How AI is Changing Hiring in 2025
In this episode of the HR Leaders Podcast, we sit down with Jonathan F. Kestenbaum, Managing Director of Technology Strategy at AMS, to explore the biggest trends in talent acquisition and AI’s growing role in shaping the workforce.
Jonathan shares his 2025 talent trends predictions, covering AI-driven hiring, the future of job applications, and how organizations can rethink recruitment in an AI-powered world. He also dives into why large action models (LAMs) will replace traditional workflows, the shift to skills-based hiring, and how AI is transforming the way companies attract and retain talent.
🎓 In this episode, Jonathan discusses:
Why AI is transforming talent acquisition in 2025
Why skills-based hiring is the future of recruitment
How AI is reshaping workforce planning and internal mobility
How Large Action Models (LAMs) will replace traditional HR tech
What HR leaders must do to stay ahead of AI-driven hiring trends
Deel is the all-in-one payroll and HR platform for global teams.
Deel helps companies simplify every aspect of managing a workforce, from onboarding, compliance and performance management, to global payroll, HRIS and immigration support.
Deel works for full-time employees and independent contractors in more than 150 countries, compliantly.
And getting set up takes just a few minutes.
Chris Rainey 0:00
Hey, Jonathan, welcome to the show. How are you? My friend?
Jonathan F. Kestenbaum 0:18
I'm doing well. Thank you for having me
Chris Rainey 0:20
nice. How has it taken us this long to connect? It feels like everyone that I know and everyone that you know knows each other, but we only now 20 years into the internet space, have managed to connect. It's wild. I know it's crazy, yeah, so I think sometimes it like it's meant to be at a certain point, and I feel like it based on our conversations in private, like it's probably now is the perfect time for us. I agree completely to connect before we jump in. Tell everyone a little bit more about you personally and your journey to where we are now, because you have quite an unconventional route, in many ways, to where we are. And then we can jump in to some of the topics
Jonathan F. Kestenbaum 0:58
awesome. Like, broadly, I feel like I've been on a mission to transform the future of work. I started my career as an attorney. I'm still a licensed attorney, but very much that was a means to an end. I've always been an entrepreneur, and both my parents were attorneys, and I felt like if I failed, I needed to go into the family business. I needed a license, and so first business out of law school was brick and mortar tutoring company, which I decided to bring online. We built a reverse auction. So essentially, a student would ask for help and then all of our tutors would bid against each other. The student would ultimately get help at the cheapest price. I thought I was running an ed tech company at the time. But what I learned pretty quickly, as I was running a talent AI technology company, I was getting tutors jobs. We had about 13,000 tutors globally working for us, and I get, I would get calls from CEOs of staffing firms saying, How can I leverage your reverse option system for my contract staffing business, essentially, with these con with these staffing firm executives wanted to do was to get higher margins when they would put contractors on placements, and I ended up selling the company in 2012 got really passionate about investing in talent acquisition technology. I met gene Holtzman, who is the CEO of Mitchell Martin and it healthcare staffing firm in New York. He was a really progressive guy, and Gene was interested as well, in understanding talent technology and the impact he could have on his business. And together, we started an organization called Talent tech labs. At first, talent tech labs was very much meant to be a venture fund to invest in early stage work tech companies. We had fantastic companies that came through the space beamerie, which was at the time, called C dot jobs. I still have their original pitch deck. One day I'll embarrass Ava car with that same thing with phenom people. They were eye momentous. I mean, wow. Their pitch deck was, was even worse than that job. I
Chris Rainey 3:02
need to send you our pitch deck now. I need to get someone who's seen them all. I need
Jonathan F. Kestenbaum 3:07
to send them over the years, yeah, and all kinds of greenhouse came to the office and and, you know, ultimately, in an effort to get more companies to take us seriously as investors, we start to write a lot about trends in talent technology. And as a result, again, I started to get calls from CEOs of staffing firms and heads of talent saying, what does this HR tech stuff mean to me? And I looked at myself in the mirror, and I said, Man, you didn't want to be a lawyer, so you definitely don't want to be a consultant. So what do you do? Now? You know what I learned in the tutoring business was, each time I'd have a student that was needed tutoring, we'd have a different headache that would come up. And so this way, I wanted to have the headache once and and so what we did was we created research reports, put them into a research library, and then sold subscriptions. And we for that business. We raised capital from allegiance group and Mercer, and we did a great job. Scaled that business five years into telling someone they need a CRM or an ATS. And having spent, you know, I dealt with 1000s of talent acquisition technology tools they want to know which one and can you implement it, we started leading the digital transformation for some of these organizations, doing the consulting work. And, you know, long story short, AMS made me an offer. I couldn't refuse to come join them and help them with their tech strategy. We ended up selling talent tech labs, and here I am today. Wow,
Chris Rainey 4:34
wow. So you've seen the full evolution from the very, very beginning, early days to and you were part of that, a pioneer of that, early days, in your first business to where we are now that that's pretty incredible. Not many people have lived it, implemented it, built it and operated on both sides. Yeah,
Jonathan F. Kestenbaum 4:55
what I've tried to do is actually drive the market forward, really like i. Think many of the categories that people use for budgets around technology I created with my team. I mean, some of the names of these categories, social search, you know, job distribution. I mean, these are things that, you know, candidate relationship management, that we came up with.
