Episode Transcript
Speaker 1 00:00:05 Good morning everyone. Today I'm going to talk about a long article I recently published about the skills-based organization. And so the reason I published that article is I think there's been so much interest in so much discussion and so much activity around skills. I wanted to give everybody that follows us an update on what we've learned and what's working and try to prevent you from getting too distracted by some of the shiny objects that are going on in this area. <laugh>. So fundamentally, the idea that we're all working on that really started at least six or seven years ago, is to build a skills based infrastructure in our companies so that the management decisions and the human capital decisions and the recruiting and the pay decisions and so forth can be made based on skills, not credentials, not looks, not pedigree, not who you like, and really to use this skills or capability based people model to not only improve the operations of the company as it exists today, but also to plan for the future, to identify the gaps and to deal with all the transformational changes that are going on in virtually every part of business.
Speaker 1 00:01:29 I mean, oftentimes people use the word skills to apply to it, technology, ai, data management, data science, security, cybersecurity, so forth. But it's much, much bigger than that because skills-based analysis can be used in sales and marketing and supply chain and finance and HR obviously, which was we've been doing. So it's a very, very, very valuable idea. It's a very important idea, and we have all sorts of new technologies that enable this. So the first thing I wanna say is it's not a new idea. And if you read the article, you'll see my little historical perspective, but I mean, companies have always done this regulated companies, telecommunications companies, oil companies, manufacturing companies have certified skills and they run their companies on these certified skills, and they don't need AI to do that. These are, you know, regulated safety based, compliance based programs where if you don't know how to do this, you're not allowed to use the machine.
Speaker 1 00:02:31 You're not allowed to check in to the manufacturing line until you've been certified and trained. And so let's not, you know, pretend like this is the first time we've thought of this, it's been around for a very long time. But what's very, very different now is we have tools, technologies, AI, to infer and analyze and study skills of any type, soft skills, technology skills, business process skills, industry skills, functional skills, and analyze them and compare them and see what the adjacencies are in ways we really didn't have before. And as many of you have heard me mention this old idea that was around when I started as an analyst of using proficiencies that were designed, picked out of a book more or less, and attached to job descriptions, we can sort of let that go and, and build these skills-based organizational models in a very, very different way.
Speaker 1 00:03:33 And before you think this is a simple problem, it's not because what you're gonna find is the word skill has many, many definitions. You may say Java is a skill. And as you'll read about in our new AI white paper that's coming out, um, next week, you know, is it Java programming? Is it Java to make coffee <laugh>? Is it the city of Java? Is it Java used for front end? Is it Java used for backend? I mean, that word is sort of vague. So all of these words in skills, but that's what a skill is. It's a word or a phrase, have different meanings in different contexts. So we have to think about skills in the context in which they're used. So it isn't as simple as simply saying, you have this skill or you don't. Now, I mean, the way most companies implement skills projects or programs or systems is they pretty much have an ab yes, you have the skill, the know you don't have this skill.
Speaker 1 00:04:33 And as much as we'd like to assess them on a five point scale across the organization, that's very, very hard to do and probably not worth the effort, except in the compliance and regulatory area where it really is important. But you'll see as I talk a little bit further, there are lots of cases where you are gonna wanna do that. Now, the first point I wanna make about what we've learned is in the early days of this idea, it actually started with Degreed. I have to say they, they kind of pioneered it in the learning space. Companies really did believe, and you know, this is a very natural tendency that they would build an enterprise-wide skills model using one of these tools. But I think what we now know is that's very, very hard to do, may not be worth the effort, and really requires an extensive amount of organizational ingenuity as we say, because the skills in hr, the skills in finance, the skills in marketing, the skills in operations, the skills in sales, whatever, are very, very complex and industry specific and company specific, you know, the tools that I use in our company are not the same as the tools that you use in your company.
Speaker 1 00:05:47 So my skills to use the tools I have are not the same as your skills to use the tools you have. So we have to build, if you wanna build an enterprise-wide skills architecture, which eventually many companies will do, that's a governance process of setting up skills councils, skills committees, or what we call capability academies where business people work with you on an annual basis to review the skills models and determine what's new. One of the companies I met long ago that was very sophisticated at this was a large defense contractor. And the reason they do it is they have to keep up on the technologies that are new in order to win government contracts. The government comes to them and says, we need to build a bunch of drones and they need to know a lot about drones and the advancements in drones and the technologies that are evolving in drones if they wanna win that contract.
