the whole time? I mean, which can feel quite a lot like that with some L&D people. Or has your view of the field changed over time? And if so, how? I mean, it must have done over It has changed.

And I mean, I'll be genuinely honest and say that when I started managing a team of people in my early 20s, I was as green as a green thing on a green day. I mean, somebody says, oh, this is how you do it. And I go, okay, I'll do it like that then. There was no kind of real understanding of how the world worked and how things could be different and could be better. So I was spending a lot of time there. But I think over the course of the last last 15, 20 years, that's kind of crystallized into, I suppose, a couple of really important things. So the phrase that I use a lot is that learning and development is about enabling people to do things differently and do different things. And as soon as you get that as a kind of framework for what you are doing, then that stuff that we were talking about a moment ago around how does this look like when it's applied to work becomes self-evident. So it's not a question anymore. It's like, obviously, that's where we're going to focus around. Then you get into the idea of, well, what enables that to happen for people. And I suppose the thing that I've been talking about a lot is around reflection. And it's partly a kind of reaction to learning styles and all of that stuff, which I'm not going to rehearse again. But within the rush to both adopt learning styles and then in many cases reject learning styles, The reflection bit is the bit that's never been that popular. And yet I think it's really, really important. The idea of being able to look back at what you've just done and go, okay, what have I learned from doing that? What am I not going to do in the future? What am I going to do differently next time I'm faced with a similar situation? And that fundamental component of learning And getting better, whether that's as part of an established course or program or just something that you are experimenting with off your own bat. The reflection piece is essential to stop making the same mistakes twice. And that idea of constantly striving to be better and being curious, I think, is now taken up more by Lots of people within the field. So good thing is ticking box, lots more people are talking about learning in the workflow, learning from experience and having a structure to that, and being able to think about how we develop capability through, first of all, a curiosity, a drive to be curious and to find out how else you could do things, but also from— a perspective of imagination. And I use that advisedly because I think one of the key skills that good L&D people have is the ability to imagine what it's like not to know this stuff or not to be able to do this task or to perform this skill. And that makes the difference many times between whether or not what they do in order to enable others to develop their their capability is effective or not. Because that first thing of putting yourself in the shoes of the other individual is really an important process. And to really understand what I would need if I didn't understand this stuff or didn't know how to do this or couldn't perform that task is really an important part of that learning designer's armory. And I think I think we've got partway there to recognizing that learning in the flow of work, doing stuff in the workplace, changing behavior at the point of need, point of application is important. I think we haven't yet got all the toolkit assembled, which would help us do that. And I think the reflection piece is the piece that some people still struggle with as a kind of discipline because they think it's more complicated than it is. And it ain't.

As you said earlier in that answer, fashions come, fashions go in L&D. But one of the things that never seems to change is that leadership development is always there. In a way, it's one of the oldest parts of training ever, going back to Aristotle and Alexander, maybe, if you want to be fancy.

Yes, indeed.

Yeah. Leadership development is the top L&D priority almost everywhere. And you say it isn't making leaders better. You cite Gartner here, making a manager's job manageable has 5 times the impact of a leadership program. And yet companies spend loads on this stuff. So is leadership training the most expensive placebo in corporate life? And what should Elodie do instead? Fix the job, fix the selection, fix the content?

I think the problem is it's all 3, isn't it? I mean, I think the first thing is, as the Gartner research shows and as other research has shown as well, is that you've got to get the job doable. It's got to be manageable. You've got to actually be able to practically, physically do it within the time that you have available. And so that's an organizational development issue. It's a workforce planning issue. It's about really being clear about what the role of the leader and the manager of a team is within your organization. So I think there's some real clarity around I think the issue with leadership development as a set of programs is that first of all, they often come too late. So the research shows that on average, people wait 9 months having been promoted into a team leading role before they get a sniff of any leadership development training. So what do they do for the first 9 months? And how easy is it then to change the behaviors and the practice and the routines that they've got into? You're not going to do it by spending 2 days at a business So what's going to change that if we wait until then? So I'm a big fan of developing leadership capabilities and skills amongst people who are not yet leaders and may never be leaders in the traditional sense of part of the hierarchy, corner office, and all of that stuff. And that's really about understanding how you communicate influence, get people to do things, get support for your ideas, all of those things which are part and parcel of leadership, but we hide that or it gets subsumed under finance for non-financial managers and stuff like that.