Chris Rainey 5:14
Wow. So I feel like, what's really going to be interested for our audience then, which is why I was excited on the show. What are you seeing now is the biggest talent acquisition trends in 2025
Jonathan F. Kestenbaum 5:26
so I think I should first reflect on my 2024 predictions.
Chris Rainey 5:30
Good. You did.
Jonathan F. Kestenbaum 5:34
Yeah. So in 2024 I had three predictions. The first one is that, you know, the elf in the room around AI, was that candidates would start to use it. And I very much think that that one came true, you know, I think the big question still looms as to, Is it cheating? You know, everyone has a different view on that. I think if these candidates are going to have access to AI when they get into your company, then they should have access to AI before they get into your company. And what we need to do is find new ways to assess candidates so that, you know, because basically, AI is becoming the great equalizer of knowledge work. And so we need to start assessing candidates for their agility, for their, you know, ability to learn new things. You know, adaptability, resilience, right, exactly, exactly and so, and train them in, in being stronger at those things. So, you know, to me, it's not cheating, but definitely, I mean, there are, there softwares for candidates that help them apply to 1000 jobs at one time in three minutes. There, you know, there's applications that help them, you know, quote, unquote, cheat on their assessment, that you know, that you know, you asked me a question, I could look at GPT real time the answer to it, I mean, but the truth is, these, they're gonna have access to this technology and on the inside. So, so that was one, I probably think it came true. The second one was the rise of open AI systems. We are definitely seeing. You know, this happen for a while. You had these kind of siloed, basically large scale platforms that didn't want to integrate. Look at work day. They've now obviously opened up their system, and it was a challenge to integrate through API into some of these larger platforms. You now have third parties like merge, dot dev stack one, there's a number of them that integrate into all the HR platforms. And so you as a provider that wants to plug into them can plug into one AI and get access. And I think that's going to evolve into 2025 and I'll talk a bit about that, but broadly into 2025 that I don't think we're going to be using interacting with UI as much in the future. We're going to be interacting with, you know, an AI agent, and as a result, you know, data is going to meet us in the flow of work, inside teams, inside Slack, you know. And so I do think this goes a step further right. So we have now the systems are connected data flows, well now you know the information actions are going to send the full work. And then the third was around blockchain. And I know blockchain, oh, everyone's like, Oh, Blockchain, here we are again. And but the idea really was about, I believe, as AI, obfuscates our identity. You know, I've created a deep fake for myself. This is the real me.
Chris Rainey 8:26
It's not you.
Jonathan F. Kestenbaum 8:29
I did create a deep fake of myself. And, you know, I think we're going to need blockchain to make a comeback. What didn't happen in 2025 as I predict, predicted in 2034 I think it will in the future, to prove who's a real person.
Chris Rainey 8:45
Is that a way of doing that? Is that? Because I was talking about that recently, and I didn't know the answer, like, how do we what technology do we use to understand whether this is generated by AI versus being real? Is that what blockchain is a use case for?
