Setting people analytics free!

• Podcasts can be fun and still pack quite a bit of learning into the mix.
• People analytics is a vital component in understanding and improving organizational dynamics.
• The COVID-19 pandemic has accelerated the need for robust people analytics.
• Technology and AI are becoming integral in advancing the field of people analytics.
• There are ongoing challenges in data privacy and quality in the use of AI and analytics tools.

Unedited Transcript

Transcription for:  

“S4HCole.mp3” (Uploaded File) (New  Transcription) 

Speaker 0: Welcome to Science for Hyre. With your host, doctor Charles Handler. Science for  Hire provides thirty minutes of enlightenment on best practices and news from the front lines of  the improvement testing universe. 

Speaker 1: Hello, and welcome to the latest edition of Science. It’s for hire. I am your host  doctor, Charles Sandler, and the parade of all star guest just keeps coming. Today, my guest is  Cole Napper, I will say he is a fellow Louisiana person here, which there’s not that many of us.  We all have to stick together.  

And so he’s a veteran of the Louisiana Tech program, which is really a great program. It didn’t  really exist when I was looking for grad school. So glad that we have options in the state. And as  I always do, I’ll let Cole introduce himself. But one more one more kind of thing to mention is  that I was on calls directionally correct podcast, which was the most fun podcast I’ve ever been  on, and he’s really inspired me to make mine more fun, so we’ll see what we can do.  I don’t have any silly sound effects or anything like that other than the ones I can, you know,  make on my own. But but we’ll be able to, I think, have a really fun journey today through a lot  of different topics. So without further ado, my guests, cool, never. 

Speaker 2: Yeah. Well, we can we can make some sound effects if you want. I don’t know if the  audience will like it. Though. 

Speaker 1: There’s a dog in the office and I’d be able to get it to bark on q. 

Speaker 2: But, yeah, like, maybe I’ll do kind of a nontraditional introduction just because of the  the context you said, but I am originally from under Louisiana, which is probably a big part of  my identity, and I appreciate you bringing it up because I’m very fond to the state of Louisiana. I  actually live in Texas now, but I live a stones throwaway, you might say, and I come back quite  often. I’ll actually be there later this week. And Louisiana is the only place that that I know of  that has, like, drive through deckeries. And So I’m a I’m a big fan of that when I go back.  It’s a it’s just a cool part of the culture and everything. But so like you mentioned, I I graduated  from Louisiana Tech’s Io Psychology PHD program. I was actually in the first cohort there, and  so that definitely dates me in terms of the the time frame of how long that program has existed,  but I’m really a huge fan of that that program and and some of the faculty there like Tilman  Sheets was my adviser. He’s just a phenomenal guy. And I think they they’ve really done things  the right way in terms of bridging the scientist practitioner gap.  

My day job, I I’m the VP of People Analytics for a company called Organic. And as you  mentioned, also I’m the co host of the direction correct podcast about People Analytics with Scott 

Hines as well who’s another graduate of Louisiana Tech’s program. And we do try to have some  fun. And so I hope I hope we can have some fun on here today, Charles. 

Speaker 1: I know we will. Absolutely. So I’ll tell you you have to drive through Jacklyn. 

Speaker 2: I did have a question actually. Yeah. Is You said it was the most fun you’d ever had in  a podcast. Was that the only podcast you’d ever been on? Because I was like, I mean, we’re we’re  a low bar. 

Speaker 1: Well, you know, I mean, I’ve been on a few. I’ve been on a few, but not a whole lot. I  expect to probably be on more as time goes on and people find me, etcetera. I don’t know. I  mean, you know, I’m pretty I’m not a buttoned up guy to begin with, but on these things for  whatever reason.  

I’ve I’ve kept it a little bit more business like and you guys had a lot of little fun non sequencers. I  will say, hey, and this is I don’t know when this episode exactly is gonna air, but I am working on  taking what you inspired me to do and turning it into almost like a little quiz show live video  stream called take it or leave it. And in in that, 

Speaker 2: I’m intrigued. 

Speaker 1: Yeah. In that, it’s gonna be, you know, I’ve actually played a game show host before  in a lot of different, you know, roles in terms of dramatic productions and things. And so I I’m  good in that role. I can really I can really get into the character. It’s just gonna be current, you  know, news articles.  

I’m a have a guest and we’re gonna, you know, thumbs up. Take it or leave it. There will be sound  effects most likely, and I’m learning to do, you know, video production. I don’t know if you’ve  ever seen something like this. This is a called a stream deck.  

I had never heard of one of these before. But I 

Speaker 2: don’t know what that is. 

Speaker 1: Well, it’s got all these buttons. Right? And so you as you’re doing whatever your  thing is, you have macros loaded on there. So cue the intro, bank, press the button, you know,  cue the 

Speaker 2: Oh, it’s like a sound board. 

Speaker 1: Yeah. Kinda, but it’s all macroed out so you can just, you know, do what exactly you  need pre programmed for someone like me who’s not a sound engineer. So I’m gonna be playing  with that, but that’s gonna be a lot more a lot more whimsical and a lot more centered around,  okay, here’s some latest news articles, you know, about stuff we do. Right? Technology,  especially, you know?  

Is it bogus or not? And we’ll just we’ll just kinda have a vote. So stay tuned. You might get  invited to that. And it’s a safe bet.  

