Episode Transcript
Speaker 1 00:00:04 I have, uh, a couple of really big things I wanna talk about this week, and I might po uh, publish the podcast a little bit early. Uh, the first is our organization Design Super Class. I'm gonna tell you a little bit about that, and then I'm gonna play a recording from Andy Billiau about it and why that's such an important topic. And then I wanna talk about what's been happening with AI because the AI spaces moving ahead at light speed, and we've been having lots of conversations with companies about it. So on org design, most of the post-industrial issues I talked about in the last week's podcast, which are part of our keynote from Irresistible, and a whole theme of research we're gonna be talking about in the fall, have to do with changing roles and skills and moving from a job centric company to a work and skills centric company.
Speaker 1 00:00:57 And that sounds very glib and sort of high level, but this is what's going on. All of the HR issues around talent, mobility and career and skills that you're working on, you're not doing them just for their own sake, you're doing them because we need to build organizations that are more adaptable and more customer focused and more flexible because of the nature of change going on in every industry. And that domain is called organization design. Now, organization design historically means boxes and sticks and spans of control. How many people are working for who and who's gonna be the boss? And if you read our research on org design, what you find out is that most of the org design initiatives and companies are along the lines of, well, this guy's failing, so let's make that group report to this guy or this woman and let them figure out how to do it because they're superstars and they'll make it all work better, which is really not a business centric design process.
Speaker 1 00:02:01 It's, it's really sort of a celebrity oriented org design process. And when you do org design carefully, which you find out is there's actually a science to it. You look at the business problem and the business strategy, you look at the go-to-market strategy, you look at the work that needs to be done, and then you design the organization around the, the strategy and the work and the skills and the people and you and the automation that you can bring to the group. And of course, the automation is changing so fast that the org design has to be very flexible. And what we found in the research, which Kathy published a year or so ago, is that the real issue is not who's reporting to who. It's the level of skills and the clarity of accountability of who is gonna be responsible for what. Because organizations are getting flatter.
Speaker 1 00:02:50 And the flatter your organization is, frankly, the better it probably works and the more productive it is. In fact, uh, most good org design projects result in the need for fewer people in companies. And AI is going to make this a very, very, very common occurrence, common practice. So what we do in the org design super class is we take you through, it's part of the Josh Academy, is we take you through a real org design project. You, you meet the executives in this pharmaceutical company, you get to know their problems, you diagnose their problems, and then we give you the tools that we use in our advisory practice to do this. And by the time you finish this course, you are gonna understand <laugh> org design and you're gonna be able to do it. And as Andy talks about in the audio that I'll play at the end, it's both for org design professionals, people that do this on a regular basis, but also for people that don't do it.
Speaker 1 00:03:50 Because org design is a black art and, and most of us who haven't done it a lot are kind of mystified by it. So we wanna demystify it. And so we've taken the research we've done and a lot of the consulting work we've done and a lot of the tools we've developed, and we've put them into this super class. The reason we call it a super class is because the J B A has, uh, four to six hour courses that are the certificate courses and the super classes longer and more comprehensive. And so already 350 people have signed up for the super class. There's literally no limit to the number of people that can join because we group you into cohorts. So at the end of this podcast, go to to the berson academy.com website and sign up because the superclass on org design is only one of 25 other courses that are also available and thousands of resources and assets and other things.
Speaker 1 00:04:41 And we're gonna be announcing later this year, all sorts of amazing things coming to the academy. The academy's turning into a really, really amazing place because of all the work we've been spinning on it. So a little bit on that and you'll be, be able to listen to Andy at the end. Now the second thing I wanna mention this week is ai. I mean, the AI market is frothy overhyped, but it is really, really hot. And there are multiple, many, many large language models. Google, Microsoft, Amazon, OpenAI, I B m are all offering enterprise corporate solutions for you to build AI solutions and buy AI solutions. I really recommend you read our deep dive on ai, uh, research report to get to know the domain. But let me just tell you what's going on. We're moving ahead very quickly with our project. We're not ready to release it yet.
Speaker 1 00:05:36 We're gonna tell you more about it soon. But what we're interviewing a lot of companies and what we're hearing is lots of experimentation going on. Lots of teams are coming together trying to figure out how to use AI tools in different ways, and there's a few consistent themes. The first is the technology vendors who sell AI tools, those that are not part of an existing product, don't really know how they're gonna be used. So if you go out and you start using OpenAI Chat, G B t, Bard, whatever it is, those are generic systems in some sense. It reminds me of the relational database industry. When the relational database industry was new, the database vendors didn't know what kinds of applications people are gonna build with these databases. So they just responded to the needs they had. And so Oracle and Sybase and Informix and Microsoft and Ingress and all these other companies that built relational databases ended up building different features for different use cases based on who bought it.
