Episode Transcript
[00:00:00] Hey everybody. Today I'm going to spare you the geeky technical stuff about AI and talk to you about some business things that impact us in hr.
[00:00:11] And that is that despite what everybody thinks is going on, we are not implementing AI because we have an AI strategy. We are implementing AI because we have a business strategy and we are deploying and using AI to advance the business strategy. In other words, rather than looking for use cases for AI as if we're being pushed to use it, we need to take a good look at our company and where we're trying to go and what we're trying to accomplish and then decide how to best apply AI to accomplish those goals. Which also means that this original concept of turning on the AI and suddenly improving productivity and laying off a bunch of people is really not the right approach. And I want to give you three examples of this that'll make this very simple. Number one, I was in Vegas this week and by the way, I'm going to be in Vegas next week at the Transform Conference and we're going to have a whole session on AI architecture and interoperability between agents with a company that we're partnering with by the name of Lena. So for those of you that are in Vegas or coming to Vegas, please come to that session. It'll be a whole hour. Okay, so the first company, this is a large services company, multi tens of billions of dollars of revenue, publicly traded. The Chro is a very seasoned public company. Chro, we had a two hour conversation at dinner and what they do is they're in the professional services, outsourcing and strategic services business. So they're technology enabled services. I won't mention who it is because I'm not sure if they want me to talk about them that way. But you can get the gist. And her philosophy is that the HR function is responsible for the supply chain, the inventory and the product because the product is people. And by the way, engineering companies that develop software have a lot of the same issues. If the people are not aligned to the right strategy, the product fails. But in this case it's a very direct relationship. So what they did over the last couple of years, couple of years is they did what most companies do is they looked at their job architecture. They found out that they had two or three job architectures. One used by recruiting, another one used for internal, and a third one used for pay and rewards. Consolidated them together and spent a good 12 to 18 months building a skills taxonomy and agreement on the skills process so that they could understand the capabilities that they are selling to clients directly related to the projects and the revenue that they deliver. So if a client asks for a service that's of a certain type, they want to know what skills they need to deliver that service and therefore the economic value of those skills. And if there's a lot of demand for that type of service, then then they will train more people or hire more people that have those skills. So this is a skills as a product strategy makes perfect sense. A lot of you may not have thought about it that way, but that's the way this works. A capability that people have turns into a capability that the company has and that turns into a product or a service or a revenue stream. Now, in the process of doing this, of course they discovered what most of you may have discovered on your own, that this isn't very easy to do in a vacuum. In fact, it's kind of a bad idea. So the only way that it's going to stick, since the skills are constantly changing, is to find experts who can be the skills owners or domain owners or functional career experts in each of the technical and professional domains that they do business. So, which is, by the way, something that's come up in many, many projects and many, many companies that are professional, service and engineering centric companies where you need a domain leader or domain expert to define and maintain the currency of the skills. So anyway, so they went through that and they're in the process of creating that. And the third thing they're doing is they're going to use this data to price and automate the process of developing proposals and analyzing revenue because they use the project management software from their HCM erp. So the project itself is staffed with people who have skills in the system all, you know, something that contractors do all the time. If you've ever had a big remodel in your house and gotten a really good contractor and they give you a priced out proposal, what you're going to find is that in that proposal for, you know, rearranging walls and building bathrooms and redoing lighting and changing the roof, there is a list of skills and the number of hours for each skill under the covers. And they know either the person or the rate they need to pay for that skill. And they apply overhead and various profit margins on top of that. That's very common in construction and engineering. That's where she's going now. The reason I bring this up now is not only is that interesting, but now how does AI