And so you mentioned the drive through deckery shops and so we do have those here and I I  belong to this in a car club with where we we all have vintage cars. And so we have a thing every  year called the decker y rally. So If you know about, like, the Paris to the car rally, it’s a take on 

that. And it’s like a true rally start. You you get a an envelope.  

You open the envelope. You stamp your time, and there’s heaps of people. And then in there, it’s  a it’s a map to four different daiquiri shops around the area that spread out. So you’re you’re like a  thirty or forty mile radius. Yeah.  

And you have a coat, you know, a coke pilot navigator. And you go in, you get the daiquiri, you  bring them back at the end, everybody brings them back and then, you know, you got doctors to  drink. No drinking while you’re driving. But and then you stamp the you stamp the time stamp  when you get back. So it’s like a true rally, but you’re actually collecting deckeries from different  decker shops, and we just did that.  

It’s every October. So it was super fun. 

Speaker 2: I don’t have a classic car, but that would be so fun to participate. 

Speaker 1: Yeah. You don’t have to. Some people there were people in, like, just an SUV, you  know, just you don’t actually have to have a classic car. I will say that I’ve done it maybe four  times in my success rate of actually Finishing has been about half. Seems like my stuff breaks  down a lot, but, you know, that’s that that just happens.  

It’s I’m used to it by now, man. The AAA is a really good investment. So let’s hear a little bit  about people analytics. I mean, it’s a term you see everywhere. If you’re just seeing people’s titles  on LinkedIn and stuff, it’s always, you know, some not always.  

There are a lot of people analytics, people analytics. And I’m sure it has a pretty pretty  umbrellaed meaning. But I bet it means a lot of different things too. And so as someone who is in  people analytics and as a podcast on people analytics, it’s probably damn good at people  analytics. Tell our listeners what it means to you.  

What should they think when they see somebody’s job title, you know, that says people analytics.  And why would you be out to somebody like that or something. You know? 

Speaker 2: Yeah. Yeah. I think I think it’s a good question. And there’s definitely not consensus  on what it means. And so I think it’s it’s it’s worth asking.  

And I’ll I’ll kinda pick on Scott here for a second just because he’s my cohost and we’ve talked  about it before on the podcast. But I think in his mind, I o psychology and people analytics are  somewhat synonymous. Uh-huh. I I am actually not in that camp. I I would say that Think of it  like a Venn diagram.  

There’s definitely overlap between the two, but there are I o psychologists that are not doing  people analytics and there are people analytics folks that are not doing I o psychology. And so I I  take a big tent definition of what you might call traditional people and analytics. Now, like, the  biggest tent would just be any analytics related to people, but I I don’t think there 

Speaker 1: That’s an easy out right there. 

Speaker 2: I I I don’t think it’s accurate either. Because there’s lots of analytics that go on in the  world about people that aren’t called people analytics. Right? And, like, just an example would  be, like, analytics related to sports players. Well, they don’t call that people analytics. 

Speaker 1: Right.

Speaker 2: They they just don’t. But that would be that would fit the definition. Yeah. Generally  speaking, it would be any kind of analytics or data about employees of an organization or maybe  even the labor market that’s outside of the organization and trying to use that information to  inform business decisions. It’s probably the easiest way of putting 

Speaker 1: it Gotcha. 

Speaker 2: In the aspects that are related to IO, is a lot of what we learn in terms of, like, the  research methods, the multi variant stats, and just understanding the cognitions of employees in  the workplace Right? The things that are people analytics but not really related to I o psychology  is things like data engineering, data management, some of the machine learning techniques, like  some of the more esoteric or sophisticated things that are going on out there you There are other  psychologists out there that know those things, but it’s not a part of our core discipline that we  really learn about in graduate school. 

Speaker 1: Sure. 

Speaker 2: And so I I’d really say that’s kinda where the Venn diagram loses its overlap. 

Speaker 1: I like thinking in Venn diagrams and, you know, I just wrote down on my notepad  here, data scientists. So, you know, now I’m starting to to visualize a VIN diagram that has all the  different things that, you know, IO’s do in terms of titles. Right? We we certainly work in  different areas, you know, executive coach or curriculum developer, whatever it is, but but more  just in the idea of, you know, how we’re we’re labeling ourselves around the the things that we do  with data, I guess, really. Right?  

That’s that’s pretty interesting. And, you know, data science is one that for the past probably five  years, I’d say, it’s really started to enter into the vocabulary you’re starting to see. And I think  even in my program, you know, there’s such an overlap between IOs and, you know, use of  statistical analysis even in more sophisticated ways. Right? We’re we’re trained To do that,  luckily for me, it was pretty simplistic back then.  

You know, I’m not really a good programmer developer type person. And it really crosses over  those kind of skills when you’re looking at, you know, writing code to do all kinds of interesting  things. It’s it’s a dimension I wish I had in my back of tricks, but, you know, I know people. So  that that’s how you get that done. You know people.  

But that’s that’s cool. So and I’m channeling you here. What what’s your wackiest people  analytics moment? What’s the craziest thing as a people analytics person that you’ve ever dealt  with that you’d like to, you know, tell that story? 

Speaker 2: Well, Ken, before I get into the funny because story. I think it’s an important  clarification on along the lines you’re saying about data science. And and I find a really practical  way to translate this for I o psychologist to understand kind of some of the similarities and some  of the differences and what it would take if you wanted to bridge that gap. Is thinking about  traditional research versus traditional product development. Alright?  

