The AI Revolution in Talent Acquisition, and Perspectives On Tech and M&A

Good morning, everyone. It's the week of the HR Technology Conference in Vegas. So for the next couple of weeks, there's going to be a lot going on. The Workday Rising Conference is taking place at the same time. SAP Connect is a few weeks later, Unleash in Paris. I'm going to be doing an HR Technology Conference in Hong Kong and lots and lots and lots of things to talk about. What I'm going to talk about this morning is three things. I want to talk about the tech market in general. I want to talk about the revolution in talent acquisition, which is a big body of research we're just publishing this week, which you can download. I'll put the links in here. And then I want to talk about this very thorny issue for the big HR tech companies, ADP, Workday, SAP, Oracle, DAFORCE, etc., of buying companies versus building. And that includes UKG and many others. So I got a lot of things to cover in 30 minutes. And I hope you find this interesting. Okay, let's start with the tech market in general. If you read the article that I wrote last week, there was a $300 billion deal announced that appears to be open AI purchasing data servers and data software from Oracle. Oracle's a $59 billion annual revenue company. So that's the equivalent of five years of revenue in one contract. I don't know how many of you have ever done a deal like that. I don't know how any company can actually execute a deal like that. But the two things to take away from that, number one, the number billion doesn't mean anything anymore, because these companies are throwing around billions the way we used to throw around thousands. It's not clear whether open AI has $300 billion or will have $300 billion to execute on this. But you can see the speed and volume of investment going into the AI technology infrastructure. If you total it all up, if you read the article, that deal, all the announced acquisitions and rather infrastructure deals by Meta, Google, Oracle, Amazon, Microsoft, it's about it's almost $800 billion. And that is more than 2% of the entire United States GDP. And I'm not including the energy infrastructure and the talent cost of all of this infrastructure. I don't think based on my research, we have ever invested 2% of GDP in anything since the Apollo space program, which was taking place when I was born to put a man in space. Because at that point in time, we believed that if the Russians put a man in space before we did, we were going to be destroyed. So this is a very, very interesting unique situation we're in. You could make an argument that there's a lot of overselling here and overinflated valuations and prices. But I'm not the one to make that judgment. Investors will make that judgment. And obviously investors don't care. Although I have to say the price of Nvidia sort of flattened out because now that they have some competition. So be that as it may, for those of you in HR and HR tech, this is a massive, massive, massive change going on in the infrastructure of our companies. And you're going to see a lot of things about it. By the way, I had a very interesting conversation with Kathleen Hogan from Microsoft. I won't tell you any specifics, but about how Microsoft views HR and AI transformation very similar to the way we do. Okay, number two, recruiting. Now, you know, you could argue, and I think it's not clear to everybody that talent acquisition is a niche of HR. But actually, it's not talent acquisition is a business function. It happens to be run by HR, but it's really a business function. Because every company has to hire all the time, even when you lay off 4000 people like Salesforce did the week before last, you're still hiring because you're hiring particularly talented people in unique areas. There's turnover. There's need for managers and leaders in certain areas that are new. So we're always hiring. If you look at the economic data on hiring, roughly 20% of the US workforce turns over every year, even during recessions. Right now, there aren't a lot of jobs being created. But that doesn't mean there aren't a lot of people changing jobs, even if there are no jobs being created, even if the economy is shrinking, there are people changing jobs. And that means there's recruiting going on in order to find those people. And for a very large segment of the workforce or the business community, those companies that are primarily front office organizations where they're in retail, hospitality, transportation, entertainment, healthcare, logistics, where a lot of the employees are directly serving customers. They tend to have very high turnover, especially companies like food service and retail. So they're hiring all the time and the speed and quality and efficiency of their hiring is a direct tie to their revenue growth. If they can't hire fast enough, they can't keep up with the demand of their customers, then they lose market share. So this talent acquisition stuff that seems very interesting and arcane to us in HR is actually very important to the company. When you meet tech companies early in their stages of growth, the first maybe five to 10 years, most of HR is recruiting. I mean, that's really what they do. So recruiting is a big complicated hairball process. There's