Yeah.

So the leadership development happens too late. The second thing is that most leadership development programs in organizations are a reward. They're not actually designed to change anything. And quite a lot of them are education-based. So they're teaching people about what they should be doing. And we're going to use a few case studies, which particularly if we've gone with some organizations from an academic background, are going to be case studies that are 15, 20 years out of date. So we're going to look at, well, what should these managers have done in this situation? Well, in 2005, what they should have done was this, but I'm not quite getting how it's relevant in 2026. So So how are we enabling those people to take those ideas, some of which are great, don't get me wrong, some of which are fantastic, but we don't follow through on how we enable people to use them in the same way as we've talked about other learning and development initiatives. So the content, questionable, some of it's okay, some of it's no longer necessary. And certainly, a lot of the stuff which people learn in the leadership development program, give them a book for goodness' sake. get it that way. Let's get them into situations where they practice, where they're put into real situations as close to reality as possible, but give them that safe space in which to make those errors. That's where leadership development could work. Make it problem-based, make it functional in terms of related to the work that the team does that they lead as close as possible to reality, but with that safety net, which means that there are no consequences for making a complete pig's ear of it. Let people get on with it and do it. Those things have a real impact. It's not how most leadership development programs are designed and developed, or at least it's not how most of them are done. And of course, leadership development is the high-ticket item in terms of budget within most organizations because they subcontract it out.

Hmm.

It's very rare that you have an internal leadership development team. Maybe if you're in one of the Big 4 consultancy companies or you're in government or something like that, you've got access to internal management leadership development resource. But for the most part, we buy it in from somewhere else and we get a generic, here's a one-size-fits-all leadership development program, now go and change things. And that seems to me to be a disconnect from the way in which learning and development happens elsewhere when it's really effective. So let's look again at that process. But yeah, I think fix the role, fix the content, and yeah, the selection of people to be leaders is universally terrible. It's either the promotion thing that we've talked about or it's somebody who's done it somewhere else in pretty much the same job with a similar team and we bung them an extra 10 grand a year and poach them from another organization. And that can work, but very often it doesn't. Very often they're coming into a new setup with a different culture, a different way of working, different sets of priorities, And they struggle to really bring the experience that they may or may not have had. Patch A, all of the stuff that we've just talked about, about how poor the development of leaders is in any case.

Yeah.

And they come forward and replicate the same errors that they've used in the past. So not great. So yeah, selection process starts from well before people are even remotely on the ladder of moving into leadership, we ought to be training people in some of those, what I call leading behaviors rather than leadership behaviors. Because now individuals who are not in the corner office, not part of the hierarchy, might be expected to lead a project or might be expected to influence people to get resources for their team or might be managing a small area of the activity independently of any fancy job title or a letter C in front of their name.

So your closing keynote at World of Learning conference, which you're chairing, is called AI Adoption and Agency. Everyone's got a view on the first 2 words, the AI and the adoption, but talk about the third of those. What does agency mean? in this context when the machine's doing the work for the learner, for the manager, and for L&D. You said that L&D moves from producer to orchestrator with AI. Is that a promotion or a demotion? Just by way of example from another register, being the producer on a movie is probably the most powerful role in film. So it's an incredibly powerful role. Being the producer on a podcast, I can tell you I've got a lot of power, but then it's only me I've got power over. Being a writer in movies, there are many kind of rather cruel jokes and anecdotes about the power that a writer has in movies, which is none. Is that kind of where we're headed with AI?

