Speaker A We've come into using what I'll call Stratified systems theory be not from a requisite organization's point of view, but from the BIOS point of view of assessing capability. So we've gone from the other way into using requisite. And so I want to just tell you a bit about that and share it with you. I also think what I have to share might also throw a little bit of light on the conundrum of somebody who talks big and doesn't deliver. I think we also have a way of assessing that or knowing where that fits. So just little bit of history. SAB Miller if what you see up there puts us as number two global brewer in the world at the moment, started about 1895 as a South African organization in Johannesburg, providing beer to the miners and prospectors that came there to mine the gold. For about 100 years it remained a South African organization. And I'll tell you a little bit more about that whole process of once Nelson Mandela was freed and the first democratic elections happened. So in the early to mid 90s, we began to have the opportunity to expand outside of South Africa. Up until then, sanctions had meant that businesses couldn't do that. So it remained a South African organization up until then. And how it grew was through acquisition of smaller breweries in South Africa initially. So starting in Johannesburg, obviously they opened other breweries, but also there were other people who also opened breweries and gradually these got amalgamated together. And when there was no scope anymore to grow those, then it became a conglomerate that opened hotel chains and casinos and hotels and clothing store chains and department store chains in almost all sorts of segments in the South African economy. Now I guess the story that I think is interesting from the point of view of Stratified systems theory as I'll refer to it, requisite organizations, elliot Jack's theories, that is the whole process of how did this help to fuel the growth? How could you manage to have available the capability to actually create a very complex organization? And so we had in fact the creation of a role called consulting psychologist in the early 1960s and a person called Dr. Simon Bissieval was in fact the first person in that role. And what was set up was a process of assessment to make sure that we brought into the organization very high capable people, particularly at the senior levels. So in other words, the concern was not so much at the lower levels. There were lots of ways you could make sure you had people with the right, shall we say, skills to perform lower skilled jobs. But the problem is leadership and capability at the leadership level. So at senior recruitment level, we applied stratified systems theories to create the kind of roles we needed in the organization. We always have used this in parallel with other job gradings, with our actual job grading system. Okay? And you might think that that's strange and certainly sounds heretical if I say it in this place there's always been the worry of, as it were, the tail wagging the dog is the way it was felt. So there's a sense in which senior management have continued to have in the back of their minds or in their thinking the Stratified systems theory but not applied it directly into the grading systems. And instead the process was to say using the CPA let us see if we can bring into the business an inordinate number of people with high capability so that when the opportunities arise, we've got people that we can take chances with and give them jobs to do that they haven't ever encountered before. I think more yes, it would be that sort of concept. Okay. So as the business developed during the 19, we actually during the late 1980s, it was realized that we needed that the opportunities were going to open up and that they were going to open up in the 1990s and that we needed, therefore, to go on a major recruitment drive where every selection, every vacancy at senior level had to be filled with people with high capability, even if that was hard to do. Okay? And so they developed a talent management model and an assessment model as well as trained a whole group of external psychologists. I was one of those at that stage in the very early 1990s that was trained in this and there was this deliberate implementation of our talent management model of getting large numbers of people in. So you know what happened? 94, nelson Mandela was well, Nelson Mandela was released before that but 94 was our first democratic election and suddenly the world opened up. So by the mid 1990s we were able to start buying breweries. So we had already by that stage divested ourselves of the other interests. Not all. We've only just got rid of the hotel interests in South Africa and we've continued to retain our soft drink businesses which the soft drink businesses are linked quite closely with Coca Cola. So what it allowed us to do was to start buying breweries. And we had a deliberate strategy of buying breweries in Eastern Europe where they'd been run into the ground under the communist systems and in Africa, where also under various nationalizations that had happened, breweries had run into the ground as well. Knowing that our strength was in bringing technical expertise and actually being able to make those breweries run efficiently providing in many of those countries high quality beer that they'd never had before or hadn't had for a long time consistently. And we could take cost and redundant processes out of systems. And that was probably the story throughout the 1990s as we were on the acquisition trail and getting bigger and bigger but flying under the radar. Nobody was aware of this until we bought Miller Brewing Company in 2003 here in the United States. Now, at that point, it required a bigger group of more capability than we had to provide the assessments. And that's when I came in house and we started to look at finding psychologists in America. Our decision was to use psychologists who could be trained in the requisite organization systems and who could also do the assessment process. So we needed to establish networks of these people, and my role was in fact, to do that. So during this