And so Traditional research, let’s say let’s say just for the sake of argument, you’re gonna do a  research project. You go through the entire scientific method. It takes six months to do, you  collect your data, you do all the the things that we know how to do really effectively. It just takes  time.

Speaker 1: Yeah. 

Speaker 2: Well, and and part of that process is usually modeling. 

Speaker 1: Yeah. 

Speaker 2: Right? Now a data scientist comes along and maybe they have a different lexicon in  terms of how they talk about things, but they they’re doing modeling as well. But chances are the  modeling that they’re doing isn’t going through the scientific method and process. It’s probably  putting what they call a model into production. 

Speaker 1: Yep. 

Speaker 2: Which means it’s just running natively in the background using software. And so the  way I put it is what data scientists are doing is just an extension of what we do. It’s just doing it  in real time. And so science shouldn’t take as long as it takes. And so how could you shorten that  time frame?  

It’s by putting things into production and building products around it to do it automatically. In  terms of, like, wacky people analytics stuff, let me think for a second because that that’s really a  good question. 

Speaker 1: So think about it while I come in on what you’re saying if your brain can work that  way. Hopefully, it’s not a cross cognitive processing challenge. I don’t think it is for you though.  So the first thing I thought of is sometimes this is not a knock against data science. This is the  way I’m using this hyperbole is is maybe sometimes it’s soulless, like, you just look at the data,  you follow the data, it’s almost purely empirical.  

Right? And and I’ve worked with a lot of data scientists. And the hole is greater than the sum of  its parts when you do that. I mean, there’s a lot of common procedures, but there’s just a  fundamental philosophy. If you’re not trained in any kind of psychology or, you know, more  mental measurement, whatever, that you’re just looking for features and, you you know, what  they call validation is, oh, the the model predicts, you know, it works, but but it doesn’t have that  external validity of, oh, this is actually predicting something that’s really important for us in a  specific way and accounting for variance in this.  

There’s subtle differences and I might even be butchering those, but but I feel like the best  combination is again this VIN diagram of of an individual or team where you have both of those  mindsets. Right? So you’re good at exploring and building the models, but you also understand  the ingredients of the models features of the model, you know, how important those are and what  they relate to. It’s it’s just goes back to the, you know, basic I o grad school idea of Dust Bowl  empiricism, which we had a lot of in our field a long time ago. And when you follow those  things, you know, what I think you lose a lot of time is the generalizability of the model too.  Right? You have this model, it’s very trained on a very specific set of data. And then you try to  say, okay, we’re gonna select all call center representatives across the world with this model we  built in, you know, Akron, Ohio, and it’s it it may not generalize, especially the more faceted it is  and everything. So that’s that’s kind of my my take on that. You know, it’s about the external  validity And I would say, you know, we’re trained to do the research methods if you think about  doing a validation study.  

Like an IO validation study. I’ve probably done a zillion of those. I love them every time I can 

because you can prove the value hopefully of of what you’re doing or show the company that  what they’re using, never something built by me, of course, isn’t working. So from that  standpoint, it’s great. And that process takes it does take time.  

It’s it’s not a laboratory research project, but it does take a lot of time you know, if you have the  archival data available, that’s great. The the projects where I’ve been lucky enough to work with a  corporation that has an internal HR data you know, warehouse and team that only does that  where you send them an empty spreadsheet and they send you back a pretty clean data set or set  up data sets for you to analyze? That’s amazing. Like, that that 

Speaker 2: that that has to be the best. 

Speaker 1: Yeah. It’s a unicorn, but it does happen. So hopefully that gave you some time to  think about your wackiest story. If not, let me keep talking and when it comes up, just just just let  me ask. 

Speaker 2: No. Actually, I thought it’s pretty it’s pretty wacky. I’m not gonna get into the  specifics just because of agreements I have with prior employers, but I remember when COVID  happened, a company I worked for actually had some essential workers, and our team got tasked  with trying to model, like, build, like, epidemiological models to see how much COVID would  spread throughout the teams that had to work in these facilities. And that was wild. First of all, it  was wild in the sense that our models were completely wrong.  

But it was also a while in a sense that, like, I remember there was, like, these famous p like,  researchers, I think, in Britain somewhere who had built these models of, like, how COVID was  gonna spread around the whole globe and that was, like, kinda what went into the the reason why  lockdowns were put into price. And we had kind of used those exact same techniques internally,  and it was wild just to be feel like you are part of history. 

Speaker 1: Yeah. 

Speaker 2: Yeah. Like, I don’t know. 

Speaker 1: It’s cool. 

Speaker 2: It it was it was crazy because, like, I’m not epidemiologists or epidemiologists. And  so we probably did everything wrong, but Just to make a plug here, it was probably directionally  correct, let’s say. 

Speaker 1: Yeah. There you go. Plug for your podcast. But that’s super interesting. I mean,  everybody’s I think, got a COVID work story.  

I mean, modeling that. So I’m just thinking about, okay, what variables are in the equation? Like,  do they have PPE? What PPE are they using? Like, there’s a lot of things like that.  Right? And I would we would all hope that that, you know, wearing a hazmat suit would would  keep you away from from the COVID thing. It’s really interesting. Man, when COVID started, I  I watched so I’ve never done this before, but I watched on this like it’s called like masterpieces  series or something. It’s like it’s like academics, there’s There was twenty four episodes of this  thing with this woman of several different people, but one one professor of medieval history It 

was the entire, you know, entire story of the plague in Europe.  