Well, I think the thing around the producer orchestrator thing is interesting. I think I'm using producer in the sense of the producer is the person responsible for content. And I think that's the significant change there, which is I think that, A, we've got a lot of content. Do we need any more really? And if we do need more content, we're going to work alongside AI to produce some of it. Unfortunately, we're going to also produce it without some of the guardrails in place. What Nick Shackleton-Jones memorably said from the stage of World of Learning, we could create more crap more quickly. And that's true. We can create garbage e-learning modules. with the quiz at the end. But the fundamental thing is that the orchestrator piece speaks to that idea of managing the learning transfer piece first, focusing around that stuff and getting that idea right. And then the agency piece and that idea of understanding agencies is a double-headed piece, if you will. So the first is a lot of research recently from McKinsey's and others around the adoption of AI within organizations and the reasons why people who've got access to AI tools don't use them. And I think it was 38,000 people in the UK, 20-odd thousand people in the US who were involved in these surveys. The surveys, self-reported, big health warning around that. But nonetheless, with those sorts of numbers, one would hope there is some significance there. And what those surveys showed us was that 46%, 44% in the UK, 46% in the US don't use it because actually they like doing the job the way they do them and don't quite see how AI is going to help them and it's not going to improve things in any way. And then if you look at the training stats around how people are using AI or being introduced to the AI tools that they've now been given access to, what we find is that for the most part, they're taught how to do do basic press the button, do things, but they are not given any control over what the AI does for them. And they don't necessarily understand what good looks like. So Dr. Phil Hardman talks about this idea of cognitive offloading.

Right.

And cognitive offloading is when you basically give the thinking jobs to the AI and you become the person who just inputs stuff, which sounds to me like data entry, which must got to be not one of the highest skilled capabilities. Accuracy is important, of course. But other than that, it's just grunt work, isn't it? So why are we doing the grunt work and AI's doing the thinking? That's terrible.

Yeah.

And can we critique what comes out of it? So my line is AI agents need agency. And what that means is we need to be able to understand what good looks like. So I use AI within our learning and development designs, within our program designs. But where we put the AI is really important. It's at the start of the workplace transfer phase. The workplace transfer phase that we designed, by the way, is twice as long as anything which has preceded it. And the workplace transfer phase is where we introduce the tools because by then people have got an understanding of what a good output looks like.

Right.

So they can critique it. They can look at it and go, well, that's garbage. And once they start to look at it, even if they're using something which is a very simple generative AI tool like ChatGPT or Claude, where you're just having a quote unquote conversation with the tool, you can actually push back. You can actually say, that's not what I wanted. Are you sure that's right? Can we just go back and find some sources for what you've just told me? There's a kind of interaction which the agent requires from somebody with agency as a human being to be able to say—

Yeah.

That's fine. You can do the grunt work for me, me. You can summarize this report for me, but what I actually need is X and you've given me Y. Well, if I don't know the difference between X and Y, then I've lost agency.

Yeah. Interesting. Personal reflection on this, although it's not going to be representative because I'm kind of a sole trader and everything, is that I've been through various sort of, I don't know whether to call them crises or just a shift in the way I've worked with it, that you get to the point of frustration. I think Most recent one, I just came to the realization this thing can't write. I'm a much better writer than Claude is, but it's a demon researcher and it's getting better and better at that bit. And then what you do after that is kind of shift, or in my case, what I've done is shift it so that I'm throwing more of the research weight onto the AI and just having to do the drafting that I've always done.

Yeah.

But from a draft and the editing and so on. And so you've got a kind of division of labor about how skilled the thing is or not. In the meantime, it's changing very fast. So it's getting better at some of the things that it isn't. And then you can start to write skills around how you work together in the new way you've evolved of working together. Now, I mean, this is going to sound a bit long-winded, but what I'm getting to is I've got complete agency in that role to redefine the roles between the two of us.

Yeah.

To what extent do you think the average employee in an organization has the ability to go through that kind of process in their personal use of AI?

I mean, I think the first thing is infrequently, I think is the answer to your question. I think some organizations are very good at that and some organizations have got people building their own agents and using the tools in that way to create things and to push back. And And to do what you've just said, I was having a conversation about this. It was not a work conversation, it was actually at a family party. And one of our relatives is a high-up guy within a global organization and he's a techie. He's been a techie all his life.

Yeah.