Jonathan F. Kestenbaum 8:58
Yeah, I believe blockchain's biggest use case is identity, not like all this cryptocurrency stuff,
Chris Rainey 9:04
yeah, yeah. Because that's, by the way, that's a good distinction. That's why, that's why I mentioned that. Because it's when everyone thinks blockchain. The first thing they think is crypto. I literally that's like, the first thing that people think of, okay, that's really interesting. And
Jonathan F. Kestenbaum 9:16
so those were, like, 2024 trends. When we think about 2025 I have three big ones that I believe are gonna come true. The first one is, everyone's been talking about llms, large language models. I think we're moving to large action models. So this is basically agents doing the work that humans used to do. And I would argue that, and this is the big challenge that organizations are struggling with, AI, it's not really replacing full jobs yet. It's replacing tasks. And as a result, in order to when we think about. AI. We can't think about it as a tool. We have to think about it as a strategy. And we have to basically redefine the roles within our organizations, within our teams. And if you think about this in the context of talent acquisition teams like that, the tasks that a recruiter, a source or the head of talent are going to do are going to change wildly, and AI is going to continue to eat at pieces of that. And what we're going to essentially have are many agents that do the work of that, some of the work that we used to do. So I think we're definitely going to see the rise of large action models, which essentially is agents. That's the first trend. The second is evolution of programmatic advertising, which I think is really interesting. So you said, Hey, we've been at this for a while. It's been fun. I've watched, you know, job ads be posted in newspapers, to job ads be posted on job boards, to
Chris Rainey 10:56
LinkedIn, to
Jonathan F. Kestenbaum 10:57
LinkedIn, right to now we have, then we had, uh, you know, I wanted to post a job from my app, book, contracting system onto these job boards, so I didn't have to go to 17 different job boards. So we had job distribution systems where I could push a button that would distribute to all those those websites. Then we moved to uh, programmatic advertising, which essentially was okay. It's not only that I could distribute to them, but I could turn on and off the ad based on certain criteria. Maybe it was how many applicants I received on an application, how much I spent. But we were never able to turn on and off the ad based on quality. And now we can, and I'll explain how. And this is what I believe is the next evolution of programmatic. Essentially, when someone applies, one, we can use llms to match the candidate to the role based on their resume. But two, we can actually run a quick assessment using Gen AI to assess whether they're a good fit, by asking them questions that fill the gap between their job description and the resume. And ultimately, where that leads us is is we can actually say, okay, indeed, sent us 15 quality applicants that score above this score. So we're gonna leave indeed running in zip recruiter. And I don't mean to call it zip recruiter anyway, I don't get paid by any of these guys, so, but zip recruiter didn't send us any quality candidates. So we're gonna turn off that ad and stop spending money, right? So, you know, I think you're gonna start to see that happen, and that's exciting. And the third trend for 2025 is the shift from software as a service models to service as a software models. And we're really going to see outcome based pricing, and I'll explain what I mean by this. For the last number of years, you had, you know, kind of these three siloed buckets of spend in recruitment. On the on the far left, you had about $14 billion spent in recruitment, marketing. In the middle, you had about $2 billion annually spent on SAS software. And on the far right, you had, you know, north of 400 billion spent on recruitment services, you know, RPOs and staffing and and like there was, there was a real kind of break between those different budgets, and you couldn't really cross that line, but AI has unlocked and broken what I call the blood brain barrier between those budget categories. And if you think about it, what historically was like a workflow SaaS tool. Let's look at seek out, for example, which is a social sourcing tool, right? They went out scraped the web, they built these profiles of candidates. Historically, they would sell that as SaaS, but actually now they can start to use AI agents to do what a sourcer would do, and so now they can kind of start to get access to some of the staffing services spend. And actually, in some cases, they could sell it as a service, which the services spend in both of those those silos on the far left and far right were historically easier to access than assessment. You know, assessment was annually budgeted, annually spent, like in those two buckets, like, you know, individual hiring managers, teams have access. It's much more malleable. So, you know, anyway, I really do believe that we're going to start to see technology vendors with outcome based pricing. This is obviously something as an organization, AMS is really thinking about is, how do we align our our services more meaningfully to the outcomes that these organizations that we work for are looking for, especially because our business used to be about butts in seats, like, how many recruiters and sourcers? Yeah, and like, you know, we get a margin on that, and now it's actually, well, maybe it's going to be a recruiter, maybe it's going to be an AI agent that delivers the work. Maybe we're going to train somebody and place them in the role, and based on that, you know, the best way we could provide enough, you know, a solution to the company is to charge them based on outcomes. Wow,
Chris Rainey 14:56