I mean, from beginning at twenty hours, I watch this. I can’t believe it. 

Speaker 2: You clearly have more free time than I do. 

Speaker 1: Well, I like to watch a little TV in the evenings. I mean, it took me three weeks. It’s  not like a binge watch Yeah. 

Speaker 2: Okay. You didn’t just ninja all the way. 

Speaker 1: And she was knowledgeable. And, boy, we’re lucky first of all that, you know, what  COVID was controllable enough for us because that thing wiped out about half the population in  Europe, and it lasted on and off for about a hundred years. And so, you know, you never got rid  of it. You have these waves. So, I don’t know.  

I always like a little perspective. I’m like, okay, I feel pretty lucky that we just got the COVID or  you know, I think if we’d had obviously modern medical technology back then, we’d be fine  because it’s just it’s boot it’s it’s it’s a known thing with an antidote now, but, you know, at that  time. And then there’s a Spanish flu epidemic, which everybody kind of glossed over. You know,  that was also, I think, early early nineteen hundreds that killed a lot of people. So we’re all lucky  to be on the other side of COVID.  

Did COVID change anything about doing people analytics for you at all, like, anything? 

Speaker 2: Well, in a kind of a micro sense and a macro sense, yes. In a micro sense, COVID  changed the whole game of the focus of people and analytics teams where they went through,  like, the whole like, leaving the office and return to office debate. Right? And they had to do  supporting and confirmatory and disc confirmatory research throughout to say, where are our  employees? How productive are our employees?  

If we go back to the office, what is the sentiment gonna be around our employees, how many  people are gonna quit, you know, all of that kind of modeling that goes into these type of  decision. And a lot of that modeling was actually neglected by organizational leaders, by the  way. 

Speaker 1: Of course. Of course. That’s part of what we do, isn’t it? I can’t ignore that. 

Speaker 2: Yeah. In the in the macro sense, it it thrust people analytics into the limelight. And  we were, again, just I I I don’t mean this in a negative way, but we were just a a base sick HR  function before that. It gained a lot of momentum prior to the pandemic. But during the  pandemic, you know, I know of teams that were meeting with the CEOs every week.  You know, and that that is not a typical occurrence for a people analytics team. And so it it really  got thrust into the limelight. And I’ll actually say just because the pandemic sort of waned. I think  our field of people analytics has sort of had an identity crisis of sorts since then because We were  so important for a period of time and our importance is somewhat being diminished now because  things are going back to normal. And and I think some teams are struggling with that that they,  you know, maybe it’s like a retired athlete.  

They used to be at the top of their game. And now Now they’re kinda having to go back and sit  on the couch some, and I think I think we’re struggling with that a little bit.

Speaker 1: So you’re saying people analytics is a one hit wonder or something, but but they  ignored you. It sounds like some of 

Speaker 2: them They say you have your whole life to write your first album, and then you’ve  got, like, a year to write your second album. And so we gotta write our second album. 

Speaker 1: Yeah. Well, we know it’s valuable. You know, one of the interesting things you just  brought up, and I’ve been working a lot with this as I kinda get in more to the to the AI ethics and  AI governance, which is fascinating to me and and I’m enjoying continual education on that, but  you know, one of the things you look at is psychological safety around the use of AI in the  workplace. And you mentioned, you know, monitoring productivity and I know that there’s, you  know, if there’s a poster child for what makes you feel insecure in a workplace, you know, it’s  being monitored, although Tara, Barron, and her lab have some really great research about, you  know, that that I would like to just kinda call out is that Yeah. If the company informs people and  gives them a a reason why and they they it’s not some a sneaky thing where, you know, people  feel like they’re they’re being big brother a whole lot.  

It it really tends to be more accepted, but just in general, I don’t think anybody necessarily likes  that. So how are you if you could even say? I mean, how are you guys measuring productivity?  Did you have, like, some kind of software or data collection device that was looking at what  people were doing. 

Speaker 2: So a lot of there’s a you could anything you could think of, somebody probably tried  it out. Right? And so it’s hard to say a general theme. I will say, for some of the people I were  affected out there that did this effectively is they would find functions where productivity is easy  to quantify and then they would do research beside that to show self reported productivity  correlated to actual productivity. 

Speaker 1: Right. Right. 

Speaker 2: And then they would use those measures of self report throughout the rest of the  business to say, if we can see this correlation in one part of the business where it’s easy to  quantify and we show that there’s that relationship. We could probably logically expand that to  the rest of the business. And so I think that’s how some of the companies did it. Actually have  really strong feelings on the the whole electronic monitoring employee productivity debate. And  I’ll I’ll tell somewhat of I don’t know if this is a pocketful.  

It’s just a real story. We actually had a company reach out to our podcast who does this electronic  performance monitoring stuff, like the mouse clicking, the eye tracking, all of the like, and they  wanted to sponsor us. And I was like, Well, I mean, we’ll take your money, but we’re not gonna  be nice to you. 

Speaker 1: Right. Right. 