time, I then developed networks of psychologists in the US. It just spread out. So US latin America, as we made acquisitions in Latin America, Europe, we had already started building that network in Europe, australia, when we purchased Fosters in India, et cetera. So there was this need to build out the network of psychologists that now I think we've got around about 139, I think at the last count in my network around the globe, most of them external. And there's good reason why we wanted external psychologists. We wanted that kind of objectivity where people weren't caught up in the politics of the business, where managers can kind of put pressure on you to say so and so has higher capability than they may have. Then the problem is, we're a very decentralized business. So if anyone's worked in a decentralized business, you would know that you do not get things done by telling people to do them in a decentralized business. Basically, each business unit in each country had their own board. They ran to make profit, and they competed with each other to be seen as the best unit. So the question was, how do you get consistent practices happening across a decentralized business? And we developed a range of what we called ways, one of which was the talent management way, which we developed. And I assisted together with others in rolling these out. So I have a global role. It's one of these things where I don't have line authority. So this is an interesting one for requisite organizations. I don't have line authority over very many people. I act almost like an internal consultant and help each country or each region to implement. So they own the process, I help them to implement it. Okay? Does that make sense? And it's worked. I mean, it has its problems, but it's worked. So it's a matter of still having very clear accountability for who owns the process in each country. And so we codified a talent management way as one of many other ways, marketing way, for example, and rolled it out. At the heart of it is the use of stratified systems theory. At the heart of the talent management way, it impacts on all elements of talent management, from our talent acquisition through to the career development for people. Or we call it career development because we see it as much broader than just am I developing for my role? It's about am I developing broadly so that I'm going to develop continually in flow, if you know what we mean by flow in the requisite organization model through to the whole element of strategy. So talent strategy, how does this link and I'll show you how we've rolled that out into a global talent model probably just should take a little bit of a sideways step and say our assessment model consists of three parts. And this is probably the bit where I think we're more able to answer the question that some of that we were debating in the previous session because the three parts are cognition. And by cognition here, I'm not referring to what we're talking about in terms of what we would rather call judgment and decision making, our ability to manage complexity. So cognition is more what you would measure with intelligence tests that understands what's going on inside. That comes from neuropsychological research. And so for example, if you've got someone who is very good in an abstract way of talking about things but doesn't implement well, you likely to find when you look at their intelligence test results that they stronger on the verbal side rather than the nonverbal side and the other way around. The person who seems to have an incredible ability to make good decisions and implement them, you usually will find that they stronger on the nonverbal side and not so good on the verbal side.
Speaker B Because they often can't talk about that.
Speaker A And it's the relative mix of these.
Speaker B Things which helps you to know where to best place people with a very.
Speaker C Low validity of intelligence testing, at least the ones I see.
Speaker B Validity for what?
Speaker C Validity for predicting future behavior?
Speaker A No, it's got your highest correlation.
Speaker B If you look at the meta analytical studies, it's got the highest correlation of any single variable.
Speaker C What is that?
Speaker B Your g, your factor of g? It's about 00:50, sometimes zero 60.
Speaker C And that's a result.
Speaker B It's still a factor. And if you don't take it into account as you don't take it into account if you look simply at ability to manage complexity so if you look at the complexity but in our research, in our business, we haven't found the high correlations that are reported by BIOS and Elliot and so on, but we found the judgment decision making having a higher correlation aperture than cognition. But if you put them all together into a model but I like for.
Speaker C Those of you that don't know that the validity of level of capability is.
Speaker B Zero point 95, we haven't found that. We found it. There's only about zero point 69 in our business. But I don't know where they got the zero point 95 from that's from.
Speaker C The human capability study.
Speaker B It would be interesting to know how it's done. I can only point to our study. No, it depends on what lies underneath it because what performance measures are you using? Are you using performance or are you using long term development.
Speaker C It's the attachment of the manager once removed, the manager and the person itself of the present position, the present level of time span that you're operating with.
Speaker B Fair enough. So then you're measuring that. What we're measuring here is performance, correlation with performance. In other words, do you grow the profitability of the business? Do you create higher sales volumes? Do you know what I mean?
Speaker D I did a study for my master thesis trying to correlate cognitive ability test with a big five and also with the judgment of ability and I found no correlation.
Speaker B Yes, there's no correlation between those three.
Speaker C Things, but the correlation and the ability to predict is related to the time span of the rule, the level of.
Speaker D Capability, but not with personality.
Speaker C No.
