Michael Beygleman Live on Shally's Alley

Hey, hey, hey, what's up people? Welcome to Shally's Alley. You are live in the alley. It's Friday, April 18th, roughly 1 PM US Eastern time. And, uh, I didn't realize this morning when I put my t-shirt on, uh, I guess I just forgot that it was Friday, and I just realized that I'm wearing a pretty, uh, fun t-shirt that I probably shouldn't have worn on the show. Um, so it's my Say No to Pot t-shirt, and it's a lobster saying no to pot. You get it? My, uh, my wife's family lives in Maine, or they're from—

it's Good Friday today too.

So I just— yeah, well, he's, uh, missed the mark on a couple of notes there. But thanks for joining us today. Pretty excited, actually. Both Michael and I have been looking forward to this. For a while now. With me today is Michael Bagelman. Bagelman is actually the man with the bagels. So he does like bagels, and at some point his ancestors somewhere made bagels for a living. You know, that's just kind of how names work. So legitimately, the Bagelman. Not very many Bagelman in—

Bagelman.

But so, you know, Michael's background, just as an introduction. He currently heads up product development for a company called Wilson HR. And he got that job because he was the founder of Claro Analytics in 2014, if I'm not mistaken. And so, he's an innovative product leader, visionary entrepreneur, serial entrepreneur, somebody that I ask questions about business and industry and things like that. And we really We agree in some fun ways and we disagree in some creative ways that are, that are both good for both of us. So I thought it would be a fun experience to kind of talk about some current events today, illuminating a little bit more of the perspectives on what's going on. And if we have some time, get into some of the wonky side of, of analytics when it comes to like behavioral, you know, applicant behavior and and stuff like that, which, um, which is his expertise. So welcome to the show, Michael.

Yeah, thank you. We actually did a trial run of this at a lobby of a Marriott in Atlanta, I believe. Yeah, because I was sitting down because I was exhausted, and you were sitting down because you had a bum foot. Yep. And we were talking, and I started looking around, and people were standing around and listening.

Yeah, like live. You know, that makes me think of somebody. Actually, my daughter was there and she said, why don't you have a, um, but just like your, use your phone and record that. Um, she, so she gave me the idea and I, and I realized, you know, that's a good idea next time because she saw how many people were coming up to us.

Right.

And, and she's known for a while that I'm, you know, well-known or whatever, but she was like, wow, that's a lot of people. So she's like, why don't you just record those? So that gives me a really good idea. I'm going to buy one of those, like, um, they're the wireless microphones. Yeah.

My daughter's got one. She does a little bit of recording and streaming. They're really good.

Yeah. So I can, I can give the other person one.

Yeah. They're like little lapel things with a little fuzzy thing. You just stick it on and yeah, off you go. And it's like pretty high quality actually.

That's right.

Yeah. I was looking around. I'm like, okay, should we be like thinking about what we're talking about?

Because Yeah, with the phone on record, I think it would make us a little more careful of what we're saying.

Because this is what— that was like the rehearsal. And then we started talking about, hey, let's do this show together. I'm like, well, the dry run was pretty good.

So it worked.

Kind of like, kind of work. I'm sure we bantered about 80,000 different topics.

We did. One of them stuck with me quite a bit, actually. I was gonna tell you about that.

Yeah, tell me. Yeah, do the segue. Yeah.

So one of the things that stuck with me, remember when you were talking about entrepreneur in residence?

IR?

Yeah.

EIR.

And I actually started, uh, I didn't even know that thing existed. I mean, it was like, I didn't even know that was a job.

I'm literally the king of useless information. So for any—

well, that wasn't useless for me because I started looking into it and there are a handful of, um, investors in our, in our space that do actually have that, where that role exists.

So it's almost like I know what I'm talking about, shall we?

Right, almost. I had to double-check, but yeah, it's almost like I've done this before, right? And I actually, um, you know, I'm gonna reach out and talk to some of them about, uh, you know, about the possibility. It might be something that I can do as a consultant because it is basically a consulting job. So an entrepreneur in residence, what I've learned— I know you know this, but folks listening might not. Entrepreneur in residence is somebody that has deep expertise in an industry sector and works on behalf of the venture capitalists, the investors, as the connection between their money and the business owner, where, you know, the business owner of course knows everything about the product and the market, and the VC knows everything about business, but they may not know the product. And so it's kind of like the person in the middle that helps translate. And, um, it's almost like a product role, right? I would say it's a little bit of a product role.

It's a product-market fit role. That's really what it comes down to. And, uh, when we were talking about it, I'm like, there's really no better person in the market that understands what's happening. I'm like, this would be like really ideal for you because, yeah, investors, like, they realize that like you can't read marketing brochures, right? Because, you know, everybody has good marketing brochures. So help them to figure out product-market fit, um, and more like of a thesis fit is like super, super valuable, right? Yeah, because, uh, that— and it's like interesting work and it's really needed because weirdly, weirdly, there's a lot of capital out there. You know, it seems weird because like if you listen to news, you know, you think the world is ending. The reality is there's just so much capital out there, it's just not being Boy, I think I, I don't remember where. By the way, um, you can be my bullshit filter. Um, I think I read somewhere that we're like at a 90-year high for wealthy people, um, uh, 90-year low for wealthy people spending money, meaning like the people that have the most amount of money are at a 90-year low of spending it.