Speaker 1 00:06:42 So they became very sophisticated about it and their, their senior business executives work with the HR and the learning people to, uh, refresh and review these skills models at least once a year as I remember. And uh, so there's an organizational governance part to this that has nothing to do with the technology, by the way. So that's number one. Number two, how do you get started? Well, let me talk about the technology next, then I'll talk about how you get started. Now, all of the major vendors in hr, certainly the HCM vendors, the learning platforms, the recruiting tools, and the talent mobility tools, and soon enough the pay and reward tools are trying to build skills-based architectures. And what they do is they use inference technology, data scraping, and eventually now they're using AI to suck in skills data, organize it into a hierarchy, give you a tool to look at it and analyze it and continuously update it.
Speaker 1 00:07:44 It's a little bit like magic, but there's all sorts of good algorithms under the covers on how it works. We've talked to many of the vendors in quite a bit of detail and there's a level of maturity that some vendors have and others don't. It takes a lot of work and a lot of engineering on their part to do this well, and as you'll read about it in the article, what I, I've con concluded at the current state is that the platform vendors have different levels of sophistication based on the solution they're trying to sell. So the recruiting tools tend to have the most sophisticated skills analysis skills inference technology because of their problem is so massive because the recruiting problem is one of, one job got opened and there's 2 billion people in the workforce who's the right person to source <laugh>. So what they do, of course, if you've heard all our conversations about Eightfold and seek out and Beamery and Phenom and others, is they literally look at hundreds of millions to billions of employee profiles.
Speaker 1 00:08:51 Many of them use a time series to do that. So they look at the history of these people and they infer from that data what these people's skills are likely to be. They look at a lot of other data like what company they worked at, who else worked in that company at that point in time, and eventually they're gonna get into project related data to see what projects people have worked on. Because I, I think the real indicator of somebody's skills is not what company they worked for, it's what projects they didn't. So that's something I've been promoting a lot. So, so if you buy one of those recruiting products, you'll find that it's doing a lot of very sophisticated skills analysis in order to match and suggest candidates. And that means it can do skills adjacencies and it can say to you, and in fact a lot of them now have pretty sophisticated succession tools and they can say, well, here's a person that seems to have very similar to skills to this other person.
Speaker 1 00:09:46 So they could be a successor, here's a job title that looks like the same job as this other job title. So we can do an intelligent job architecture project. They're, they're very good at that because they have the most amount of data and the most amount of experience. The second category of skills tech is the talent mobility systems, the talent marketplace tools. They are also good at this, but not as sophisticated because they don't have to be, what they're trying to do is basically this person in the company, this employee is likely to be qualified for this opportunity, so let's show them the opportunity and present it to them. So they're similar to the recruiting tools from that standpoint. They do have matching technology, but they tend to be a little less sophisticated because they haven't looked at some of these second and third order use cases.
Speaker 1 00:10:36 And then the ones that are the, the most early stage I would say is the training, the learning ones. If you look at greed and EdCast and Cornerstone and all of those companies, they have now tried to build more integrated skills technology, and Cornerstone in particular is pouring a lot of money into this through the acquisition they made of a French skills inference technology. But the general off the shelf learning tools are m more fo focused on identifying the topics and and learning content and then, you know, creating matches between you as an individual and what training you should take. But they're getting there too. But, but my point is that any one tool may not do everything you expect it to do. And I have not found a vendor including many of the ones who are sort of cross domain middleware types. That really can be the only tool you have because the problem is the skills-based organization idea has to be applied to at least four major use cases.
Speaker 1 00:11:36 The recruiting use case, the talent mobility and career use case, the training and development use case, and the pay use case. And there's a bunch of other use cases, succession management and eventually leadership development, poaching and other things. So most of these vendors can't make any money selling you a generic tool unless they focus on one of these domains. So they're starting in one of these domains. That's just the way the market works and time will tell if anybody becomes the end-to-end sort of Uber technology here. As far as the ERPs, Oracle, Workday, sap, you know, they're doing a lot of work on this too, but honestly they are not as good at the AI and the inference as any of the other vendors. So those of you that have one of these, these companies products, and I know a lot of you have Workday, they're opening up their platforms to take data from these other systems so that they become more of an aggregation center and less of an actual inference engine.
Speaker 1 00:12:33 And that to me makes the most sense for them too, because these specialized vendors frankly have more experience at this and they have more domain specific expertise. So there's a lot of infrastructure decisions and I know a lot of you're issuing RFPs and looking at the different tools. And that gets me to my third point is that skills-based organizations are not companies that built a taxonomy and just use it. They focus on particular problems. They're very good at doing skills-based problem solving or projects. We call this falling in love with the problem. And this is a big, big issue in hr and I understand why this happens. A lot of you guys build stuff and then you try to find a place to use it. Well that's kind of a bad way to do product management. The way to build something is to find a problem and design a solution that solves that problem.