Speaker 1 00:06:34 And that's kind of what's going on at LLMs. Not every l l M does everything the same way. And so, but the vendors don't know all the use cases that you are going to use them for. So it's a very immature market and that means that as you get involved in any kind of an AI project where you're building something, not necessarily buying it, you're going to have to experiment and iterate on the process. For example, in our copilot, which is actually coming along really well, we've been firing all sorts of questions at it and it behaves differently than we wanted. And so we're steering it and we're teaching it how to be better at what we want it to do before we give it to you guys. 'cause we don't want you guys to mess with it when it's not giving you the best answers.
Speaker 1 00:07:21 And we're changing the UI and adding features to it to make it more useful based on the use cases that we think are the most important. And that's the really important second point I wanna make about AI is that making a comment like, we're gonna have an AI system to help our salespeople. We're gonna have an AI system to help our HR people. That's not the right way to think about this. These have to be use case based systems. Is the purpose of the AI chatbot or the AI tool to send candidate emails, is it to answer questions and facilitate onboarding? Is it to validate and support compliance? Is it to source and select candidates? Is it to find a new job inside of the company? Is it to evaluate pay? Is it to qualify sales leads? Those are each different, very different use cases.
Speaker 1 00:08:19 No one tool is going to do all of those things well, because all of them are based on really narrow, but deep data problems. And what the research and AI is now proving, and you can see this in a lot of the articles are being written, is that the really big, big, big big AI systems do tend to get stupider over time. This is why chat sheet PT four is sort of making mistakes, is that it's, it's very hard for the system to be expert on everything. And when you give it more and more and more data, it becomes less and less expert. I was talking to a, a PhD I know very well in this domain who's been doing it a long, long time, and he said, you need to think about AI like a teenager or a college kid who's learning, he's gonna be an expert or she in some areas, but not in every area.
Speaker 1 00:09:09 And when you ask him or her a question, there are times when they're gonna have very detailed information and there are times where they're not, and they're gonna wing it when they don't know the answer. So don't think about this as one massive AI system that's gonna solve every problem in HR or, or, or the, or the domain of your client. It's really going to be a much more narrow solution. And it actually works better that way because the way the neural networks work is when the domain is narrower, it gets smarter and deeper better. It's just like in the human world of intelligence. If you call me and you ask me about hr, I'm probably gonna give you a really, really good answer. If you ask me about compliance processing in the healthcare industry, I'm not gonna be able to answer at all. I'll wing it, but I'm, I'm gonna be actually a pretty poor ex source of that information.
Speaker 1 00:09:59 So, so we want these AI solutions to be narrow and defined because they're going to be iterative designs. And maybe that's the third point I wanna make. And this is something we've learned from the client conversations we're having, the vendor conversations and the work we're doing internally is that we're not building AI systems are not the traditional technology platforms where you build it, you test it, you go through a beta, you roll it out, you turn it on, and you monitor it. These are living, breathing, learning systems. They change, they evolve, they get smarter. You're always adding new data to them. You're always looking at the results to see what you can do to make them better. You're steering them, you're tuning them. It's a very different type of technology, which means that if you're going to build or implement an AI solution, even if it comes from a vendor, you want to have the data providers in your company on board upfront.
Speaker 1 00:10:59 You want to have the IT people upfront for any security issues. And you're going to have to train users that this new thing that you're rolling out isn't perfect. And you're going to have to be able to interpret the results in an intelligent way. There's a belief and certainly a lot of testing out there that these AI systems are going to make employees, you know, we're not gonna need as many employees and they're gonna automate a lot of the work that employees do. Even the AI systems that order hamburgers and hotdog, and I forget the name of the company that was doing it, but one of the retailers rolled out an AI for ordering makes mistakes and it sometimes puts the wrong things in the order. So the person who's actually processing the order actually has to be a little smarter to understand how to interpret or fix the things that the AI didn't get right.
Speaker 1 00:11:51 So I think there's gonna be a lot of end user training that's gonna have to be done on these systems too. Now, in terms of the the vendor landscape, it's moving very fast, but it's not going to consolidate very quickly. You come to the HR technology conference in Vegas in October, I'm gonna talk a lot about this also at the Unleashed Conference the following week in Paris. But there's many, many vendors and it's gonna be confusing. And I personally think a lot of the smaller vendors that are doing generative AI off the shelf tools will disappear in a relatively short period of time and they'll be consolidated because the, the real power of this is the deeper neural network based systems that are tailored and narrow in specific use cases. So if you're talking to a vendor, if you're working with a vendor, ask them what is the real problem you guys are trying to solve here?