Speaker 2: That’s right. I and so they said, oh, no. No. No. We wanna come on because we  wanna explain how we’re the good and ethical kind of doing this, and we’re not the evil, like,  dystopian, you know, or wellian type of company who’s doing this.  

And I said, that’s fine. Bring your data. I promise you. I I have a scientific orientation I will  definitely give you a fair shake. But just know, I’m predisposed to think that you’re probably 

doing some screwed up stuff here.  

And so I I I come down on this stuff very difficult. And the one the reason why I mentioned them  because you said, like, is people analytics doing this? The primary customer of this organization  was actually IT functions, not people analytics. And that’s why they were trying to get a name in  

the people analytics face by coming to our podcast because no people analytics teams were  buying their products. 

Speaker 1: Interesting. Did you change that forum? 

Speaker 2: No. We we ended up not doing it. But yeah. 

Speaker 1: Gotcha. Interesting. So one thing that, boy, this is another, like, I need a your There’s  things in future episodes I might do again inspired by you. But I need some kind of like a sign or  a sound effect that’s like it’s it’s LMM chat 

Speaker 2: It’s LMM time. 

Speaker 1: It needs to have, like, a a there’s a game show. You bet your life with you know,  gotcha marks. I don’t know if you know that or anything about that. It’s it’s from the fifties, I  think. There was a he would say say the magic word and the duck comes down, and duck would  would come down from the ceiling, and that would be like the, you know, a pee wee’s play  house, the word of the day, everybody’s excited.  

So I 

Speaker 2: I I gotta say this, man. You seem like you would be fantastic to have a beer with like  the podcast because you got a lot of interesting stuff I’d love to learn about. 

Speaker 1: Oh, yeah. Well, I’m I’m sure you do as well and, you know, we can always make that  happen for sure. You can’t do that over the over the ether, though. I I went to some beginning of  the pandemic. I went to some virtual cocktail parties with, like, my friends from college.  It was great to see them because I haven’t seen a lot of them, but, you know, it’s it’s it’s nice to be  able to hug people. Yeah. Anyway, so One of the things, you know, in looking at kind of the what  LLMs are doing to help people be more efficient. Right? There’s a whole thing of people out in  the workplace using them to get their jobs done faster, not telling other people because they  either feel guilty about it or, you know, they don’t they feel like their employer would be upset  with them because they’re getting paid.  

So I read an article where, you know, it’s just different stories from people and one guy’s like,  yeah, you know, I I work remotely I can get my all my assigned work tasks done in in three  hours. And the rest of the day, I just goof off because I’m getting everything done that I need to.  And so, you know, there’s kind of a little bit of a of a ethical dilemma there. Right? And a  dilemma for the company I think if they knew that, they would probably not be super happy with  it.  

But at the same time, the guy’s getting his work done. Right? So I feel like a lot of people would  say, hey, I can do more. So I interviewed a chat GPT for a podcast episode. I asked it a little bit  about that.  

You know, hey, is that what would you do? How how should companies handle that? And is it a  problem? And, of course, nothing that GPT does is a problem in its opinion. But It just said, well, 

that guy needs more work responsibilities.  

You know, the company needs to give him more stuff to do, but they may not know. So I guess  the point is, you know, you could take it either way. You could you could be more productive  with it or or less. But, you know, I wonder if you know, maybe that’s another reason why  employers might wanna monitor, you know, what’s going on with their employees at at work,  especially remote. Employees because maybe there is more time they could give, but but I I don’t  know.  

That maybe that was a little bit of a noncommcommuter. But these are the modern problems that  were starting to have around productivity. And if you track that guy’s productivity, I guess,  depending on how you did it. Right? Either you would see that he was logged on to his work  computer for eight hours and only three hours of stuff got done, or you wouldn’t be tracking that.  You’d ask him how productive he was, or you’d see that his objectives are met. So the data  behind that’s very different depending on what the source is. That guy could get away with not  getting exposed for a very very long time. But with some monitoring, he might be busted, you  know, right away. I don’t know. 

Speaker 2: So I I’m not even sure that some of the employee performance monitoring software  would actually catch it. What they, again, are usually doing again, if the if the person just, you  know, left their desk for hours at a time. They’ll catch that, but they wouldn’t actually be able to  diagnose what the person was doing or why they were away from their desk. That the the the  software from from what I gather just not that intelligent to be able to say, oh, this person just  wasn’t working in one way or another. I will say it does bring up a lot of interesting points on in  terms of what does it mean Like, is there like a winner’s curse for being really good at your job?  Alright? So presumably, if you’re able to automate seventy, ninety percent of your job. That’s  because you were really good at your job. And it’s very a curse that now you have to work even  harder because you were so good to automate it whereas maybe your peer group wasn’t as good  as you and was not able to automate it. Alright?  

But then there’s the other side of it, which is the organizational perspective, which is If ninety  percent of your job can be automated, is that even a job anymore? Right? And should automate  all of it for all the people who do this type of work. And and I mean, I feel like there’s a variety  of different perspectives beyond what I the two I just mentioned in terms of this debate, like,  what what should an AI be doing? What should a human be doing?  

Yeah. Do we lose our humanity? Do we lose all of our jobs? Like, what what happens here with  the outcomes of some of this? 