Got it.

Um, it was just like some crazy stat. I remember where I saw this. Maybe I'll find it later and I'll share it with you. You can share it with your audience. But, um, and it actually validated what I've thought because I see there's just, there's so much capital on the sidelines. It's not being deployed right. And it's not because there's no capital. It's not because the last lack of desire. It's just, uh, I think the industry is moving too fast.

It is moving very fast. That's one of the things I wanted to do.

So I think an EIR role is actually really, really cool in that regard because you can help sort of bridge the gap. So I'm glad that you're really good. Yeah, it was Double-click on it.

That's right. I learned that too, the double-click concept. But, you know, they just in the last week, maybe it's because I just started paying attention again or not, but like, I feel like there was at least 5 or 6 new products that just popped into the market out of nowhere. And when I went to their websites, almost all of them to a tee made some sort of claim of having been around for a long time. I, I mean, either I completely missed them or like maybe they were called something else. I don't know how that happened.

People incubate ideas as an entrepreneur. It's like really interesting. It's such a double-edged sword, shall we say. So it's not like you're not going to go out to dinner and come home and go, oh, I have a brilliant idea for a startup, you know. Yes, this happens, but that happens like one in a trillion, right?

Right. More often than not, you've been evolving and evolving and evolving a while, right?

You incubate stuff, you tinker around with stuff. And let's say you get like some friends and family money, but like you have to set up a company. So what happens is a lot of times somebody will set up like a corporation, they'll get, you know, $50K, $100K in seed money to tinker with stuff, and they may be tinkering and stuff for a while, and then they, you know, run across the right person who helps them move it from tinkering to something more interesting. So it's not uncommon for some of these things to form like a while ago And then you kind of create this a little bit of a, do we want to tell people that we're a rocket ship and we just came up with this idea last week? Or do we want to tell people we've been around for a while?

We've been around for a while, even though it was a different— yeah, that makes sense. I get that.

Yeah, it just, it happens like that, like that sometimes.

But I think that the other thing that I've seen, and correct me if I'm wrong here, it's between HR Tech which was the previous time that we met, and then when we recently met, and now I've continued to evolve this idea that there's almost like a generic language that they're all using that is not— it doesn't identify what they do very well. It's almost like they're all—

I've heard phrases that I have no idea what they said, and I'm like relatively smart.

Yeah. So I'm right about that. I mean, it's not just me.

1 billion%. The thing of it is, it's like, it's really interesting. The— there's a conundrum. You know, I've, I've done a few startups. I'm a bit of a mutt. I've done the corporate world thing. I've done the startup thing. I built a couple of companies, been involved in selling a couple of companies. Right. So it's just I've not— I haven't done it 50 times, but I've done it 3 times.

Hey.

Yeah. What I've realized is that you have to walk the line between speaking in soundbites that people understand.

Yep. And giving them something that's actually new and valuable.

Yeah, because the people— it's very interesting in HR tech. By the way, you could stop me at any time if I'm digressing.

But in HR tech, just to be clear for folks, I mean, not everybody's in our space, right? So We're talking about, let's just say, 4 market segments in HR technology. That is, HR technology encompasses human resources and HRIS, but also talent acquisition, recruiting, sourcing, and everything. But there's 4 market segments. You've got, right. So you've got the agencies that are firms that do recruiting on behalf of, as they act in agency of, they are an agent of an employer. So, that's one market. Then you have the in-house, which is the employer directly. Those are the enterprise organizations that do their own recruiting for themselves. Then you have the small businesses that these could be like a consultant or a solo practitioner that sometimes they are embedded at the company or whatever, but they're kind of a consultant, right? And then there's the I guess the fourth segment would be like the technologists, I would say, right? So those are the four segments. And in HR Tech or TA Tech specifically, you've got this, this like conversion of various needs that they've all identified, apparently all the market share, and they don't. So agencies and in-house and the technicians and the small businesses, they, they have different needs. But these products that I've seen out there, they all offer AI-powered matching, outbound candidate automation, efficiency and work optimization, CRM and data enrichments, right, with talent intelligence. They all offer these 5 things, but only 1 or 2 of those things are really important to each of those sectors. And I think it's a miss.

Yeah, I think the, the— by the way, thank you, that was really good. The, the paradigm that I've kind of like observed and often participated in, to be honest, in HR tech, perpetuate the buyer and the consumer are not the same. And the issue is, oh my God, that is a huge By the way, I'm stating the obvious, but it's really profound.

It's a huge deal. Okay.

And so what happens is the buyer is typically on it. The economic buyer is typically uneducated, uh, in the details of what it actually takes for the solution to work. But they know because they just read an article or came out of a meeting or spend time with a Gartner analyst, whatever that is. And so in order for the economic buyer to say, okay, you're in the realm of things that I think may be interesting for my business, you have to use a certain lexicon. And so what happens is, is this kind of like word dictionary, if you would, that a lot of the tech companies use tends to be kind of rinse, wash, repeat, like the same.

That explains it very well.