many, many steps starting with defining what the job is, which is also tricky in and of itself, but usually it's just a job requisition, posting the job, defining the job, writing the job description in a way that somebody will apply for it, sourcing the job, looking for people that will apply or posting it on various different networks, buying job advertisements, and then the complicated part of dealing with the floods and floods and floods of candidates. Most of the companies I talk with and we talk to a lot of them tell me they're getting hundreds to thousands of candidates within minutes of posting a job because job seekers have access to AI too. So they're using posting systems to put their resume on 100 companies at the same time. So we've got this flood of candidates coming in and this flood of candidates to source through. And for a low-skilled job or a semi-skilled job, the employer is more likely to just take the first person who comes along. But for a very important, maybe highly skilled, unique, specific job, you got to sort through all that. And if you don't sort through it carefully, you get sued, like the big lawsuit that Workday is going to have to deal with where they're basically being accused of using AI to filter out people that are below a certain age or above a certain age. And we'll see what happens with that. So you've got that. Then, let's suppose you found a set of candidates that you want to interview. You've got to screen them. You've got to give them some form of assessment to figure out how to score them or use AI to score them. Then you have to talk to them or schedule somebody to talk to them or use an AI to talk to them, which is more and more taking place. Then you have to make them an offer, which means you have to understand where they're located, whether they can get to the office on time, what their relative pay should be for the job, and then for their particular situation to get them to join. Then you have to sign the deal. Then you have to get them to start. Then you have to give them an onboarding program. And then you have to give them goals. And then you have to do performance management and on and on and on and on in job training. So this is one very large business process that really is from job idea to hire to performance. Each part of it is done by different groups and different technologies in companies because we designed our companies around tasks and operations, not around workflows. I mean, everybody tries to design around workflows, but you can't find a workflow system that does all this. Even Workday doesn't do all this. SAP doesn't do quite all of this. They do a lot of it. And so you end up buying all sorts of different tools. The applicant tracking system was and really has been the sort of core platform that is designed to do this. But as you see in our research, there's innovation all over the place because every one of those steps can be completely transformed by AI. And I won't go through all of them because it's a fascinating discussion, but creating of a job description can be done by AI, not just to write the words, but to identify what the actual opportunity skills or needs are of the role that the manager may not even be able to explain very well because he's not probably very good at conceiving it in a general form. Then there's the assessment. Companies like Maki People, who we're big fans of, have built an AI assessment that can be tailored to each company that's so effective that it does screening at orders of magnitude, higher quality, and better and faster screening than any other process. And it's not an invisible AI giving you a score of one to five. It's actually a company specific assessment driven by AI. AI interviewing. There's a new tool from Eightfold and others that will literally open a window and show you a character or an avatar who will interview you and talk to you. And it's actually really, really human-like. And you can do a thousand of these at the same time. You don't need a thousand interviewers, so you can deal with those thousands of people that are applying. Then there's the issue of talent intelligence and looking into the job market and seeing what the skills are in this job and whether we're even scoping it correctly because there might be competitors that have actually been looking for things that we didn't realize we needed to look for. And maybe this person actually isn't a good fit, even though they seem like a good fit for the way we scoped the job. That's done by AI. Obviously, interview scheduling if you're doing a regular interview is done by AI. And then on and on and on and on. So where we're going, which I'm going to talk about at the conference, is to an end-to-end series of integrated agents. And since we don't really have an agent-to-agent protocol yet that works, you know, you're going to be looking at tools that either work together in different parts of your TA function or potentially new vendors that are building the end-to-end solution. I know Maki is trying to build the end-to-end solution as his eightfold as are others. So, you know, I don't really envy Workday Oracle SAP trying to do this because they have, you know, this problem about 100 times over in every other part of their stack. But that's what's going