And he's exceptionally good at it. And he just said, look, you've got to treat AI, it's like having a really good intern. And you just think about that as the context for what you're doing, which is you're not going to just, every time an intern gives you something, you're not just going to sign it and send it to the board. You're going to read it, you're going to check it, you're going to make sure that there's no horrible errors within there that committed you to spending hours and hours and millions of dollars doing something that you actually don't want to do. You're going to go through it with a fine-tooth comb if an intern produced it, perhaps. But you need the knowledge to be able to do that. So that's the agency bit, and you're absolutely right. I don't think organizations get that, and I don't think people have necessarily the skillset to do what you've just described, Jon. I think things like, I can write better than the AI, from some of the things that get sent to me, clearly that's not the case.

I'd rather have the badly written stuff than some, I must admit, than something I instantly recognize as Yeah.

And you're absolutely right. And you can tell they've all got little tells, little signature pieces that they use in there, whether it's the em dash or phrasing that they particularly use.

It's not that. It's more about the voice, the rather overdramatic, breathless. I just said it is like working like an intern. Maybe now it's more like a kind of first jobber. It's got cleverer. But it's still, it's management skills. It does come back to management skills, I think, because I've managed people. It just I fall back on things like that. The only difference is that if I tell Claude exactly how shit I think its first draft was, I'm not going to ruin its weekend.

No, that's true. It's not going to take a hissy fit and not come in on Monday because it's got a headache. The thing that we've got, but it is that capability to push back to say, that's not very good. That's the agency part. And then there's the skill and the capability to work out how this thing is thinking. So one of the things that I always do is when it produces a piece of work for me, piece of research, or it gives me a summary of something, or it helps me to think through a particular process that I'm working on, when I have then edited it, I always repost the output that I have created from its original draft. And go, by the way, this is what I did with it. Stick this in the memory bank for next time.

Yeah, me too.

And that's just a very simple process which people— I don't know how many people do that. I have no evidence to suggest that that is infrequently or very frequently used, but it seems to me when I talk to individuals and go, well, have you done that? They go, oh, I didn't know you could do that. Oh, I've never thought of that before. So that's just one of those practice things which helps you to think To kind of get under the skin a little bit of how the AI thinks. And I'm using thinks in massive inverted commas.

Yeah, I know. It's difficult, isn't it? It's very counterintuitive because I've got into the habit of kind of, as you say, feeding back the outcome. And I occasionally think, well, I better tell Claude about what happened with that sponsorship thing or whatever. And then I think, what am I doing? Why do I have to tell this machine? Because that just doesn't make sense. And yet, as you say, It will hone and improve the outcome as you move along.

Yeah. And paradoxically, it does better if you say please and thank you as well.

So they say.

Yeah. Which is what it seems to from my perspective. So I just get into the habit of going, please, can you?

Yeah, yeah. It gets harder and harder to keep up with that.

It does. But then the flip side of that is you've got 5-year-olds starting school and things like that who don't understand that you say please and thank you because their entire interaction is with Alexa.

Yeah.

So actually, if I'm being overly polite and terribly British about the way that I interact with AI, I kind of prefer that to the alternative, which is it might leach out into my interaction with real people in meat world.

Yeah. Talking of which, my wife walked into the kitchen the other day and I called her Alexa. I'll never live it down in the family. Yeah.

Not good.

You work for a company that sells training. If training doesn't stick because the system's broken, then the honest conclusion is that the money should go to job design, IT, and operations, not to L&D, and not to Huthwaite. Do you accept that?

Yeah.

And if they follow your logic, would a learning consultant who's drunk the Robin Hoyle Kool-Aid and everything you've been talking about today, wouldn't they not be in danger of consulting themselves out of a The job of all consultants is to make themselves redundant.

So as a principle, we're not a member of staff, we're not there forever. And I see consultancies and they say, we have been working for 47 years with this organization. I say, oh God, you must have done a terrible job then if they still need to ring you up every Tuesday to find out what they should do next. That's problematic, isn't it?

Yeah.