there's a lot to take in. Yeah. There is so fascinating. Let's start from the beginning. So you're talking about large action models. And by the way, I smile because I use that terminology a lot, and then I have to explain that to people because it's relatively new. And you saw there also, there was some companies that claimed that they did this, and it kind of blew up like rabbit OS, yeah. Was it rabbit? OS, is that the was rabbit? Though, there was like, A, yes, the like, the like, a little gadget, like, you basically could talk to, and it could then go book you a flight and do your hotels and but that was the whole big it blew up, right? And the whole pitch and premise was, and that's the first time I ever heard it in that keynote was around large action models as well, but now we're starting to see this come to life. You know, even within your iPhone, if you've got an iPhone right now, you have Apple intelligence, and you can start to start to start to use that to start to take actions, whether it's booking, you know, a hotel, or ordering a meal, or just or just training it on a specific task that typically you're doing manually that you can train it on one time, and it just repeats that. So now you can just remove that from, you know, the equation, and I'm already doing that of a couple of tools here internally. For example, in our video editing and production, there's some really amazing use cases to be able to just take, ingest the content, take an action, go and post it, then, like on it, and it kind of covers your end to end product. How far do you think we are away from this being and obviously I'm seeing it in consumer but I haven't seen it that well established in businesses yet as well. How far you think we are away from really starting to execute on that? Yeah,
Jonathan F. Kestenbaum 16:38
so the big challenge, there's two big challenges. One is, I think most organizations are in, are in the testing phase, like, you know, almost like launching little AI accelerators to test use cases, and not like, at a place where they're building business cases, like, like, you know, and going to their CFO and saying, This is the cost savings you're going to see. And I think that's a result of the technology is moving at such a rapid pace that, you know, it's people are too scared to, like, make a bet on something. Also, it's like, you know, there's kind of two sides to the story, like everyone's talking about, AI is going to make us be better, you know, humans we're going to be able to engage, and computers are going to be able to compute. And you know, I was at a conference. I was at the LinkedIn talent conference. It was, it was, there was this awesome panel with the head of talent, Amber girl of BCG, and one of the experts at LinkedIn. And they were talking about just that humans are gonna engage. This is such an awesome time. And then some guy in the audience raises his hand and says, This is all great. And that sounds awesome, but let me tell you how this is putting out in my organization. I bought chat GPT, and my CFO didn't say, Okay, how are you going to now train all those people into being better at engaging, or, you know, doing this not my CFO is saying, Where are you going to cut the cost so I can get the EBITDA improvements from the technology? And so there's, there's that dynamic of trying to figure out, you know what, what's the impact? What is the business case like, where are we going to cut people? Are we going to retrain them and re skill them to do more engaging work? And then the other side of this is the legal and ethical side. So I think you know, if you look at AMS client base, like only probably 50% of our clients have policies that allow them to even use AI. Of the 50% you know, a certain portion of those are are looking at AI, and that's because the, you know, legal system globally is kind of evolving to catch up and make sure that we could implement AI effectively and ethically. Yeah.
Chris Rainey 18:48
What are some of the specific use cases from a talent acquisition point of view, when you think about large action models, what are some of the tasks that you foresee that this could help replace or enhance? Let's say so
Jonathan F. Kestenbaum 19:01
I'll give you some of the use cases that AMS is exploring. One would be, I mean, there, and I'm going to start from the most basic. So, you know, I'm not trying to wow you with the first one. The most basic would be like Job job description creation. You know, historically, people would go on Indeed, and copy a job description and paste it, and that would become their job description. So job description creation, you know, we're using it for matching candidates, you know, parsing information. We're using it for engaging with candidates, so we're able to reach out to candidates. The messaging is crafted for us. We're able to write emails to hiring managers, selling candidates, pitching them. We're looking at image based general AI generation as well for job advertisements. So creating image. And advertisements, both for job boards and just broader social networks. We're looking at voice based AI agents. I'm most excited by that because it's just me
Chris Rainey 20:11
too. By the way, we're literally a few weeks away from launching our AI, our voice capability in Atlas, and I'm literally so excited for people to be able to just talk to Atlas and have a conversation, ask questions, get answers in any language on any device, it's just a game changer.
Jonathan F. Kestenbaum 20:30
And by the way, this I engage with with chat GPT.
Chris Rainey 20:33
By the way, I don't know in the chat, wherever you are right now, I'd love to know in the chat are using chat GBT voice, because I realized there's not that many people I personally speak to that are but so you're one of the few I always use it. I actually had a full podcast conversation with chat GBC the other day. I was like, you are ahead of HR. I want to understand these trends. Let's have a podcast. You're going to ask me questions. And I had a whole 30 minute conversation with as well.