on. And the other thing that I think you have to add to this interesting kind of combination of technology issues is that if you look at the United States in particular, and I think this is true in a lot of countries in Europe, the white-collar jobs, the highly educated, highly skilled, high-paying jobs are somewhat diminishing in quantity in the blue-collar front office operations that you would call skilled jobs are going up, nurses, truck drivers, salespeople, operations people, retail people and so forth. And so we need systems that can handle this with much more of a skill based approach as opposed to a credential and expertise-based approach. Because, you know, when you're a McDonald's or a FedEx, you're not looking for somebody who's an expert at working McDonald's. You're not going to find it. But you're looking for somebody who has certain skills and certain learning abilities and certain levels of, you know, they need to be close to the office and they need to be able to get there and things like that. So it's very, very complex. When Workday decided to acquire Paradox a couple weeks ago, I told them that this was not just an enhancement to their recruiting platform, but this was an opportunity for them to get into a new market segment. I don't know if they heard me because in the endless briefing we just got their positioning Paradox as a recruiting technology, which it is. It's very sophisticated at candidate experience, and I think it has been for a long time one of the leaders in chat-oriented or conversational oriented recruiting. Once they're part of Workday, we'll see what happens. Hopefully, Workday leaves them alone and they continue to thrive in that segment and they don't get sort of sucked into the whole Workday infrastructure issues. But they will, you know, they have to do that too. But so you got a lot of options. There is no best system. There are a lot of great ones. I met, you know, so many entrepreneurs in this particular domain. It's one of the most creative and entrepreneurial parts of HR. And you have to think about it from the standpoint of your company, your industry, your jobs. If you're Delta Airlines and you're recruiting pilots and flight attendants, if you're Disney and you're recruiting people for the theme parks and designers and movie professionals and business people, each of those segments is going to require a different talent acquisition process. And there may not be one tool that does every part of all of those processes. There may, but there probably won't. I don't know of one that's quite that versatile yet. And then there's this issue of trusting the AI to make the right decisions. You know, the reason that Workday bought hired score, and one of the reasons I think SAP bought Smart Recruiters was that these products had a lot of intelligence on qualifying candidates for different skills. That's, you know, an emerging and sort of a 10-year-old journey we've been going through. You have to be careful because as the Mobley lawsuit is beginning to play out, you're going to see that your legal folks are going to be very nervous about you relying on a piece of software to decide who to hire. So, you know, this is a very complicated area. The way I think about it in the high level is basically the transformation we're going through is from a handcrafted personal human-centered process to what I call precision talent acquisition. Because what we really want to do to grow our companies most effectively is pinpoint every hire in a unique, optimized way. Sometimes, by the way, there's an internal candidate that's perfect for this job. So, the system or process you use should also look for internal candidates. Sometimes, it should be a contract worker. Sometimes, it should be an expert. Sometimes, it should be somebody out of college. Sometimes, it should be somebody who's had this job before and is certified and is credentialed. Sometimes, it should be somebody who comes from an adjacent role. I mean, you've probably heard me talk about this before, but there are a lot of companies, American Express, AMC, theaters, others, that have learned that they don't necessarily want to hire people that come from their industry. Because in the case of American Express, they want to create a hospitality-like experience in their sales organization and their customer service and their call centers. So, they hire people from hospitality. AMC theaters discovered that the traditional hiring practices for movie theater managers, where they're looking for people with good grades out of school, was also wrong. I've had this same conversation with Liberty Mutual in another area. So, all of these uniqueness fit characteristics that you have in your company have to be considered as you implement these systems. I don't think we're ever going to end up in one perfect system. The guys at OpenAI have put out this sort of grandiose position on creating a job search engine for AI people that is going to give you access to AI skills. I don't really know the folks over there yet, but I think they're a little bit naive. As you all know, AI skills do not mean you're going to be a good fit. Technical skills are only one part of the criteria of a great fit. I think even Mark Zuckerberg, who's paying $100 million for engineers, is