So there's an issue there which says, actually, yeah, part of our job is— the second thing is I don't work for a training company. Now I'm going to explain that in slightly kind of gnomonic terms in a way. Well, not gnomonic, but I'm going to explain it in a slightly pedantic way, if you will. I'm employed by Huthwaite Research Group and we are a research company. And the problem is the only way we've found of monetizing the research that we've done, and it's very expensive to do it and continue to do it, is by selling it to other people in the form of learning and development interventions. So that's what we do. So the reality is, yes, I do work for a training company, but we don't just do training. And that's the important part of it, which is that part of our role in service of our clients' objectives is to say, right, well, training's part of the solution. We get that. We very much believe in this idea of having a common language across the team that you're working with. Everybody understands the terminology that we use. Some of the research that we have done over decades now has come out with counterintuitive recommendations, things that we seen successful people do that less successful people don't do. And if you say that to people, they go, oh, are you sure? That sounds a bit weird. That's not what I've read elsewhere. And you go, yeah, I know, but we have the largest single body of commercial research in the world. So we're confident that it's right and it's observation. We've not invented this stuff. We've watched what people do and now we've codified it and here it is. So that process of helping people adopt those behaviors, adapt what they do to fit with that behavioral frame to make it more likely that it lands goes beyond just training people to do stuff. It's about how do you organize the teams? What's your strategy? How are you targeting particular organizations to work with? What's the process that you're going through in terms of briefing your salespeople? What's the marketing team doing? How are they communicating that? How are your product development or service development teams interacting with the commercial end of the organization, all of that stuff. How are the finance people getting involved in this? So we will work with teams which might involve the FD, the product design specialist, the head of operations, but in actual fact, we're doing a sales training program. But we're looking at the entire ecosystem around the sales organization and saying, You need to think about this. You need to work on how can we change that? How can we develop that? And that process of working across the organization, we don't do it on every implementation that we do, clearly. It would be nice if we did, but we don't get the opportunity to do that. But we do strongly advise people to say, look, make some space for this to happen. And help people to do this. And yeah, if we've trained everybody within the organization, thanks very much. It's been nice knowing you. Send us a Christmas card. We don't want to hang around like a bad smell because we're always there. But we do, I think, if we've achieved our goals, keep in contact with people, but the relationship shifts and it changes and moves. And it may be that we continue to work with them in some other sphere, or it may be that we say, thanks very much, job done. That's fine.

Lovely. You've chaired World of Learning since 2013, I think, for well over a decade anyway.

Yeah, yeah. About 15 years we've run it.

15 years, right. So a conference program is by definition a menu of what's fashionable. This year we've got skills-based organizations, learning in the flow, AI strategy, strategy with evidence, not guesswork, learning transfer, not completion, which we talked quite a bit about. I wonder, well, 2 things really. I mean, how much has that changed in the last 5 years? Because people complain that conferences do tend to be a bit repetitive. You don't need to go to them every year because they'll be asking the same questions and probably giving the same answers as well. And the other thing is, what do you want people to leave the NEC with after this year's show, the one thing that if they did it differently on Monday would make the trip worth it?

Well, I think there are some changes. I think the technology piece is obviously one of the things which has changed significantly. And AI is sticking around because we actually canvass the people who've been to the event, not just the last event, but over a number of years about what's important. So then I don't get involved in that, The team at venture marketing do that and they come back with a kind of long list of these are the sorts of topics. Let's now go and see if we can find interesting and exciting people who represent the good practice or experience or stories to tell and experiences to share across the L&D world. And so the technology thing, I hope in the last 15 years, with all conferences, there is a tendency to go, here's a new shiny thing, let's have somebody talk about that. And I hope what we've done is moved away from the new and shiny but not proven into let's talk about the stuff that people are really interested in and focused. And it's why AI is playing quite a significant role. I mean, A couple of 3 years ago after the ChatGPT launch and all the hullabaloo and everybody, every session was about AI and everything was very exciting. Not just, not I hosted within the world of learning, but in every other conference program it was AI this and AI that and AI the other. And we did a conference where I said, look, last year we did a lot on AI. This year, let's not do any stuff about AI this next year and see what happens. And inevitably it kind of crept in a little bit, but the reality was it keeps coming back because people are still puzzled. They're still hungry to find out, what do I do? How do I do? What do I do differently? In terms of what I want people to take away, I think it's that idea of looking at learning and development as one part of the jigsaw within improving the performance of their team.

Right.