Speaker B Isn't ultimately your ultimate measure though, is performance, surely? I mean, you do want to know whether there's a fit with capability. I'm probably running out of time. You do want to know whether there's a fit with the way the time span is. But if the person still doesn't perform and the business doesn't grow and how.
Speaker C Long do you have to wait?
Speaker B Well, you'll get various time spans for that, so you can measure it over time.
Speaker A If you've got people in the business.
Speaker B For a long time, you can have a longitudinal study. But I'm not saying therefore that correlation isn't correct. I'm just saying that you also want to relate it to performance. It's not the only thing that predicts performance. It's one of the highest. Okay, sorry, that was a bit of a d to us because I don't think this is so important. This is the black box that sits behind what we do. So if I can quickly run through the last bit.
Speaker A So what we did is we recruited a talent. We look at trying in terms of talent pools to get the right mix of talent based on Stratified systems theory. And I think that's what everybody does here. It's really trying to understand what roles are at the different levels. So what does our mix of talent look like so that we can actually have the right mix across all these levels and have that for long term planning? We look at development in terms of individual development and leadership development and using the stratified systems theory as a very strong input into that development so that people can be put into the right kind of development for their particular capability. Obviously we use it for succession planning and I'll show you in a minute the way we aggregate that up to XCOM level. And we use it for talent analysis and strategy development of our people strategies in the business. So the rollout in our global businesses I'll go through this very quickly, starts usually with assessment at the board level. If we take on a new business, we will usually start by assessing the people at the board level of the business that we're acquiring. We will cascade down as appropriate to the business priorities. We will develop capability in the local businesses, HR, talent management experts, custodians managers, et cetera. And we will develop a network of psychologists and CPA practitioners. In terms of roles, it's very interesting that I've conducted a number, and maybe this is where it would be more like what most of you are doing. I've conducted a number of studies across different countries which are really our business units, a country unit, and I found that the different roles don't always have the same level of capability requirements. It's obvious I guess, because some country units might be. For example, if I take an example of say Angola. When we started in Angola, it was fine to have the MD there operating as a high level three because basically we didn't even need to worry about marketing. All we had to worry about was production, get beer, produced as much as we could produce sold the moment you got it into outlets on the street. As simple as that. That's very different. It's not like that now. The moment the whole economy grows and the moment you have competitors and the whole external environment changes, now you will have a lot of different needs and suddenly the role is elevated. The whole business then needs to change. And so you have what we've been talking about of businesses moving from one level to another. On the other hand, a very highly complex country. You're usually going to find that the MD of that country needs to operate at level five. Does that make sense? So you'll see, MD roles are not all the same. And you could do that for any one of the roles. I've done those audits across a number of countries to see where we stand. That would be the application of the requisite organization model to that. Okay, right. And then let me probably just talk about the global talent pools and then I'll stop. The other thing that we've found necessary, which is one of the problems of a decentralized business is, if you think about it, who thinks they own the people? Each country. Each country they own the people. What tends to happen if you have one country needing someone, what happens? The MD usually holds on to his good people and he lets available, he offers to the other country the people who not so good because they're competing, they literally occur. They MDS competing with each other. So we needed to have global talent pools, to have MD succession, to have succession for the various boards, to have succession for marketing directors. You can take it all the different functions, technical, marketing, sales, you name it. And so we had to create a global talent model where we now have global talent reviews and forums for critical roles. We've got about 300 critical roles across the globe that we monitor in that way. We have global functional talent reviews and forums because we are trying to make sure we've got talent pipeline for the long term across all those functions. And for example, if I take one simple example in the technical functions where nowadays there's a big difference between ability to attract high level technical people into a brewing business from what it was in the past, nowadays those people go where do you think? Where would you predict they go? To Silicon Valley? Yeah, they go to all those kinds, so they're not as attractive. So we have to build a whole strategy around how do we now create pipeline? We were talking earlier on about HR and our HR directors. At senior level, we are expected to make level five input. Okay. And so where do we find those people? You have to have a long term plan to find those and grow and develop. And ultimately, it's overseen by our board having XCOM talent reviews. So they have a board session once a year that's focused only on talent. And they have other points at which the committee responsible will check this out on a quarterly basis. I think that's all I need to share. It might be just worthwhile just mentioning two lessons. The one is that we would see capability in terms of stratified systems theorists only, forming one part of what you need to know about people. And the second lesson would be the ongoing calibration development training of the people using it. If you don't do that, you lose the power, and I'll stop there. Any questions?
Speaker D You Sam.