help that? Well, as I talked to her for a while and really got into this. What I explained to her, and she didn't really understand this yet, is that an AI engine learning like any of the LLMs and Galileo is particularly good at this because it's trained on all the skills. Data can actually do this in near real time. And she can subscribe to or buy data from Lightcast or Drop or Talent Neuron or other companies that will literally keep her experts up to date on the trending skills every single day. And the way those companies do that is they're scraping job postings and they're aggregating them and they're selling that data. And by the way, a lot of this is also in Galileo, this data. And so she can take this process, this business process that they're building, and they can actually get data feeds on the new skills needed in the market and in areas like technology and AI. This is happening all the time. New products, new technologies are being invented and customers either do or don't know about them and will pay extra for people that can develop or deliver those kinds of services in, in life sciences, the new sciences and gen genetic engineering, and in various biologics and drugs and RNA and MRNA and all that are also constantly changing. So I explained to her that this can become much, much more dynamic than she had really realized and that the AI can identify and actually I wouldn't say quite certify, but pretty close the skills in the workforce by monitoring their work and monitoring their activity and their documents and their emails using the type of technology we have in Sana and the technology in the digital twin. So anyway, we had a long conversation about that and at the end of that dinner, she was pretty excited about applying AI to this problem. So that's number one. Number two was a bank that I talked to today who's one of the top 20 banks in the United States. They used to be a regional bank. They're much bigger than that now. And they're going through a significant transformation to improve their technology, improve their customer experience, and improve their advisory capabilities. The business that they're. By the way, there's lots of banks, different banks specialize in different things. Some focus on investment banking, some on consumer, some on commercial, some on real estate and loans, et cetera. This bank is a very successful bank. Price to earnings ratio is way above the average for banks. In other words, they're performing at a higher rate. They've acquired a lot of banks, which is very common in the banking industry. And they are focused on a whole series of business initiatives to improve Profitability and differentiate themselves from their peers because they're, you know, there's some big players obviously out there and that those initiatives, which there's a lot of them, but the main one is improving customer experience in a positive way and improving advisory capabilities for high net worth individuals and business leaders which in turn results in reduced risk and higher margins in different parts of their business. So anyway, there's that and then they said, she said, well you. And then of course there's AI. And so she's not looking at AI as a strategy in itself. She's saying how are we applying AI? So there's a whole series of projects in the company to use AI in hr. The more significant ones are an entirely integrated employee portal to allow employees to get a fantastic experience so that they in turn can give a fantastic experience to customers. By the way, this works in banking. There's a bank in Western Canada that I interviewed years ago that it turns out has the best customer SAT rating of any bank in Canada. They're not super big, but they're very focused. And I interviewed the head of HR there a while ago and she said the reason our customers are happy is because our people are happy. We spend a lot of time making sure our people are happy and that in turn makes our customers love us. And for those of you that have done a lot of business with banks, complex things for your family, or maybe you have a death in the family or you're opening a new bank for your kids and you're trying to figure out this, that and the other thing, getting a mortgage, managing your business, paying your taxes, et cetera. When you get good service, it is way different than when you don't. So anyway, that's one of their mantras and they're very big, so it's not easy to do that when you're growing fast. The second area of course is this issue of advisory services and quality of advisory. And what they're doing there is really fine tuning their recruiting and their internal mobility. They've built a talent marketplace. They have a high powered skills engine that they use for internal development so they understand the skills needed in various jobs. And they are now investing in a much more integrated end to end recruiting system. And they've outsourced a bunch of their L and D to a specialized firm to help them with highly focused L and D in new and some of the new topics they need in the financial services industry.