Not because they have shitty products, but, but it's hard to say the things to the economic buyer in order for them to put them into the map of like, this may be interesting. Then this is why often there's this kind of like dichotomy. It's like on a demo you often hear it's like, oh, I didn't know you can do that. Oh, I didn't know you can. The reason being is the demos are usually with the consumers, not the economic buyers. Yep. And the consumers are people like yourself that actually run the tech, that do the outcomes, right? The economic buyer is not responsible for the outcomes. They're responsible for the infrastructure, for the people to get the outcomes, right? But then the practitioners are the ones that are responsible for the operating of it and to get the outcomes that are desired. And those people want different things. And so the challenge is those people aren't the economic buyers, and they're not the ones there— you should do the due diligence. So this is why the disconnect. This is why I mean, you can look at 60 websites, they all pretty much say the same thing, the same thing. The truth of the matter is the product is actually very different, right? And unfortunately, you know, you can't really market your differentiation because honestly, it's not believable, right? And this is like, it's weird, you kind of have to see it to believe it. So a lot of times they do kind of like generic marketing communication language to appeal to the economic buyer. Did they find, like, the influencers and the recommenders and the advisors? And then they'll kind of tailor the messaging more to them. And then when it comes time to actual the product, hopefully they could, you know, get some traction with, you know, the user, the consumer, right? With the user and then make the recommendation to the buyer.

So it's like, Michael, so I don't want to cut you off. It just said this, this has always been in, in my head here, like, uh, our industry is rather, is rather unique in, in a couple of ways. There are other industries that have, um, a relationship between the economic buyer and the practitioner or consumer, and then the quote unquote customer, but we have a fourth dimension. That is forgotten. And I think that that's that, that one thing, because the typical triad applies to most businesses, but in ours, and there may be a few others where this applies to. So I'm, you know, not saying that we're the only industry in our industry. It's a quadrant, not a triad. So you've got the economic buyer, then you've got the user that's going to use the, you know, the buyer never uses the tool. The user usually doesn't buy the tool, right? Typically, unless that's the SMB, solopreneur, they are the ones that buy those, but they're buying a different product. And then you have the supposedly client, which is the hiring organization, the hiring manager.

Yeah, they're the beneficiaries of the output.

But in our industry, we have a fourth—

The candidate?

That's right. It's another vector. And that tends to—

It's the participant in the supply chain. Yeah.

And so in other supply chains, the other supply chains are supplying products products that stay delivered. Okay, when you supply refrigerators, our products change their minds, right? The refrigerator gets to the consumer's house and you plug it in and there's no refrigerator that's going to look around and go, oh hell no, get me out of here. I want— right? So we have this other dimension and we don't measure it. I'm going to tell you, I'm going to give you like—

this is—

I'm going to redact it as much as I can, um, and if I redacted too much pull me back in. But a customer client of my consulting client of mine, because that's, you know, what I do. I work with firms to help them improve their internal talent acquisition. One of the most pivotal and important discoveries during the discovery session of my consulting engagement was completely unexpected. OK. And it was that despite all the things that we know could possibly go wrong, despite all the things that we know we could improve on, there's always ways that we can improve. There was this one thing that just stood out when I did the data analytics and it was shocking. And it affects everything else in such a deep way that you could fix— I'm going to just say this bluntly. You could fix all of the problems and improve all of the processes and still have a problem because of this one thing. Okay? So this organization had a 75% 75% of their requirements, their requisitions, had to be reworked. So it's like this, what was the root cause analysis of that? That— so the— I, I can go back to that, but I'll tell you what, what I identified as the symptoms that was really interesting. So they had an offer acceptance rate of 80%, they have, which is respectable. A lot of companies strive for you know, 80%, 90% is amazing, 100% is unheard of, 60% is a little bit on the low end, but somewhere between 60 and 80, or I don't know, what do you think, 60 or 90, somewhere in there?

Yeah, I think 3/4 of your people accept you're doing well.

Yeah, 75% of your people accept. Great, perfect. I like that. That's a great margin. So there's that. And then they're like, okay, so we need— we know that 25% of our recs will have to be When we get to the end zone and make an offer and it's not accepted, we'll have to kind of start over, you know, usually because at that point you've already eliminated all the candidates and that. So that's 25%, right? But then in addition to that, and this was never calculated, there is also a 30-day, 60-day, and 90-day drop-off. There are people who accept the offer and show up to work and leave after 30 days, 60 days, and 90 days. Very few people leave after 90 days until they've been there for a couple of years. So it wasn't a retention problem. In other words, when someone stays, when someone passes the 90-day probation period, they usually stay for a while, but they had a pretty high bounce rate before 90 days for people that literally turned around and, you know, I made a mistake. This is the wrong job. Not from bad hiring. It was, it was not, we fire all these people. It was these people like literally got here and left. And guess what that number was? About 50%.

So half the people were leaving in the first 90 days?

In the first 90 days. Some of them didn't show up to work. So there was a small percentage, about 10% of the people that accepted the offer and didn't show up on the first day. That's a little weird, but It's the industry, in the industry that happens. So that's normal for that. So now you got 25% plus 10%, that's 35% already. And then there was another 10% at 30 days and another like 9% at 60 days and another like 9% at 90 days. And so when you add all that up, they have to rework a lot of the recs and they're looking at numbers like, for example, the absolutely ludicrous time to fill number. I was just like, this is what they told me. They said, well, we have a time to fill. At least this is what they're communicating to their leadership. We have a time to fill of about 47 days. Great. That's pretty, pretty good. 45-day time to fill is good. You want to try to get it to 30, but in a country where you don't really need to give a lot of notice, if you just do 2 weeks notice, You can see how it might take 2 weeks to fill the job, 2 weeks for the person to start. Okay. But usually it takes a month to hire someone and then 2 weeks. So that's where the 45 days comes in. Well, anyway, that number was completely off, Michael.