Jonathan F. Kestenbaum 20:59
Yeah, there are awesome companies in the space. I've recently demoed class it. They so basically you put it, you know, we can put a job ad up, and then you can actually interview, right there, 24/7 and, you know, with AI agent, you know. So, you know, it really exciting opportunities ahead. I think you know, where we use these technologies is still up in the air. You know, is that a high volume solution? I don't think executives are going to want to interview with an AI agent
Chris Rainey 21:30
today. No, no, no. I don't think it's there yet. And our head of engineer, Marvin, that's how we first connected. He built one of the first AI avatar agents that you could do interviews with and he just sent me a link on LinkedIn. I didn't know him, and I went in and I had this interview, and I was like, You are a head of this that we're going to do an interview around x. And I just sat there doing it. I was like, I have to call this guy back. I was like, this is, this is. And it's like, two years ago where the voice didn't match the lips and the avatar wasn't that great, but now it's got to the point where it's almost indistinguishable between is it actually a human or not? So that's going to be interesting. Yeah, I think at the executive level, not so much, but I think for high volume hiring, I think that there's a place there for high volume hiring. I
Jonathan F. Kestenbaum 22:13
missed one. Also, we're transcribing and summarizing interviews, which is helping us gather a bunch of information. And if you look at companies like bright hire, you know they're you're able to not only transcribe and summarize interviews, but you can get insights. You could assess the candidates against the criteria based on what they say, but then more importantly, you could coach the recruiters on how to be better interviewers based on what you know these transcripts are showing. So I mean, there's so much to be excited about. Yeah,
Chris Rainey 22:48
and also in that interview process, if you have a multiple stage process, that next person who's going to be interviewing has all the context
Jonathan F. Kestenbaum 22:54
they know what's missing from what wasn't asked. It's nothing worse
Chris Rainey 22:58
when you're a candidate and you get asked the same questions by every person in the stage of the interview process. So you can start to see, okay, this is what we got from the first few this is some of the gaps maybe we want to explore as well. And then you can align all of that with your culture, your values, the competencies, and start to, yeah, that's and that saves a lot of admin, a ton of admin. I think one of the things I want to add to what you said, though, is deep everything you just mentioned. One of the things I'm excited about is that all of the stages of all of those use cases, you can customize it to the individual. So it's not that just you're creating job ads or you're doing the communication pieces is tailored to that specific audience. In the past, we had approach of just blasting out one one message, which wasn't tailored to Chris and to me and what matters to myself. And I think that is going to be one of the most exciting opportunities that we have to be able to do that at scale 100
Jonathan F. Kestenbaum 23:54
so I always joke I live like, two years in the future, right? Because I'm looking at the technologies that will make an impact, you know, two years later. But, by the way, the future is awesome. I'm super excited about upcoming but, but, you know, one of the, you know, look at recruitment. I mean, the issue for candidates historically has been access, access to opportunity. You know, access, in that case, is like one. It's sometimes about who you know, but it's also about having the time to look for a job. It's about being able to find the jobs on the job boards you know that you need access to. Like access will be solved by AI for candidates. Like candidates will be able to get matched to the right opportunity, even if they don't use the right search term in the job board to find it and at the same time. So the experience will be way better for candidates. At the same time, companies will start to implement new assessment platforms. You know, a lot of this wasted kind of brute force stuff is going to go away. I think unfortunately, what that means is we'll need less people potentially. Essentially, to deliver this, you know, the combination of human labor and an AI, you know, Agent labor will get rebalance. You know, this is what happened in the Industrial Revolution. The Industrial Revolution was the great equalizer for physical labor. Now we're, we're in the the intelligence revolution. This is the great equalizer for knowledge labor. You know, in the short term with the short term, with the industrial revolution, we lost some jobs because, because, again, it was a great reshuffle of of the tasks that humans had to do, where the same thing is about to happen now, but ultimately, I believe AI is going to create more opportunity. It's just a matter of how quickly and can we all adapt and adjust to be able to, you know, take advantage of that. Yeah,
Chris Rainey 25:41
one of the challenges that keeps coming up in my podcast, and I'm going to throw it to you, is a lot of serious are telling me, with the evolution of the ATS is and the programmatic advertising, because it's becoming so easy for candidates to apply, they're getting an issue where they're just getting one person apply to like, 20 jobs at the click of a button, and it's causing them to have an insane amount of applications. So it's like, yes, we made a candidate experience really easy and sort of one touch and super simple. In some cases, don't even even need to add an email in there right to eat to get this going. Where how do you how do you feel that we solve that, because that's becoming now they're overwhelmed with application. Yeah, so easy.