And therefore, using their influence of being the custodians of that bit of the jigsaw effectively to make sure that there's a slot for it to fit And if that sounds a little bit highfalutin and a little bit kind of theoretical in the sense, what I would really want people to do is to very definitely get hold of one or two things that they can do differently this week. And that's what I— when I brief speakers, I said, that's your job.

Yeah.

Your job is to give them the one or two things that they could do do the day after or the week after they've been to the conference. So, what is it? What's the takeaway that people could actually do? And some of that will be about getting their data right. Some of that will be about understanding the organization strategy and where that is effective and working, and sometimes where it's not and where there isn't enough evidence to support the strategic direction, and therefore push back and question. I think some of the role that L&D should be playing, people teams generally, should be playing is to be a kind of thorn in the side of the organization a little bit. Jiminy Cricket, we're going to ensure you won't do that. That's our job in part. But I also want to help those individuals to kind of make sense of the change which is going on.

All right.

I think we have a lot of organizations which are staffed by a lot of people who are ever so slightly confused at best and scared to death at worst. And that level of change, whether that's geopolitical, whether that's to do with technology, whether that's just to do with the economy, is infecting and affecting the way that people do things. And so part of L&D's job is to say, look, here's the direction, here's the direction of travel, here's how we're going to help you get there, here's how we're going to help you individually cope, succeed, thrive, delete as appropriate. And we're going to help you do that in a way which makes sense to you and is deliverable and scalable. And if we get that sort of conversation going around the organization rather than just, here's this course that we want you to do, please sign up on the learning management system and complete by the end of the month.

Sure.

If we can move from that into something which says, well, hang on a Wait a minute. Is the solution that's required to the challenges that you've just outlined to me really a learning— is it really about capability competence, competence of the individuals? Is it stuff they just can't do, or is it stuff they're not able to do because you don't let them? Let's ask that question first, and then we'll decide whether or not learning and development interventions are necessary. If we get people who can ask those questions and be confident that they've got a solution and they've got a community behind them who understands, and hopefully some contacts so they can ring up or message or get on WhatsApp or whatever it is and go, help, from time to time, then I think we've done a job.

So always the last question. Well, not always the last question, but for fun guests, this is the question. What's your go-to karaoke song? Now, your musical taste I read somewhere runs from Malian blues to Ska and reggae to Underworld. So what is it? What do you reach for?

Well, it won't be a surprise to anybody that I'm not going to try and sing along to Underworld, or indeed Chemical Brothers or any of those other greats of superstar DJs. I'm also not going to get involved in trying to sing any of the Malian blues, which will be a great comfort to everyone. But I do think that if I were ever unlucky enough to find myself in a karaoke environment, I have to say it's one of my worst nightmares. But I have done it from time to time. And my go-to tracks are either Take Me to the River, that's the Talking Heads version for those of you who are supposed to know this.

Not the Al Green version?

No, not the Al Green version. And it's too high register for me. And I may also go with Mustang Sally, and that will be the commitment.

Oh, that's right.

I quite like that one. Yeah. So those will be my go-tos. I think I'd probably belt those out without embarrassing myself too much.

Well, I'd love to be in that pub. And we should all be in the pub. Thank you very much for that, Robin. I hope the conference goes well. And thanks for taking the time today.

You're more than welcome. Thank you for inviting me. Thanks for listening to In Your Feet for this time. Thanks to our guest and to our sponsors. We'll be back In Your Feet in 2 weeks' time. Meanwhile, stay classy, learning people. Now I finally get it. Get it.

Let me ask you, have you learned anything from this podcast? And are you wondering maybe whether a podcast could help your people learn stuff? Well, when I'm not making my own shows, I walk organizations through making podcasts for learning of their own in-house. And now I've written a paper setting out the evidence for where it works and where it doesn't, together with 10 questions that will help you decide. It's free to read or download. No form, no follow-up, at johnhelmerconsulting.com. The Learning Hack Podcast is supported by Learning News, the learning sector's newswire. Rob and his team are good friends of the podcast, and we really value the help and advice we've had from them, and they do a great job.

For the very latest news from around the learning sector, for interviews with learning leaders, the latest from learning sector vendors, and features on workplace learning, go to learningnews.com.