It has to be off because the job was filled twice.

So really, they're in some cases 3 times. So when I actually calculated their time to fill, average was 175 days.

Time to stick, right? Not time—

nobody saw that. And I was like, you can't— we, we can, we can hire faster and you'll still lose people. We can hire better and you'll lose other people.

I've been around a while, you know, and I have some gray hair. Um, me too. You know, the, uh, they don't have a hiring problem. This is another problem. There's another problem because what I, uh, I remember I'm gonna put on— I used to be in the recruitment outsourcing business. I ran Pontoon, which is like a big RPO. And I remember talking to a client. I'm going to be funny here for a moment. So stay with me along this line. So I remember going in and there was a big room of people. It was in Portland, Portland, Oregon. And And they're like, you know, we hire like 4 to 4,500 people a year. Um, so I'm like, holy crap. And they're like, how long you been doing that? For years. I'm like, okay, so would you have like 30, 40,000 people? He's like, no, we have about 4,500 people. I'm like, well, that's not good.

So you're rehiring everybody?

I was just like, I swear I have to go I went to Columbo mode because otherwise I was going to be like offensive to someone. So I'm like, so what you're saying is you've been hiring for years, you hire 4,000 to 4,500 people, you're about 4,500 people. So they're like, yeah, you know, we have a lot of churn. What I found was, and this is just in that specific instance, because, you know, you get the blame game, like everybody's getting blamed, but at the end of the day, like It was really interesting. I'm like, you know, your recruitment marketing talks about how you're a paternal maternal organization, how you take care of people, you know, bring your dog to work on Thursdays, wear Birkenstocks, you know, everything's hunky-dory. But I've been here about 3 hours. And to me, it seems like it's all about metrics. It's all about dollars. And it's all about corporate compliance. The issue that I see is somebody thinks that they're coming to work at a place that values work-life balance. And it's all about like the feel-good, which, you know, great vibe, great culture. But when they get here, it's like sterile, white glove, every hour, what have you done? Every week. You have one-on-ones to account for every minute of your act. You see what I'm saying? So it's like, this is, um, yeah, I told him I'm no— I'm not a, a PhD in psychology.

I don't need to be for this.

I were being interviewed for a job where I could bring my dog to work on Thursdays, and I got here and I couldn't bring my dog to work on Thursdays, I'd be kind of pissed. So Anyway, the CFO was cracking up laughing. The TA people weren't.

But the thing, it's like, in a funny way, you have to cash the check that you just deposited.

Yeah, I made my point. My point is, the real issue is the non-alignment. And there is no authenticity between the promised job and the real job. And, and we literally reduced churn by 70% just by aligning the real work environment with the advertised work environment. And they had to completely change the kind of people they were recruiting because I'm like, you don't want those people. You want the kind of people that want rigid structure, the numbers.

Yeah, right.

And they were afraid. They're like, well, there's not a lot of those people. I'm like, well, if you start looking for them, like, you know, you, you get there. So I said, you know, you got to hire people that are wired slightly different behaviorally because you have this aspiration of we think these kinds of people are, you know, but the reality is if you look at the benchmark profile of somebody successful in your organization, very little bridging of that gap to the kind of people that you actually are hiring for.

Aspirational but not actual.

Yeah. And I'm like, there's no fault here. Like, no one, no one gets up in the morning, go, aha, today I'm going to lie to all my candidates. It's not that. It's not— it was nothing nefarious. There was no ill intent or anything like that. It was just something that was happening that no one really questioned.

Yep, that's exactly what that— what I just encountered. It was they were trying to fix the upstream problem, not realizing that the downstream problem, um, wasn't— they didn't know about— was actually partly why the upstream problem looked like it was such a big deal.

Yeah, it's, uh, it's really interesting, and To me, it was really all about when I hear churn like that in the first 90 days. Hey, somebody used to work for me. Oh no. Hi, Jen. How are you? When I hear churn like that in the first 90 days, my antenna goes up. And I immediately start thinking, okay, there's a non-alignment between the promised culture and the actual culture. Yeah, yeah, yeah. A Deco RPO days.

That's a big gap.

We're going back a few years, right?

That was a Deco.

I have some gray hair now, Jen.

Um, but yeah, so that's kind of, uh, so if you were, you were at a Deco back then, do you remember Harriet McCormick?

No, I don't remember.

About 25 years ago.

I, I started at Deco like 2006.

Oh no, no, no, this is like '96.

Uh, and then I left in 2014. Into 2013, '14. Yeah, I was there for like 8 years, something like that.

This was right way, way, way before that.

Yeah, yeah, no, I wasn't there that long. Um, yeah, I started there, uh, I was actually a consultant, weirdly. Um, not, not weirdly, it wasn't weird at all. I'd say weirdly because I sat outside the CEO's office and I spent a lot of time in there. Everybody thought I was like his replacement, which I was. I couldn't figure out why for 6 months everybody was afraid of me. I'm like, what the frig is the new boss? And you're like, no, thank you, John.