Jonathan F. Kestenbaum 26:25
So listen to this idea. So first of all, we're seeing that too, right? We're seeing especially in the early career space, because at AMS, we have early career RPO, and we're seeing tons of applications come in. Because, you know, early career candidates, I think, are just more schooled in, you know, leveraging these tools that they read about on Reddit and Twitter and, you know, or x, excuse me, so I recently got pitched a company called match two. Dove berg is the CEO. Their vision is, candidates come to companies, websites, career sites, you can actually apply to a job so that there's you can see the jobs, but there's no apply buttons. So what happens is you have to put in your resume and basically build a profile of yourself with their little wizard, and then they'll match jobs to you based on the information you put in. So, no, that solves two problems. One, the AI can't just apply to jobs. Two, I, if I didn't like use the right search term to search for an opportunity as a candidate. I don't have to worry about that, because the LLM is now going to match me based on doesn't matter what you know, language I used in my profile. They're going to match me to the opportunities that are relevant for me to apply to. If it requires me to take an assessment, they'll take it. So they're flipping it, like flipping the whole, I think they're onto something. And the best part about their tech is that it you don't need to, like, it's not some big ATS CRM platform. It sits right on top of those ATS CRM platforms. So,
Chris Rainey 28:02
so sitting at the front end, so they're not trying to build their own ATS, they're basically saying, hey, on the front end, we're going to use this to filter everything and then filter that back into your ATS. And I'll
Jonathan F. Kestenbaum 28:15
take it a step further, if I'm a candidate, and I you know when to match choose widget on Deloitte website because they're a client, let's say, and then I go to match to maybe, let's say Home Depot is also a match to client. I go to Home Depot as a candidate. I don't actually need they'll know who I am, because there's like, this super cookie that follows me around. So,
Chris Rainey 28:35
oh my God, that's a game changer. It also makes the
Jonathan F. Kestenbaum 28:39
candidate experience better. I don't, you know. I mean, just within one company, sometimes you have to go through a login experience for an ATS and a CRM and an HCM, like, you have to keep it's a disaster. Now. Now think about across companies, you'll be able to do it. So, I mean, would
Chris Rainey 28:56
companies be okay with that? So, like, I for the candidate, I'm like, wow. Like, the ability that, if they can get this on the front of all of the major ATS platforms and and have the agreement that they can share that data so the candidate can then almost be, hey, we've got this job, this job, this job suggested from different areas. What would the companies themselves not be like, Hey, you're, you know, you're taking away my you're giving my candidate, candidates, our competitor, and not us. How'd you overcome that objection? Yes,
Jonathan F. Kestenbaum 29:27
I don't. I don't want to speak for them, because I don't know all the you know, I'm sorry but, but my understanding is, can clarify. My understanding is, they're not, they're not like pro marketing candidates from one company to another in any way. They're not crossing that line and saying, Oh, you didn't match for a drive. Here you go. Here, nothing like that. It's just, if they have a profile, a match to profile for. Can you just like, I can log into a website with Google Gmail, and they know who I am. It's the same kind of idea.