Speaker 1: Yeah. Well, I mean, I think it’s what PT said, maybe that person needs a promotion.  They obviously showed some good problem solving skills and some, you know, some applied  some good characteristics to figure out how to make their job more efficient. I mean, if  everybody did that and companies took advantage of it, I think there’s value that’s that’s created  there. And people that have those skills.  

Those are transferrable skills, I would imagine. Right? I think there’s classes of jobs that will be  eliminated just because they’re so repetitive and and, you know, don’t require people very much,  but there’s also kind of you know, the the creepiness factor, the caveman effect, whatever, where  there’s certain things you just don’t have the warm fuzziness of a machine. We’re gonna have  

synthetic fuzziness Yeah. Soon enough with agents and stuff.  

But I did

Speaker 2: I got a question for you, Charles. I I heard this the other day. I thought it was  fascinating. Do you know what the number one employee job prior to, like, the invention of the  automobile was? 

Speaker 1: Elevator operator? 

Speaker 2: No. It wasn’t. It was Whaler. Whaler. Yeah.  

And the so I guess or maybe it wasn’t the invention of the automobile. Maybe it was invention  of, like, extracting oil from the ground or something like that. Yeah. Yeah. 

Speaker 1: Got it. 

Speaker 2: And so they were using oil for from whales for lamps. 

Speaker 1: Yeah. 

Speaker 2: And then immediately once oil started being extracted from the ground that was  overnight those jobs went away. And I think you’re gonna see something along those lines with  some of the repetitive jobs you see now. It’s like, there’s gonna be a time remember, like, people  used to be typists. Like, that was like a thing. And like now, everybody knows how to type. 

Speaker 1: Well, you know, I I always enjoy going through onet. Right? And one of the jobs in  onet, there’s a bunch of jobs. I don’t know if they’ve take an amount that just don’t exist anymore.  But I said elevator operator because elevator operator was a job and an important job because the  logic of what floors to go to and stuff was in your head?  

Well, you know, now there’s there’s circuit boards and all kinds of things that do that. So you  might find an elevator operator in a in a place that’s trying to be super luxurious and old school,  but, you know, as a host for the elevator ride, probably heard a lot of elevator pitches. Right?  Didn’t No. What? 

Speaker 2: Nice one. 

Speaker 1: Yeah. You know about the the Peter principle. Right? People rise to the level of their  incompetence. So what does AI do for that?  

You know, we talked about people like making their jobs easier and then, you know, could they  get promoted? Is there an AI Peter principal where, you know, incompetent AIs get get lauded  and people are excited about it. I I don’t know if there’s an example of that or not, but, 

Speaker 2: oh, you see, you’re in a totally different direction than what I was thinking. I was  thinking that AIs in conjunction with human beings that actually raised the ceiling of Peters to  get higher and higher because they would be more you know, have more power at their disposal  using these AIs. 

Speaker 1: Ah, very interesting. I like that or, you know, but in an organization where there’s you  know, AI agents doing stuff. I wonder if I don’t know. There’s so many offshoots of all this stuff,  but let let’s pull it back into, you know, people analytics and our audience. So I think a good 

amount of our audience are non IOs.  

I’m always strive for that, you know, and 

Speaker 2: seven man. That’s what we tried to 

Speaker 1: Yeah. Yeah. Yeah. And, you know, sometimes, I worry, oh, we get so esoteric with  the IO stuff. But but I That’s just who I am.  

You know, I live a dual life of what not really a dual life. I’m a scientist practitioner for god’s  sake. 

Speaker 2: Yeah. So we’re we’re double agents by nature. 

Speaker 1: Yeah. Exactly. And I’ve always been very diplomatic and able to work the the middle  ground of stuff, see both sides of it, etcetera. So if you’re if you’re talking to people, you know,  that are in talent acquisition, who I would think would be our primary audience, you know,  because we’re science for hire, of course. You know, what would you say to those people about a  people analytics function, how they could you know, think about utilizing that function because  I’m sure a lot of people analytics serve as internal consultants internal center of excellence that  that move around to different business units and problems.  

So how should an internal, you know, person in hiring think about their people analytics function  or the need for one, etcetera. Right? 

Speaker 2: Yeah. I’ll say recruiters and servicers don’t just automatically come to thinking about  people analytics. But I will say that people who lead talent acquisition are Yeah. Almost always  some of the first adopters of people analytics internally because operationally to run their  functions, they need to have those analytics, whether it be as something as simple is, like, time to  fill or how long a candidate has been in stage, like, you know, all all of those kind of basic type  of measures, but even when it comes to things more sophisticated, like, how many people should  we hire next year? What is our hire what are our hiring plans and projections?  What are the projections of turnover and the gaps that’s gonna create in the organization? And  just more workforce planning, more generally, People analytics can make those talent acquisition  leaders really shine. And and so I think they are usually some of the first adopters of this type of  work. 

Speaker 1: So here’s a thought I had. If you walk the floor at HR tech and you said that all those  things that you needed, there’d be a lot of people from companies with large booths who would  say, oh, you don’t need a people analytics platform for that person for that. You need our  platform. We have all that data together, and we have all these dashboards and, you know, why  even bother hiring somebody for that? And so, you know, what would your response to that b,  you know, I could tell you what mind is, but you have more perspective problems.  Can you automate a lot of that stuff you just talked about to wear to our earlier conversation, a a  people analytics specialist wouldn’t even be needed because the dashboard would show you all  that crap. 