I appreciate some praise here.

So I, yeah, only 6 months later realized I'm like, oh, everybody thought I was gonna be their new boss. When they found out I was just a lackey, then everybody started abusing me. But no, I'm joking. But I was there during that, that time. And we had built, you know, we built an MSP and RPO business. I then focused more on the recruitment business. And Laurie Hawk had taken over the MSP business. And then we acquired a company called Talent Track. Yeah, with Kim and Wayne, based in Toledo, Ohio, and spent a bunch of time there. So it was, it was really interesting. I, you know, I lived the recruiting world, we had a lot of fun stories, because that was like the, you know, when you could put fog in a mirror, you can get a job.

Has breath, can do job. Let's— sorry, yeah, go ahead, finish.

No, no, it's okay. And but we also lived through a downturn where the market crashed. And, you know, people were carrying boxes out of, you know, Lehman and those things.

So, um, at least 4 of those already.

So yeah, we've seen that. Yeah. Jen, it's okay.

So I want to go, you know me, I want to go unfiltered for moment. What is a bold, audacious take that you have on the way companies are using, actually using labor market data and talent intelligence today that most leaders would disagree with mightily? So what is a bold and different take that you have about how companies use labor market data that the leaders would disagree with?

That's an interesting question. Let me warm up.

And it can't be everything.

It's okay. Let me warm up into that because I need to ease into that. Okay. Because I don't want to give you a knee-jerk reaction.

Okay.

So let me start giving you a stream of consciousness and I promise I'll get there.

Okay.

So I think the top of mind right now, and when I say right now, The problem is, is you get up every day and it's like, I want to turn on the TV, right? It's like every day there's something different, right or wrong, it doesn't matter. I'm just saying it's different, right? And what freaks out leaders is different. See, leaders aren't necessarily freaked out by the what, they're freaked out by the different because leaders want to plan and they like to have feng shui and harmony. And the different every day freaks them out. And so this is what I mean by when I say right now we're in this period where there's change. And when you think about change and how that impacts talent, some leaders are thinking about it more traditionally of like, how do I geo-source talent and put people where they should be, right? So like talent location strategy. Yep.

Okay.

That's not like the bold, audacious take. It's strategic. And not everyone's doing it. But it's not the bold, audacious take, because it's something that should be done. And it's a strategic way of doing it. More the bold, audacious take is basically looking over at my neighbor's lawn concept, because you could use labor market data to track what your competitors are doing. Yeah.

Okay.

Um, and, uh, understanding how they're constructing their workforce so you can compete better. So maybe like you want to go to like what Josh talks about, like the super worker model, where like maybe instead of figuring out where we should locate our people and how much we should pay them, what we should really figure out is How do we replace the people that we currently have with better ones that can do more work with less? And no one's really going to talk about that because it's not— it's not politically correct, politically correct. But it's not that they're not thinking it, Joe. Because at the end of the day, you know, if you have, let's say, let's not even take software, right? Like, step back and do it, be practical. Let's just say like accounts receivable. Every company has to send invoices and collect invoices. Do you agree? Even the smallest business. Yeah.

Even a solopreneur still has to. Yeah.

So let's look at basic shit, right? It's like the basics, you know, the very, very basics. Every job is ripe for disruption. And so the question is, their thinking is that like, not how do I hire 4 better accounts receivable clerks? It's like, can I just hire 1 instead of 4 that has like better digital skills? And this way I could cut my cost and cut my workforce. Because the thing is like with AI, you know, it doesn't require PTO. Um, you know, there's no— you know, right now AI is not being taxed. That's, by the way, a whole nother conversation, because my hot take is we don't realize this, but we're at the eve of like the government taxing AI agents and bots. It has to happen. The reason why— and people like, wait, what, Bagelman smoking crack? I assure you I am not. Is if you radically reduce the size of the workforce to make it AI-enabled, that workforce isn't going to pay more taxes, and the government sure as hell isn't going to want to collect less taxes. So something has to be taxed. And right now, companies aren't dumb. Like, let's say I'm employing 5 people. There's employer-side of taxes I have to pay, right? Matching FICA, FUTA, all that crap. So for the first, like, whatever, $100,000 plus, I have to match, you know, up to 14% of your pay. But what if I was to replace you with an— with a bot? Let's say that bot cost me exactly the same salary. I at least save the employer side of taxes, and I don't have to pay any benefits. So let's say I have to pay you $100,000 in salary, but your fully burdened salary with benefits and taxes and all that crap cost me $135,000. Even if I have to pay $100,000 a year for a bot, I'm still saving $35 grand. And like, honestly, like, a CFO would be a liar if they didn't think about that, right? If they didn't admit to that. Because the reality is that— so I think that ultimately, not to digress, if I'm just being practical about this and understanding macroeconomics, the advent of, you know, AI enablement doesn't matter what industry you're in, is going to transform labor and the size of the participation of the labor, which is going to impact the government's ability to collect taxes on the labor. So AI is going to have to be taxed. It just is. And when it does, that's when it becomes really interesting, Joe, because companies will start comparing the cost of AI versus the cost of a person. And when we think about the world of recruitment, I think recruitment is the petri dish, it's the test bed for this concept, because a lot of people are prognosticating the disruption of recruitment based on AI, that, oh, you don't need to hire as many people, or you could do a lot of the automation. I think this, my stream of consciousness is going to get tested here in the world of recruitment sooner. But I think the hot, the hot take is, is like, how do I get rid of 75% of my workforce and replace them with better workers with digital skills? That's the hot take. Not a popular take.