Chris Rainey 29:57
Well, I get what you're saying. Now set up. If they're going to go and apply for another job on another site that and that already uses that same platform, you already recognize they have the cookie stored, and they can just log in and it knows the context of who they are. I misunderstood that so that make that again, just improving that experience, making it seamless as well. I'd be interested to understand the candidate experience on that, because they may feel like, not helpless, because they just sort of now in left in waiting, at least when you apply, you you kind of like, I've taken an action to apply, as opposed to now you've reversed that on me to be like, now I have to wait to see if there's any, you know, anything that matches me. I agree
Jonathan F. Kestenbaum 30:37
with that. It could be frustrating, especially if I think, I mean, that's part of the problem, like, you know, everyone who applies to a sports team thinks, you know, to work at the sports team or sports league, thinks they're the most qualified person ever, and that's what I mean. And I don't understand why there's no opportunities for me, but it's probably better than burning a bridge with them and, you know, not getting back to them after they apply. You know, making that they don't want to, you know, this was a big challenge at Disney. I was doing for work for them years ago. Is like, you know, everyone who wanted to apply to a job was a client that they didn't want to lose. Yeah,
Chris Rainey 31:09
could you delve a little bit more into the third trend? Because I think that was one that threw me off, and I was writing notes about it in terms of, you know, just dive a little bit more. What you mean by software as a service, to serve, to service as a software. Could you and talk about maybe some of the implications for companies that don't do this and some and what are the positives of people that do? I really just want to understand a little bit more about that. Yeah,
Jonathan F. Kestenbaum 31:34
so let me I'm going to tell you a thesis that I've used like to drive my investing over the last number of years so, and this is basically just a I'm going to try to reframe what I said in a different way, to kind of give it a different angle. This whole idea that technology companies should be valued at a 10x multiple, and services businesses should be valued at a 4x multiple. Or I'm just making up numbers for the sake of purposes here, right, to me, is is going to go away, okay? Because I believe that in the future, and not the distant future, like the very near future, every company is going to be a technology company, right? We're all going to be leveraging technology in some capacity. And in that regard, I think that typically, it's easier to sell, serve a service into a large organization than it is to sell technology like technology is usually sold like annual budgets, like, you know, you have to, you know, get there before they budget. It's a large expense. It's not as transactional services are more transactional but, but but I think that because what historically was technology which was like a workflow engine, you know, where, you know, think about like a CRM or an ATS, these are just workflow engines. You're now actually going to be able to add a service layer on top of those technologies, you know. And I think that they're going to be able to start to position them. So let's again, look at seek out for an example, like seek out theoretically could sell sourcing as a service today, instead of a SaaS subscription, if they wanted to, because and it still would be the technology that they go for powering it, right? I should get paid by them for this idea, but I'm sure they've thought of it. I
Chris Rainey 33:22
think every SaaS company listening right now is like, Wait a
Jonathan F. Kestenbaum 33:28
minute. So I think that essentially, when you move to service as a software, the market gets so much bigger your target, you know, market gets so much bigger. Yes, in some cases it might look more transactional than recurring, but actually, there's ways to overcome that. But I and then this, there's kind of two things I mentioned and I kind of completed that I'm sorry about that. There's, like, this move from software as a service to service as a software one, and then two, I believe we're going to start to see some outcome based pricing, instead of just, like, buy my software and it's going to work. You know, I can't tell you how many times I saw because I was always the guy who was called in by these large enterprise companies. You know, X company told me they're gonna save me 40% on my sourcing. It's not working. Or I implemented this CRM, and my recruiters are working 80 RECs now. Like, and I'm like, Well, hold up, hold your horses, dude. Like, you can't just, like, put a system in and not change your people and process around the technology like you gotta. I believe that you're gonna start to see these, these systems tie themselves the outcome that the organization is trying to achieve.
Chris Rainey 34:33
Yeah, yeah. You just made me completely rethink how we're approaching a few things right now, and I think it's super, super I think especially in this day and age where everyone is now trying to sell their AI SaaS solution to a company, and you all get put in this big pop this bucket over here, right? And we're having the same challenge with Atlas right now. We're speaking to CHROs in the. Teams. They love what we're doing. They love the value. And you get put over here and now, and every single part of companies has been off is being inundated with these solutions, right? Whereas, more recently, and this is so crazy, we're having this conversation, we shifted to literally a service as a software approach, and all of a sudden it's now become transactional deals, and nothing changed. It was just a way that we positioned the product
Jonathan F. Kestenbaum 35:28