probably going to find the same thing. I certainly have seen this in my career. Many times, technical skills are important, and they always are. But learning agility and ambition and fit with the culture and fit with the mission and fit with the team are just as important. You could argue that they're more important, really. As technology changes so fast, we're always learning all the time. I actually think the number one thing that I look for in people, and we find this in our company, is energy, ambition, learning, ability, learning agility, fascination with the work itself, and collaboration with the team. If you have those things, pretty much everybody can learn many, many things. Then, of course, there's jobs in science and engineering where you really do need to have a lot of technical background, too. Read the report. We spent almost a year on this, and we're going to give it away to all of you. You can also go into Galileo and ask Galileo a lot of questions about talent acquisition because it has access to all of the vendors. You can literally do a talent acquisition assessment in Galileo because it builds assessments dynamically. We're also adding a course to Galileo Learn on the revolution in talent acquisition. You can not only read the report, but you can actually take a course on this and have your team take a course on it. It'll really help you sort through this complicated area. That's topic two. Topic three is M&A. I've been involved in M&A a lot. I've worked with probably 30 companies that have been acquired. I've been in a company that have acquired other companies early in my career, and then I've sold two companies myself, so I know what it's like. Because the AI world is so different and so new from the transactional world of the past and the cloud transactional world of the past, a lot of vendors are deciding to buy technologies to get the team acquires to build their AI expertise more quickly. I had a really interesting conversation with Julia Stiglitz of Upliment yesterday. We'll publish that later about how she went from Coursera to Upliment and what it took to build an AI learning company here in the Bay Area. Anyway, you go look around and you see a workday buying companies. You see SAP buying companies. You see Oracle buying companies. I actually think Oracle has a very sophisticated position on this because what Oracle tends to do is buy companies because they want their customers. That's really the model that Cornerstone has used. We as a company have a business. We're in the learning or HCM or recruiting space. We want a bunch more customers, so we acquire a company that has a bunch of customers and then we merge their technology and dark technology and we take care of them and we get the recurring revenue. You end up with a lot of operational problems. You have teams and people that quit. You have old tech and technical debt to deal with. You have integration from the old product to the new product, but Wall Street doesn't see any of that. Wall Street only sees the revenue and the profit. There's one model and I would call this the private equity model that you buy as many companies as you can. You generate the revenue. Learning technology group has done this. There's a whole bunch of companies that have done this in HR. Then you sell the appropriate product to the appropriate customer. You don't try to get everybody to buy everything. You do cross sell as much as possible. This is really the story of UKG and Kronos too, Ultimate and Kronos. Then over time, and I mean a long time, you start to integrate the technologies together and you find out that some of the technologies you purchased aren't as useful in integration as you thought they were going to be and your engineering department gets bigger. You tend to have a lot of engineering expertise on integration between the engineering teams and you have a lot of different engineering teams. It's really kind of a different way to think about a software company because everybody goes through mergers and acquisitions. Even Microsoft spot a lot of companies, of course, and ideally you want to take the team or the engineers and the software that you purchase and integrate it into the one you have. But what if the new stuff is so much different in architecture and capabilities than the old stuff? Should the new guys run the old stuff? Well, that is where we are. If I think about the paradox guys coming into workday, other deals that are going to be coming out, the smart recruiters guys coming into SAP, and many, many others that you see out there. In the case of Oracle, you look at their acquisition of Cerner in the HCM, the health care market, usually what works best. And if you think about peak on how they were acquired by workday, what seems to work best is when the acquisition vendor leaves the acquired vendor alone more or less and lets them run their business and keep their team intact and over time maybe slowly builds integrations between the two products. Sometimes though, what happens is the old product needs to be replaced with the new one. And now you have not only technology issues, but political issues of who's going to be responsible for that and can you get the team to work together. And I think in the world we're in today where everybody's