So then that's not a popular take. But I think it's realistic. And I think that's kind of what it just going back to in my lifetime, what is a similar transition to this, which is a transition that fundamentally shifted the balance of power in the proletariat, if you will. So, you know, and there may be another one that I'm not thinking of. So, you know, this might not be the best example, but it's the one that came to my mind when, um, When Al Gore first plugged in the internet and put fresh batteries in it or whatever, you know, claimed that he invented the internet. But when the internet really became a, um, a utility. Okay.

So it used to be basically like a public utility.

Used to, yeah, it used to be a luxury at first and, and you didn't need it in order to operate a business. But then at some point it became something you have to have in order to operate a business. That point, that transition shifted the power from, um, those who held organizational knowledge and had no— those who had the knowledge and those who seek the knowledge, right? So that transition, that disintermediation of knowledge, had a huge impact in the workforce. It changed, it basically created the knowledge worker economy. It became important. Organizations built— organizations that were built exclusively on intellectual property or knowledge that didn't even have real property, like they didn't even have a real asset, as in there was not even an office.

It was an intangible.

It was an intangible. So, that was a huge shift. And what that did is it created a whole other type of worker. Before the internet, we didn't have I don't even know how to tell you, thousands of the job titles that we currently have. Yeah, software engineer, web dev, I mean, there's so many. So it created a lot of jobs. Those people who do those jobs or did those jobs to transition into, they came from another industry. So when the internet came around, other industries died because they evolved out of business and new industries were created. That I think is similar to what's happening today, and it's not I don't think it's even close to what people are talking about with regard to social media, because I'll tell you this. When you talk about social media, it used to be, used to be social media was something you talked about and it was a different thing. So you have, you know, uh, advertising and social media, right? And now it's become part of the background noise. So it wasn't a utility. Social media didn't become a utility. Social media merged into media and it's now just called media. So media has always been media. The format changed, but it wasn't fundamental. Internet created a whole new industry. Anyway, a long way around to saying what I think is happening right now is that change because it's once again shifting the power of knowledge. In other words, the organizations had knowledge that people wanted. The internet gave people knowledge. Now, the internet, think about this, the internet has knowledge, the chatbots are the, you see what I'm saying? It shifted again, right?

Yeah.

You no longer need the internet for knowledge. You could literally right now get a lot of what you need out of, you know, you could learn, you could do a lot with just a basic chatbot and a basic chatbot The technology that we call AI right now is brutally antiquated. It's like so old that there were people with cranks and not even electricity that were operating machine learning intelligence 50 years ago. So this isn't really revolutionary, but what is revolutionary is the fact that it is democratized outside of the boundaries of any containment. So it used to be the knowledge was contained by the organization and it broke out and everybody has it. Now the bots have the knowledge and it broke out again.

It's true. I would say this is— remember we started out saying some things we agree on, some things we banter about, right?

But this is my opinion and I think it's just—

I don't disagree with the, the, the, the backplane of your thesis. The premise I think is slightly different in how I view it, which is, is that I think it's very powerful within a corporate firewall where the data is—

Oh yeah, absolutely, absolutely.

Versus I see too many executives essentially use ChatGPT now instead of Google.

That was kind of what I was—

Yeah. And they literally base all their knowledge on that, which is not correct. And this is like very dangerous because you have very smart people learning wrong things.

But Michael, the same thing happened with the internet.

No, no, but what I'm saying is— but this is why I disagree with you, okay? The, the same thing did not happen with the internet. The internet requires the user to be the arbiter of what is true. ChatGPT represents it as truth. That is a fundamental difference.

Oh, it's not representing that. It's an opinion. It's telling you.

No, no. So when you do a Google search, there's 39 links. You as the user are forced to go through all 39.

You have to determine which one has authority.

And make up your own opinion about which one you believe is true.

I got the difference, but ChatGPT—

When you're using ChatGPT, it is the truth. So do you—

It gives you one truth.

From the oracle. That's the fundamental difference. And that's why I think this is very dangerous, because you have very bright, well-intended leaders that are developing a whole new concept of truth based on something based on hallucinations of a large language model versus a search engine requires the user to use their IQ, judgment, discernment. It just requires memory versus traditional search and reading. Like, why do you read a book? You learn, you have to absorb, you have to process. They even test for reading comprehension in school. Did you understand what you read? Can you apply what you read? Does that make sense? But when—

absolutely does.

And I think he's giving you this stuff. So I am a huge believer of firewalled off internal use. By the way, not a plug for Claro or Wilson or the products that we develop. But when we developed our AI-powered labor market reports, the reason why we won product of the year is because you could go to bank on it. See, follow up, meaning that we have cleansed, structured, created taxonomy around the data. So when AI is processing that information, it's not reading Wikipedia.

Do you follow?

This is really important where it's unverified materials.

What you put in was already filtered for facts.