Speaker 1 00:13:32 And in the case of skills technology, this is a perfect example of how you can go sideways. Let's suppose you did a project for two years and you built an enterprise skills model. Now you've gotta use it. Now you've gotta apply it. Now you've gotta, you know, focus on one particular job category or one job family or one rec or one group or one department, and it may not work for that. So what we see as the most successful companies is they do projects and then they learn about the technology of the tools in the projects. And then we have dozens of these projects we've talked to companies about, they're written about in the article for example, we have a really hard time hiring AI people right now, right? I mean, everybody is trying to do that. Or machine learning people or data science people or whatever it is you wanna call them.
Speaker 1 00:14:19 Well, hmm, we need to build a little skills model of some kind, figure out what skills we use and need internally today versus where we wanna go, which you know, is a problem because in IT the skills are changing so fast, maybe we need to figure out who inside the company has some of these skills so we can at least get these projects started and then we can find out what we're missing by examining the domains of AI and looking for people who have that expertise. That's a really interesting project right now. And the skills models of AI are not clear. I mean it was, it was a little bit vague what data science meant four or five years ago. Everybody kind of just came up with that word, but they hadn't really defined it. We've got the same issue now is prompt engineering is skill is training the AI skill.
Speaker 1 00:15:09 You know, we're going through our own AI project here, which has been really, really fun. We're gonna unveil it a little bit later this summer. But we're learning a lot. There's a lot of new technologies and new ideas and new capabilities needed in these AI systems. So that's one example. Another example, as I talk about in the article, an underperforming group, any part of your company, sales, marketing, customer service, whatever it is that's underperforming, really warrants a skills project. Why are they underperforming? What are the high performing people doing versus the low performing people doing? What's different about their capabilities and skills? Are there systems? Are there PR practices? Are there techniques? Are there applications or technologies or hard skills that they have that the other people don't have? That kind of project always pays off and then you end up building a better skills model for your sales organization or whatever the group is, and you can go back and use it for training, for recruiting, for internal mobility to find people who are good for that job to find who may not fit in that group and so forth.
Speaker 1 00:16:13 So I'm going through, as you read the article, you'll see lots of examples of that. And what, what we believe here in our company, based on the research we've done in all the companies we've talked to, is that if you are not focused on a problem, the infrastructure project may or may not take you to a good place because eventually the infrastructure project is going to be applied in a real problem. And then you're gonna really figure out if your skills model is any good. I mean a really good example, we had a, you know, at the conference we had a bunch of people talk about this, the two that we've worked with a lot, BNY Mellon and p and g, Procter and Gamble, I mean they've both been down this journey. We helped both them a lot. We've been talking to them in the case of p and g, they built a corporate skills model and the interest of trying to get the Workday skills cloud fired up and, and useful.
Speaker 1 00:17:04 And they found out that what it was really the first thing they ran into was supply chain. They needed a lot of people to work on the supply chain problems they had. And because they had started the skills model and had a lot of governance and, and working with operations on that, that they could, they then used it to source people to improve the supply chain issues during the pandemic. And it became very useful and the team now became smarter and smarter and smarter about how to do this. BNY Mellon has been doing this in IT operations, looking at product management, project management, and has developed what we call capability councils or capability academies to talk about these skills and study them and then look at the, uh, interventions rather they can use to improve them. So there's lots and lots of use cases for this.
Speaker 1 00:17:51 Now, a couple more things. You know, the actual skills data is really a challenge because you may think there's 10,000 skills, there's probably a hundred thousand, you know, if you look at vendors like Light Cast, that aggregate skills from tens of millions of job descriptions, they have tens of thousands of skills in their system. That's an open free database you can download, look at, they update it every week and it's obviously generic because it's looking at the aggregation of hundreds of companies, but you could use it to build kitcher skills model started. You can get skills models from other companies. We have a very, very well-defined, well proven skills model for hr. It has 90 some odd skills in it, and we use it in our academy and we've applied it and you can benchmark yourself against it for, so for those of you that are thinking about optimizing and improving the HR capabilities, you know, give us a call, we'll, we'll take you through the model and you can use it.