Speaker 2: Well, I’ll say first of all, ideally, those platforms could do what they say. But  generally speaking, they are not so plug and play that you can just turn them on and it fixes  everything. Those those tools just don’t exist. And so you need somebody who can actually run 

these tools internally. That’s problem number one.  

Problem number two And this is a big and multifaceted problem. It’s just data, data quality, data  privacy Yeah. Getting data to merge from one place to another. Mhmm. None of these tools do  that.  

You have to have human beings who can do that and improve it and make these things work.  Kinda to the story you mentioned earlier about every Blue Moon. You work with a team that has  all your data in a perfect place and they just give it to you when you need it. That took a lot of  work to get to that point. And none of the technologies just do that natively, and so you’ve got to  have a a very talented team to get to that point.  

Of maturity. And I I don’t see that going away anytime soon. 

Speaker 1: Gotcha. Here, I like to think about I I love analogies. I’ve always been good at it. So  I’d say that receiving a data set a lot of times is analogous to looking at the street after a mardi  Gras parade. It’s cluttered with all kinds of chaos.  

And to be able to use the street again. You gotta clear all a crap out of there. It’s not a perfect  analogy, but that’s what popped into my head. You know, and there’s a lot of that cleaning out. I  do I’ve been hearing and I’ve never tried this, but I’ve been hearing, you know, you can feed a  dataset to chat GTP and say clean this dataset for me.  

I’m you have to give it more parameters than that. I don’t know if I trust that, but I mean, you  know, I think where we’re we’re headed. So that’s a segue into this. And we’re, you know, we’ll  we’ll we’ll we’ll start wrapping up soon because we could go on forever with this. But there’s  kind of the next evolution of all this stuff that I’ve there’s, like, the CEO.  

I can’t remember his name of oh, man. I can’t remember the name of CEO or the company. But  he talked about so I can’t give a reference people could look at. But he talked about the future of  all this stuff moving into what he calls interactive AI, which is essentially that you have an agent  and you say, hey, I need to find an answer to this problem and the agent is running a lot of little  models underneath it. It understands how to stitch them together and it gives you an output.  Right? So my point is, like, we’re we’re headed toward that, but there’s still that fundamental  question of do you trust it? I mean, you gotta have a leap of faith. So, you know, would you trust  GPT to clean a dataset for you? Or have you ever ever tried anything like that? 

Speaker 2: Yes and no in the sense that I’m trying to I have a lot of these, like, standard, like, go  to things I say around these things because I speak on it often, and I just wrote a book that or a  white paper that was, like, fifty pages long going people analytics operating models with Jynerv  AI. Yeah. What’s the 

Speaker 1: need one of these. Every time you have a standard answer, if you have this set up,  you just hit the button. 

Speaker 2: Exactly. Just automate me. You don’t need me on the podcast. Just automate it. But  the thing I’d say is, I’m waiting on these autonomous agents.  

Right? Because that’s what they’re called and I like like, here’s the example I use. Oh, well, I’ll  talk about the data set in a second, but I want an autonomous agent to manage my calendar for 

Speaker 1: me. And

Speaker 2: when since AI has gone live, people have told me over and over again, it’s like,  there’s gonna be this AI that’s gonna manage your life for me. And I haven’t seen it yet. And so I  am waiting for the day, I can turn over my life to an autonomous agent that can do all the things  to schedule my appointments, cancel things, move it around, find things on my calendar, or give  me breaks in my day that knows me better than I know myself. I want that so badly. And and so I  think there is a little bit pipe in this space.  

So, like, oh, that’s right around the corner. That’s right around the corner. It’s like, was the corner  a year from now? Is it five years from now? Like, when is it?  

Because I wanna see these things and I wanna get past the hype. In terms of the data  management example, I would never encourage you if you’re using proprietary data to just put it  into one of these open AIs that are out there. 

Speaker 1: Right. 

Speaker 2: Like, that is 

Speaker 1: I think it’s 

Speaker 2: a bad bad idea. The thing is and this is this change is occurring quite rapidly. Is a lot  of companies are investing in closed AI systems that are accessible just to their employees. I  would say if you have that type of system, absolutely try it out, but I would measure twice cut  once because it can probably do some things really fantastically and it can probably screw up  things royally, and you’ve gotta learn the difference. 

Speaker 1: Yeah. So one, again, thoughts here. So if I had an an an AI agent that handled my  calendar, it would need to know how to make up believable excuses. For things that I didn’t  really wanna do. Right?  

So is AI good at that? That’s where you start to really get into the 

Speaker 2: Yeah. But then the excuses are gonna go to the other person’s AI agent that is gonna  to to show this, like, actually, this is BS. Charles just didn’t wanna speak with you. 

Speaker 1: Right. And do you trust it? And then, you know, the I mean, I just think of so I don’t  use an Alexa or any kind of personal assistant, you know, I I keep thinking when even when you  see the commercials, like, Alexa ordered me more diapers. You know, do you really need that?  It’s like a WiFi or refrigerator.  

I mean, do you really need a WiFi refrigerator? Come on. So a lot of that stuff to me is just kinda  neat stuff, you know, where it’s home automation. Turn on the lights or whatever. I mean, that’s  kinda it’s kinda okay.  