[00:10:04] And the reason I've mentioned them is that again I'm speaking to their leadership team in A couple of weeks is they don't want to know about AI for AI's sake. They want to know how companies are applying it to solve critical business problems. Which is a second reminder that this AI strategy isn't about AI, it's about your company. The third one is another interesting one that came up today in a bunch of meetings we do on Fridays. On Fridays we have a series of meetings with about four or five hundred people where we go through various short sessions, one hour per week, which we call the Big Reset. And this is a company in the Middle east that's a delivery company, like Uber type of company that delivers people and food and other things. And they have a very, very business focused HR team. From what I picked up and what they do in L and D is they took L and D and they basically took it apart and they said, we don't really want L and D to be delivering L and D, we want L and D to be improving the business performance. And so since they have an AI native platform and they can generate content very quickly, they don't have the big overhead of the traditional publishing process of training. It took the instructional designers, IO psychologists, an AI engineer who knows nothing about hr, somebody who's actually an engineer, and a few of their business partners. And they created a pod, cross functional pod, which is one of the things we recommend in in systemic hr. It's a really, really powerful thing to do. And they decided that the way they were going to operationalize this SWAT team was they were going to give them six month assignments to focus on an area of the business that has a measurable performance improvement opportunity and go and live in that part of the business for six months and then leave and make sure they measure and document the outcome in a business language. So for example, one of the areas that they're focused on, one of the projects is sales. In their particular case, sales means the number of restaurants signed up to use their network. That's very easy to measure. Number of restaurants per week, per month, per quarter. Well, as you can imagine in sales, as I've seen in every other company I've ever worked in, salespeople never have quite enough training. It's always. There's always something they don't really understand because the market's changed or the customers changed or the products changed or the competitors changed. And so they took this team and they started focusing on sales. And sure enough, what happens when you go to market this way and you have the power of AI, which makes you more agile and faster, you don't try to sell a solution that's predefined. And one of the problems with L and D, and I know this because I've been involved in L and D for a long time, is that L and D specialists and professionals live, eat and breathe L and D. So of course every problem is a training problem.
[00:12:57] And you could argue that regardless of the problem, high turnover, low performance, reduced sales, quality problems, products are late, whatever it is, you can say, well, it's all about the training. They need to learn better skills, they need to learn better tools, they need to learn how to collaborate, et cetera. Well, of course there's some truth to that, but you could also argue that it's a management problem. You could also argue that it's a tools problem. You could also argue that it's a culture problem, a rewards problem, a team management problem. I mean, there's a diversity problem. I mean there's dozens of things that come together in every solution. And because this team has L and D expertise and has an AI tool that can generate content quickly, they can put in place a dynamic enablement system that gives people access to information in real time and maybe doesn't require any courses at all. And I think, and I was talking to Workday about this, that the dynamic enablement new life of L and D is probably a five times bigger market than training. Because as you know, I mean, the problem with training is a lot of it doesn't get used or finished or completed, but it takes a lot of money and time to build it. So we build more and more and more stuff and we have, you know, sort of like a giant wall of books and nobody reads all the books or maybe a few people read a few books and a few people read a bunch of books and then somebody, you know, opens one page and doesn't read it at all. So this cross functional team is able to leverage AI, not necessarily force everybody to use it to go to market in a much more business centric way. And this idea of only doing it for six months, with a deadline to get an outcome done in six months and then moving on to the next project is one way to really deploy these new tools strategically. Now, none of this means that you don't need an AI architecture and you don't need a long term plan. I talked a lot at Unleash about the architectures of AI and if you're interested in digging into this, please just give us a call, we'll walk you through our blueprint. There are a lot of architectural issues. There's the issue of AI system agent development, content management, supervision by the experts who control the AI or supervise the AI. There's the orchestration within the AI of how it makes decisions and then there's the orchestration between AI systems. All of that is new. It is a different way of building software because we're not building a predictable one. Size always does the same thing every time system. We're building a system that adapts to the data, adapts to the needs and learns. So you still need an architecture. But my point in the last 15 or 20 minutes is that is not the goal, that is the means to the goal. The goal is to find areas of the business with near term opportunities and then plan the AI implementation along those ways. Hopefully these three simple examples get you thinking that way. If you want to learn more about architecture and you're in Vegas this next week, come to the Transform Conference. I'm going to be talking about it there for a good hour and showing you a bunch of pictures of this stuff. Or get your hands on Galileo, which has all of our research in it, and just ask Galileo the most complex question you can possibly think of. Ask it to build you a development plan, ask it to teach you about how to deploy AI or whatever is on your mind and I promise you you'll get a great result. That's it for now. Have a great weekend everybody. Talk to you next week.