Speaker 1 00:18:50 It doesn't take that long for people to assess themselves. And you can really quickly see how to use skills and capability framework in a, a very specific domain. So don't be too enamored with the one skills data set that somebody says they have. You may like it, but it's not gonna be the end. I think what we're essentially building here is a living, breathing data system that's going to be become more and more useful and more and more relevant over time. I'm gonna give you example of this. If you look at companies like Chevron, Halliburton, other companies in the oil and gas industry, you know, there's, I used to work in an oil company. There's a lot of very, very specific domain specific expertise and skills in the oil industry. Well, the oil industry is crossing paths with mining minerals, solar alternative forms of energy, and even electrical energy.
Speaker 1 00:19:49 So they're becoming very, very domain specific. Think about an automobile company that's trying to build EVs. What are the skills needed to build an ev? They're not the same as the skills needed to build an old gasoline internal combustion engine. You have the same thing in your company. And if you wanna learn about the these skills evolution, read our global workforce intelligence project research because we've looked at this now in healthcare, in consumer banking and in consumer product goods, and we're doing pharma next. And we can show you the e evolution of these skills requirements as they move from industry within an industry. And we also can prove and show you that the pace, pace setters, the advanced, more high performing companies are better at keeping up with the new skills. And they also have a different skills profile. They have skills in transformation, skills in change, skills in training, not just functional and technical skills.
Speaker 1 00:20:47 So what you're building here is not a one shot skills library. You're building a data system and a new way of using the data for these really important people-based applications in the company. So this is all kind of covered in the article. The reason I felt it was worthwhile taking the time to write this is I think people are just a little bit too confused by vendors. I think the vendors are overselling the simplicity of this. They wanna sell you their product and they're gonna make you believe that if you turn it on, all these wonderful things are gonna happen. It doesn't quite work that way as you see. And one of the things we do all the time is we go through workshops, we call 'em talent intelligence workshops. So if any of you want to, you know, have some hands on help with this, let us know and we'll sit down for a couple hours or a half day or we'll come on site and we'll take you through a workshop and help you get this started.
Speaker 1 00:21:42 That's kind of the main topic for this week. I'm going on vacation for about two and a half weeks, so I probably won't do a podcast next week unless I can figure out how to do it on the road. One more just sort of funny thing that I wanna point out that I find fascinating is the Twitter versus Threads thing. I just wanna make a point about management. I've never met Elon Musk, but I've certainly seen enough of him. He's, he's one of the most vocal people in the world and we all kind of in HR were a little aghast at how he managed the Twitter takeover and the layoffs and not paying people their severance and so forth. And now he's got a real competitor in Meta. And you know, you may argue that Facebook wasn't a very ethical company early in their days, but they're pretty well run company.
Speaker 1 00:22:32 The people that I've talked there are good leaders and as big and as fast as they've grown, they've managed that company pretty well. And there's a lot of management practices at Facebook at Meta that I think are working very well for them. And I think what we're gonna see over the next couple of weeks for those of you that are interested in the business world like I am, is how well can the meta management philosophy compete against the Elon Musk management philosophy? I would tend to bet on meta at this point, but you know, everything that we write about and we work on and we do as an analyst is focused on this. It's focused on your company's performance and what can we help you do as an HR professional to help your company outperform, innovate, grow, thrive, maintainance productivity and profitability. So let's watch these two things in the public domain and see what happens.
Speaker 1 00:23:27 Hopefully they don't have this cage match they're talking about, but I think we're gonna be able to look back at management philosophies and see how management philosophies pay off, you know, these, these really hyper successful companies in the, in the market that have trillion dollar market caps, apple, Microsoft, Google, you know, whichever one you pick. If you think about why they've been successful, it's always about their management. It's not that they stumbled across a technology and they innovated faster than everybody else. It's about how they did that and the leadership and journey that they went through as they got bigger and bigger and bigger. And, uh, so I'm looking forward to learning a lesson from Twitter versus Threads. So have a great summer, everybody. I'm going off to Europe with my wife and we're gonna get outta here and try to take a real break and we'll be back.
Speaker 1 00:24:22 I'm gonna see, I'm gonna be in London the week after next and I'm gonna be meeting with a lot of you. If you're there, come visit with us. We're gonna be there for three or four days doing client meetings, and then I'll be back in early August. So, uh, as always, check out the J B A. By the way, we have added a lot of stuff to the academy. We're gonna do a much formal, more formal launch of the new materials in the academy soon, but the price is going up in a couple weeks. So if you buy yourself an annual membership now, you can lock in the current pandemic price and we can do that for your whole company or call us and we'll take you through the overall integrated solution that we offer. Thanks everybody. Talk to you later.