Of course. Yeah, of course. Versus the danger as I see it, opinion seen as fact by ChatGPT. Yeah, this is the danger. And this is like I, I'm an eternal optimist. Those that know me will tell you this. I'm not gullible and I don't have irrational exuberance, but I am an optimist, right? And I believe this will get solved, but it's in the short term. I think that I would recommend caution on using AI synthesized information based on the internet of data. Because I can tell you, if I scroll through my Facebook feed, it seems like I just need to take those 3 pills and I'm perfect, right? And, and what—

don't worry, it's—

so what AI is reading is those 3. Do you understand what I'm saying?

Those are the 3, that's it. So, and then you're not sick, you just need these 3 pills.

Yeah. So the challenge is I'm starting to see senior people embrace the use of open source LLMs like OpenAI's ChatGPT, and they're using it to synthesize the internet. And to be honest, it's a fucking— excuse me, it's a shortcut to learning.

Yep.

Because instead of reading the McKinsey report, they now have GPT synthesize the 79-page McKinsey report. And I'm not judging the people for doing that, but I could tell you that if I read the report versus the AI synthesizing the report, I would have more depth and fundamental knowledge than the person that would just use GPT to summarize it. Because essentially AI is the world's version of Cliff Notes. And like, exactly what it is, minus versus the book, I get an A minus. Do you see what I'm saying?

No, no, it's exactly what it is. Exactly what it is.

And so I feel that the challenge is— let me come full circle to where I say the hot take, right? As I said, I would warm up into it, brother, right? So the hot take is in utopia. I think this desire to morph to a super worker that has all these digital skills, that has AI enablement where one person can do the work of four, that's kind of like utopia. Yeah, because that relies on augmented intelligence that can educate this super employee by synthesizing a lot of information into sound bites, and that super employee is supposed to have the same depth of understanding as someone who has domain expertise. I think the hot take is, is that ain't gonna work. And, and I think the, the reality is, is it's not gonna not work Unless the world becomes okay with everything being C-.

Idiocracy.

And, and if the world becomes okay with C-, that will be the new A.

Yeah, I got—

there's your long-winded hot take.

Yeah, the new— the C- now will be the future's A.

But, um, I'm here to tell you I have both good news Sorry about the F-bomb before.

I heard nothing.

Beep.

No worries. The bad news is that what we currently have, what we currently call generative AI, is going to collapse. It is on the way to collapsing. It's probably already collapsed in a lot of different ways that we're not even aware of, but it will definitely have a catastrophic collapse. The good news is we needed to do that because we needed to grow up and be something else.

Public generative AI will collapse, but That's what I'm talking about. Yeah, I agree.

I think, like, to be clear, I'm not talking about if you use, um, AI internally for a scientific project.

I'm talking about ChatGPT and Gemini, you know, that's a good chatbot, and it's really good because it's internal. Yeah, yeah, because it used— because I think the, the BS of, hey, can you synthesize all this unverified data and educate me that will die. What will come out of that though is like the concept of like retrieval augmented generation or RAG as we call it, where AI can play a pivotal role is connecting disparate systems of quality data to give you like help with advanced reasoning. Do you see this is where—

Let me give you a perfect example. So there's, and this is where I think that the line is drawn, right? And you're right, I said, I said, I said GenAI, but I should have clarified that or qualified that as public GenAI.

Yeah, GenAI based on public data.

Yeah, based on public data. So there's many ways you can use ChatGPT, but there are two primary ways that, you know, two all-encompassing ways. There's you getting information from ChatGPT, as in asking it to tell you, like, give me something. And then there's you giving it information and asking it to do something with it. So there's two different things.

It's task automation. The task automation.

There's that and then there's the— so there's, there's who's the President of the United States, which you're not providing any information. GPT is giving it to you. So that's one thing. That's been a collapse. The other one is not. And this is a use case that I'm going to give you that's real life example. So there was a client that I'm— there is a client that I'm working with, and their system has multiple databases for the same reference index. So the ID, the candidate ID, is the same across 7 different databases, but those databases track different things. One tracks activity, another one tracks applications, another, you know, different things, and they're not really very well connected. So in order for a report to run a dashboard You have to use all 7 sheets, spreadsheets. Well, they've been manually doing that. So I came along and I downloaded all 7 sheets in CSV format and I uploaded them to my personal private GPT and I told it, merge these sheets into one. And I can— it says, do you want it this way or that way? Do you want vertical? And I said vertical. And it grabbed all of the missing columns, all of them, and put them into one. Perfectly and seamlessly without error. And now I have one sheet instead of seven.

It's task automation.

Like, that is AI, but it's AI using my information.

Yeah, yeah, it goes back to what we were saying. I think task automation—

automation is not AI, right?

Input, input data is good. Where the challenge is, is that, um, using it for, um, as a replacement traditional learning behavior.

That's where, yeah, right.