Speaker 1 00:12:43 Can you, can you narrow it down for me on where you're gonna specialize? Because if they don't know, they're probably just hacking away. And there's a lot of hacking going on in AI right now because the technology is changing so fast. Even even the core vendors open ai, Google, Microsoft, Amazon, there's a fair amount of experimenting going on there under the covers inside of their products. So this is not going to be a deterministic type of technology. Okay, so a little bit on org design, a little bit on ai. There's going to be a couple of weeks here of summer yet left, and then come September, October, it's gonna be a very, very busy season. We're preparing to launch a lot of research at the HR tech conference in Vegas. We're gonna have a special event there, which you guys, many of you'll be invited to.
Speaker 1 00:13:32 We're gonna be previewing the systemic HR research, the first findings in October in New York at the LinkedIn Talent Connect conference. And we'll also be previewing and soon launching the irresistible leadership research later this fall too. So we have actually a lot of stuff coming along here that's gonna come out before the end of the year. But I really want you to look at the J B A because not only is the org design superclass pretty amazing, but come the next around September, October, we're gonna launch a whole new release of the J B A and you're gonna be blown away. We started the J B A as a real academy, not a bunch of courses. And, and what I used to say all the time was, this is not a course library. This is a place you go to learn. And that was really what it is.
Speaker 1 00:14:20 It's filled with more than six or 700 learning resources. We're putting J B C research in there. We now have quarterly special events in the J B A. We have senior faculty in the J B A. We have a mobile course on AI that you can take on your phone that actually is really fascinating. It's micro learning on your phone that you can participate in if you join. And we're gonna add more of those and lots of, and and additions to the capability assessment to deal with new topics like ai. So you can assess your skills in that area. So I wanna encourage you to take a look at it. You're gonna see a lot more from us. We're gonna, we're gonna ramp up the noise making on the J V A because it's, it's come along so well. There's about 35, 40,000 people in there. So it's a really fun filled place of activity and positive support and sharing of information and knowledge. So that's kind of where we are this week, and we'll keep producing information for you in all these topics. And I'll keep discussing what I think are top of mind issues in the world of work. Let me now turn it over to Andy Bidu, who's our head of the J B A and also our head of advisory. And you can hear a little bit about what this org design super class is all about. Thank you.
Speaker 2 00:15:36 So originally when we concepted this idea of super classes, we really wanted to build out something that learners could take or HR practitioners could take that would give them hands-on experience with a capability. So you know that we talk all the time about HR capabilities and we talk about gaining capability through experience. And so the super classes were really our solution for when clients come to us and want to build capability in a specific area, this is something that they can give to their practitioners to get that hands-on experience. And so within the model that they built, the first step is really defining what is your business strategy? Where do you wanna play? How do you want to go to market? What are the products and services or offerings of those products and services that you need to be able to make available to customers?
Speaker 2 00:16:17 And then how do you back into that through the way that your organization is structured? And so that's really the, the big piece of the puzzle historically for org design. You started at the other end of the spectrum where leaders have built an organization because they had a preference for, for a leader, and they wanted to put them in a growth opportunity. And so they built the organization around individuals. Our research tells us that we gotta go from top down. So as you determine what your business strategy is, and then you start to think about what the implications are on your internal operations and how you function, that's when you need to start to understand how to break the work down and then assign it effectively to individuals or teams so that the work can be done efficiently and effectively. Yeah. And so
Speaker 3 00:16:59 Go ahead. We're just gonna jump in. I just finished a call with, and he has actually been tasked by the c e O to start look at using the way that they design the company. So think about this, this financial risk averse company, everything they do, but to think differently to use or design to restructure how they work. And so it's very timely when you start to talk about, 'cause we're starting to see the c e o level conversation.
Speaker 2 00:17:28 Yeah, that's exactly right, Megan. So it's, it's much more of a business driven initiative now rather than an HR team, maybe historically an org effectiveness team kind of looking at, at workforce planning models and starting to understand how do they staff and structure teams. And this is actually to your point, being driven by the business strategy. And so that was the lens that we wanted to take when we built out the super class. And so moving into how the super class is structured and how we deliver on that, you know, we really start with, uh, a business imperative around a growing pharmaceutical chain. Our operating regionally across four different regions. And it's actually driven by the C F O who comes to an HR vp, the HR VP of learning. It says, we really need to do this smarter and more efficiently, and we need to be able to do this at scale.
Speaker 2 00:18:13 And so we really try to make the simulation based on a real life business scenario, which it is based outta our advisory practice. And then from there, what we try to do is take the, the learner through and immersive simulation where they actually get to see stakeholders talking about the business problems, what's driving the need for the org design, what are the outcomes that the business is gonna achieve rather than what the HR outcomes might be. And so that's all part of the setup. And we spend, of the, of the three primary modules, we spend an entire module on just understanding the business problem. And we feel like that's so critical for HR because so often we build great, beautiful, elegant HR solutions that don't necessarily have a business problem that they're solving. So we really flip the script on this one so that the, the learner can understand the context for the solution that they're building.