like I can connect to it more me, if I'm paying for example, researchers on my team, which I we have like that are out in the market analyzing all the talent technology solutions so that we could stay on top of the trends and features and functionality. And someone came to me and said, like, two different pitches. One sales pitch is, hey, here's a software that can help you track those conversations and build profiles of those things. And then another person came to me and said, you don't need four researchers anymore. You need one because we're going to be your digital researcher, I'd be like, oh, yeah, I have a budget for research. I know how much I pay
Chris Rainey 36:07
them, and you can fit it in there, right? You're not thinking, Where does that fit? And I don't
Jonathan F. Kestenbaum 36:10
have, yeah, I don't have to, like, decide, like, is this going to save me 30% do I not need to just, like, it fits into a spend category I have, and I've seen this before. Like, budget categories is a big thing. When CRMs came out, okay, I was like, this is genius. Candidate relationship management makes so much sense. This is a specialized form of marketing, you know, just like marketing, we need to engage candidates. Earlier on, I brought some of the early CRM clinch was an example to some of these heads of talent that I was working with. I was like, how awesome is this thing? Look at this. It's, you know, look at what you could do. You can engage candidates. And they're like, this is great, but I don't have a CRM budget. And I'd be like, Well, what do you mean you don't have a CRM budget? They're like, well, maybe next year we can budget for CRMs. And I was like, Well, so, like, we got you hit a wall. So then I was like, Okay, well, do you have a career site building budget? Yes, we have that. Okay, we'll build your career site, and then we'll upsell you all the CRM functionality after. Now, everyone has a CRM budget, but, you know, that's this is another area that you kind of see this happen. And a lot of these tech companies, I think, interestingly, they all want to be one of one, right? They all want to be like, I'm the only, you know, they create a new instead of CRM work TRM, right? But guess what? Dude, like, they don't have a TRM budget. They have a CRM budget. Like, like, you gotta fit into the budget. I
Chris Rainey 37:27
feel like you're doing a master class right now for everyone listening. So if you're a company right now, solutions in this space, these are some of the biggest hurdles that you're gonna face. And and you're right. Like, the difference between I already have a budget for this, and so I can position you here, versus you're coming in with a category doesn't even exist, especially in this AI space, there's categories now that being created that didn't even exist before, let alone have a budget allocation for that. But by, but right by, but by repositioning it as a service in a bucket that already exists, versus trying to go into a company and create a brand new segment. With a budget that doesn't exist, you're just going to hit a wall. You will get some deals. But the sales sell the lead that the sales the time to it's going to take a long time that sales, sales process is going to be really long.
Jonathan F. Kestenbaum 38:18
It cost people, I would, I think, almost 10s of millions to create this talent intelligence category they created. I mean, that was totally created by them. I don't you know, it was hard. It's hard
Chris Rainey 38:31
before I let you go. We covered a lot, and I feel like we could do a whole series on some of this stuff. What would be your advice for everyone listening? Because we there's a lot we just threw at everyone, right? What would be your key takeaways in terms of what some steps maybe, or things that they should be walking away from this conversation to think about, because otherwise they're going to walk away with way too many thoughts and not take any action. If there's one action, let's say that if there's one action, you want our audience to take away and do, what would that be? So
Jonathan F. Kestenbaum 38:59
if we're talking about talent leaders, folks within organizations that are leveraging this technology, which I think is the best place to go, if I've learned one lesson over the last 15 years having evaluated these 1000s of technology solutions, it's that the tech is only as good as the people in process, you build around it. And that stands true even now. And I mentioned briefly this idea that AI is eating at tasks, not jobs today. You know, I very much believe if there's one thing you can do, it's, it's with any of this technology, it it's, don't just focus on the shiny object that they share with you, because it's, I've actually never heard about sales pitch. They always sound good. I always leave the sales pitch saying, Wow, that's cool. But in the end, you have to have a business problem. You have to have take, you know, you have to figure out if that technology can solve that business problem and optimize your people and process around it, so that that is like the biggest lesson I can share. I love that,
Chris Rainey 39:59
and. Can people reach you if they want to reach out, say hi. Connect, where's the best place? LinkedIn,
Jonathan F. Kestenbaum 40:04
you know, I'm constantly I try to get my thoughts and ideas. I try to put my money where my mouth is, you know, at AMS. You know, really, AMS is really leading the pack. I believe in changing what's possible in recruitment. I mean, you know, if you think about it, we do over 250,000 hires a year for our clients and power their town acquisition functions. And you know, we're constantly thinking about how to make the process better. So amazing. I would love to talk to all of you.
Chris Rainey 40:30
Yeah, I know man sounds like they hired the right man for the job. For anyone listening right now or watching wherever you are, whatever platform you want, there'll be a link to connect with. Jonathan, advise LinkedIn link in the chat, so make sure or in the comment section. So make sure you do that now. Apart from that, it was really good to connect with you. Man, I'm so happy we crossed paths, and I look forward to continuing to get to know you more, and I wish you all the best until we next week. Same
Jonathan F. Kestenbaum 40:55
here. Thank you for having me. Thank you. Thank you.
More from the HR Leaders Podcast
Jonathan Kestenbaum, Managing Director of Technology Strategy at AMS.