That's the risk. And I think that, you know, we're, we, you know, this is probably another topic for another conversation, but we're in a skills kind of environment. And I don't know if people just repeat that all the time, but not really understanding why this is important. But, you know, there are critical skills that are required in order to— and I don't mean just the digital skills of click this, click that, but there are certain skills that are required to make sure that— how should I say? I'm trying to be PC about this. But let's say there's certain skills that are required for things not to go south. This kind of whole skills focus kind of helps companies ensure that the people that are operating these things are best suited to do so, in very, very, very simple terms. And it's not just about like, you know how to use Excel, tick, you know how to use this, tick. There are different skills like reasoning skills and etc. And so those things are going to become important. This kind of goes to the topic of skills aren't just hard skills, there's also soft skills. Um, so anyway, I like— I'm afraid to go down this rabbit hole because that's like an hour conversation, so I'm gonna hold myself back out of that rabbit hole.

We'll have to do it next time. We are getting pretty close to the hour, so I'm just gonna literally put you back in the, uh—

yeah, I gotta pull myself back out of that rabbit hole.

And we're gonna— we're gonna do this again.

So I'm gonna send you like a boundary economy. We do a stuff in that area. And like I said, I'm afraid to go down that rabbit hole. But I think that ultimately, there is a higher level than just the headline of like, oh, you know, it's all about skills. I think this is one way that we can mitigate this kind of challenge that I foresee to kind of mitigate the hot take.

Thanks for that. I appreciate the perspective. And I I'm gonna definitely bring you back so we don't have to try to cover all of it today. I had another question that we just don't have time for. I do need to ask you my two wrap-up questions that I always ask, and one of those, which I'm very curious to know your answer to, is who should be our next guest?

Who should be our next guest? That's a great question. I would love to hear from like a product leader at a, like a chat-based company in HR tech. Because I want to, okay, hear from them about how they're— because clearly they're not thinking about chat, like Paradox, for example, maybe get But clearly they're not thinking about chat because they've been doing chat for 15 years. So their product roadmap isn't like, oh, I'm going to make a chatbot.

We're going to do that, right?

Yeah.

Recruiting.

They've been doing that for 15 years. I would love to hear from the people that were early, early in the AI enablement. And I don't mean like high-level applications, not like the hire easy kind of companies that use AI behind the scenes to enable their high-level platform. I would love to hear from an AI-native, AI-first product leader, meaning whose products are built using, uh, conversational stuff. Do you follow? Yeah, maybe somebody from Vizier that's built out like their, um, you know, uh, V platform or something like that. I'd love to hear from an AI-native product leader about how they're thinking about the maturation of their product, because that's what we're going to get the real bleeding edge versus talking to me like I have a traditional software product that I'm transforming to an AI-first product humbly, but going the right way, right? I'd love to hear from an AI-first product, how are they transforming? Me, because that's the exciting thing. Like, I'd love to hear from one of those, and I think it would be interesting because there we would get like—

challenge accepted—

we're seeing over the horizon. Yeah, how they're thinking about leveraging AI.

That's a good— I'm gonna, I'm gonna take that on because I actually, um, I'd love that take as well. I mean, you know, you're, you're a product leader, so product leader from that other type of business would be, would be fantastic. The last question, um, is just kind of, uh, what, what do you want the audience to think about over the weekend? In other words, not— I don't, I don't want to, you know, pulpit lecture.

Yeah.

But so a question the audience should ask themselves.

Yeah.

Over the weekend.

It's great. Well, first of all, it's a holiday weekend for many people in the world.

Yeah, today's for a lot of people.

First and foremost, spend some time with your family. You know, family and life is important, right? I know we're immersed in work all the time. We think that's our life, but step back for a moment and, you know, enjoy the weekend. Hopefully you have some good weather to do so. So the first thing I'd like to do is ask them just to like reflect, you know, on that and enjoy the holiday. The second thing, you know, more intellectually, you know, ask yourself if like, if your role in your profession was to be redefined, how would you redefine it? Often change is thrust upon us. It is what I call subjects of change.

Right?

We don't often think about like, what if we were the change driver? And, you know, step outside of yourself without being like greedy or egotistical or self-serving, and just say, I'm in XYZ profession. And if my profession was going to undergo a transformation— and I don't mean evolution, I mean revolution— like, how would I redefine my role? By answering this question, I guarantee when they show back up to work on Monday, they will perceive their role very differently.

Very— yeah, that's a good point. So you don't have to prepare for catastrophic change, but if you were going to change, what would it be?

Yeah, like if your role was going to change and you were going to be the master of that and do it in a way where you were a bit kind of like removed existential about it, right? It's great, you know, it will give you a different take when you show back up to work on Monday.

Yep. And hopefully it's better. Appreciate it.

Anyway, I enjoyed the time, brother.

Oh yeah, we're gonna do it again. I'm gonna, I'm gonna, I'm gonna force you back.

We killed an hour, like That—

yep. And that's gonna— that happens when we're having fun. That's what I find, for sure. Yeah, man. So, all right, have a good weekend, a good holiday weekend, everybody. Hopefully we will see you back here again next Friday. We have Alex Marcus who's coming from, uh, he's a product leader, um, from like, uh, the ATS side, iSIMS, things like that. So, um, he's, I guess also a product leader, but not the chatbot one. I wish I'd just thought of that first.

Think of it as an AI.

Yeah, right, right. Yeah, he's the founder of, um, of a company, um, that is called Integral Recruiting Design, IRD. But anyways, so hopefully we'll see you back here. Yeah, next week.

Great.

Thank you. All right, be kind to yourselves. Bye-bye.

Cheers. Bye.