So so we have that nailed down and, you know, we don’t really need it, I don’t think quite  honestly. But I’m sure we will have those type of AIs that do that. It’s just a sophistication and the  capabilities they have to manage kinda less scripted, more subtle conversation. So it’s the same  exact conversation you have about using AIs to do a lot of stuff. But you’re right about the you  know, and that’s another big thing about using these things internally.  

And I would imagine something that that people analytics folks would face a lot if if they wanted  to use an LLM or their company wanted them to use one or whatever is, is that data security  issue. So, you know, I’m sure there are applications. I was just in San Fran visiting with a friend 

of mine. I’m gonna have him on the show. He doesn’t do anything like what we do.  But he’s got a startup where he’s built a platform that you can run different LLMs on that keeps  them contained. But you can use external ones. I’m learning how it all works, but there’s a need  in the market, apparently, for this. And it made sense because I immediately said, well, there’s  probably some HR technology companies that could build something like that into their product.  Right?  

So It’s it’s a lot of applications, but it’s it’s very clear that data security and privacy is a guiding  force in all AI ethics or any ethics really? 

Speaker 2: Well, if if this technology revolution is anything like the prior ones that have  happened in the last twenty years, HR will be the last one to adopt it. 

Speaker 1: Yeah. 

Speaker 2: So 

Speaker 1: Yeah. 

Speaker 2: You know, I I look forward in other places first, and then HR will get it as the last 

Speaker 1: Yeah. I I would agree with that. You know, we don’t need to even start going into that  whole thing. But, you know, people analytics transcend HR? Do people still?  Do you sit within HR? Obviously, that’s people. But, like, is there a hope that people analytics  could rise above the stigma or whatever Yeah. Above HR. 

Speaker 2: Oh, I I am very much in this camp. I’ve talked about this a lot before, is people  analytics needs to ascend in the sense that right now, when they talk about people analytics not  being a part of HR, they just mean, well, let’s put it in IT or something like that. And that’s  usually a horrendous answer. That’s that’s not usually very good. But what I mean is perhaps in  the future HR is people analytics.  

Right? And that’s all HR is. And I actually am very I think that’s an exciting future. I think that  that’s a probable future. And and the reason why I say that is because If you’re using AI, you’re  using data and you’re training models and to to automate what HR does, what function it’s HR is  most likely to be the most progressive and the most use like, that’s used to doing this type of  thing on a daily basis.  

The only one that’s even close is people analytics. Therefore, who’s gonna be driving the ship on  this transition is going to be people analytics and therefore who are gonna be the leaders of the  future. I think it’s gonna be people analytics. And so I think that not only will people analytics be  a like, usually right now, people analytics just gives recommendations to other people and they  make those decisions. In the future, I actually think people analytics will be the decision maker  themselves.  

And I think that’s an exciting future. 

Speaker 1: And that is an exciting future. And I’m doing something right now. I’m multitasking.  I’m asking chat GTP to tell me a joke about people analytics. 

Speaker 2: It probably

Speaker 1: does fine. It’s really good. No. It’s corny, usually. But and it was also over taxed.  The server was over taxed. This morning and I wasn’t able to get through. So let’s see what  happens. Tell me a joke about people analytics. Okay.  

Are you ready for this? Or Oh my gosh. That’s really a horrible joke. I’ll read it. I’m a ask for  another one.  

Why did the people analytics team get excited about their latest report? 

Speaker 2: I don’t know. You tell me. 

Speaker 1: Because they finally found out what everyone in the office was thinking, they all  wanted a coffee machine in the break room. 

Speaker 2: Wow. I’m hilarious. 

Speaker 1: Yeah. This one is that’s, like, ridiculously not even funny at all. This is the last one I  promise. Why don’t people analytics experts play hide and seek? 

Speaker 2: I don’t know. You tell me. 

Speaker 1: This one’s making me chuckle. Because because good luck hiding when they’ve  already predicted your next five moves. 

Speaker 2: I think that’s what people worry about for sure without you. I don’t think we’re that  good. 

Speaker 1: That one’s pretty good though. It does come up with some good ones. So cool. Well,  look, leave us with some parting stuff. I always open the door for for promo, and I’m gonna I’m  gonna pump your podcast again too as a template for fun and exciting educational podcasts are  actually correct.  

But beyond that, what do you wanna leave our audience with? 

Speaker 2: Yeah. 

Speaker 1: So they walk away with a smile and thinking, man, cold napper, what about it? 

Speaker 2: Well, I mean, yeah, you know, if if you’re interested in anything I said, feel free to  add me on LinkedIn. Definitely check out direction correct. But I would say, listen to Charles.  Science for hire. Yeah.  

You got he’s he’s literally and metaphorically gonna be on a rocket ship. I’m excited to see where  you’re going with this Charles and you’re obviously a really interesting and eclectic dude, so I’m  a fan. 

Speaker 1: Cool. Well, thanks. So likewise, good stuff. Well, Until next time, bid you would do. Speaker 2: Absolutely. Thanks. Thanks for having me. 

Speaker 1: As we wind down today’s episode to your listeners, I want to remind you to check  out our website rockethire dot com. And learn more about our latest line of business, which is 

auditing and advising on AI based hiring tools and talent assessment tools. Take a look at the site.  There’s a really awesome FAQs dot commit around New York City local law one forty four that  should answer all your questions about that complex and untested piece of legislation. And guess  what there’s gonna be more to come.  

So check us out. We’re here to help.