building a startup in AI, and I can't tell you how many there are. I mean, geez, it's just kind of astounding. There's going to be a lot of this. Now, on the side of the company that was acquired, it's also an interesting cultural situation because I've been through this. In the beginning of being acquired, it's very exciting for the company that was acquired because all of a sudden, oh, we're going to get a bunch of options in this new company and we're going to be part of this big organization. We're going to get access to more customers. We're going to be able to hire more people. We're going to get more resources. We get to have a larger sort of stage for our story and our message and so forth. That's all fun, but you lose autonomy. You lose freedom of movement. Now you're operating in the business model constraints and PR and marketing sort of aura of the company you're a part of. What I found at Deloitte was that was very, very frustrating in our case. We managed to do a pretty good job of keeping our brand alive and really helping Deloitte grow the business enormously, but as most of you know in my case, within about 10 years of acquiring my company, they did away with everything. They did away with the business that we were in and the name. We were sort of an injection of innovation and an injection of new customers for them, but not a long-term value. Most of the people left. Some of them stayed, but most of them left. For those of you that are selling your company, I always kind of laugh when people put these linked ends up there about how proud they are to have sold their company. That's always great and it's an endorsement of that you built something, that you did something of value for sure, but it's also a little bit of a capitulation of saying that, well, you reached an endpoint and you reached a point where it was time to be part of somebody bigger. Either people got tired, either we didn't have enough cash, either the market changed or whatever. We're going to see a lot of that. The reason I pointed out to those of you in HR and tech is you kind of need to sniff this out when you're buying software from vendors. When you go to a briefing of one of your vendors and they talk to you about their products and their acquisitions, I think it's helpful to meet some of the people who came from those acquisitions and see how they feel about the new product strategy and how on board they are. In most cases, you're going to get a really positive perspective, but sometimes you're not. If you're purchasing a piece of software that is part of maybe one of these acquired companies within a larger company and you don't get this sense of confidence that there's going to be continuous investment in that particular module, well, maybe that's not the right vendor to do business with. That's the risk. The risk that we take as buyers is buying something that's so great that it got bought by somebody else and then it gets changed and out from under us. That is not something we can change. It's just the way these markets work, but I have sort of become very savvy to this. In our company, we're very careful not to buy things from companies that we think are about to be acquired because we know how disruptive that is for the buyers because all of a sudden, the product changes or there's no new features added or something happens and you can't do anything about it. It's just something to be aware of. This is not a new topic. It's been going on for decades, but I want to make you aware of it because all this AI stuff is so doggone new. Once these A to A protocols get settled, we'll have much more replaceability between agents. If the agent you bought doesn't work as well as you'd like and you want to get a new agent, you'll be able to swap it out and reintegrate it. Then we still don't really know how Microsoft's going to play out. We've been doing a lot of work with Microsoft. I can't give you the specifics to get Galileo connected into Microsoft. It's really amazing what they've got up there. It's really, maybe to me, the most interesting part of the whole stack from the engineering standpoint, but it's also only a small piece because they're not an ERP or a learning company or a skills development company by nature. Anyway, so I'm going to stop there. I hope to see you guys in Vegas or Paris or wherever else you are. If you have a need for help, reach out to us. If you want to think through your talent acquisition tech, download the report, get Galileo, take the course, and then Galileo or we can step you through your situation and help you. We also do a lot of work with AMS, who's a wonderful consulting firm that can help you with talent acquisition and there are many others. There's lots of experts in this space. It's a very big domain, but I think it's important to think about talent acquisition as a business function, not just an HR function. Because that gives you a broader context of how it fits into internal mobility, job redesign, AI transformation, because the jobs you're hiring for are being transformed in real time too. So the more business oriented you are in thinking about this, the better off you're going to be in the long run. I hope everybody has a great weekend and see you See you guys next week in Vegas, bye for now.