For HR leaders navigating new reporting rules

[00:00:22] Plans, Merit Cycles and the bonus that it makes, Sean runs a crap at WorldatWork, Community is Vi-Pause, The Novo Insights who helps programs come to life, Crunchy or Creamy, they litigate for days, Sleep on the ground, only one of them stays, It's Totally Rewarding Chats, Where the 10 states go, It's Sean, It's Paul, It's the rewarding time show, How long it's gonna take, there are two things Paul, how long it's gonna take for me to get used to having the intro song, but I'm glad actually in the

[00:01:05] grand scheme of things we don't have to pick one, That's right, that's one to get you in the mood, right? The other thing that's interesting is the crap thing for those that don't know will be interesting because my daughter went and watched the show yesterday outdoors and she's like, and actually I don't know, this is not great for me probably publicly, but she's like, I knew you were full of it, but I didn't know we were gonna put it in the song.

[00:01:36] And so like that's, yeah. It's the best part about your new role though, like the fact that. It is great, I will say the couple, you know, the keynotes I've had, whatever, when you throw that on and actually if you do it in a large font on the 20 foot screen, it's actually super cool. I just haven't decided if I'm gonna do the emoji to go with it. Yeah, I can appreciate that challenge for sure. I am excited for today. So this should be a good one back to pay equity and transparency.

[00:02:06] So we're also, this is the first time if anyone's watching, we've put people in time out to start the show instead of making them sit and watch us. So we'll see if we can get them back in here. Yeah, here we go. Welcome guests. Brian Levine and Helen Friedman, how are we doing? Doing well, thanks for having us. No problem. So that I don't mess it up and we start with some background. Each of you do your own bios quick.

[00:02:33] So, and Brian, you were in the top left of my screen, which means you go first out of any highly scientific method. If you can give us your background, that would be awesome. Sure. Well, glad to be with you. I'm a labor economist by training. I've been in the pay equity space for a whole lot of years. So 1998 was the first time I ran a pay equity analysis.

[00:02:59] In 2000, I had the good fortune of joining Mercer when they were building a workforce sciences group. And that's where I first met Helen. So our first work time together was at Mercer. So I was at Mercer for 23 years. I grew to lead that workforce analytics practice.

[00:03:24] I built the pay equity practice and architected the methodology there. And in 2023, I had the good fortune of reuniting with Helen and moving over to Merit and have been working on pay equity and other workforce analytics things since at Merit. And Helen? Sure, Sean.

[00:03:51] I had the pleasure of actually working with all three of you in my past. Because I've made the rounds across WTW, Hewitt, and Mercer, although longest tenure at Mercer, like Brian. About nine years, actually over nine years ago now, I think our anniversary just elapsed, we co-founded Merit Analytics to focus specifically on all things related to people analytics and planning.

[00:04:17] And we are a consulting-led organization, but we are tech-enabled. So we know that technology, I know, is very important to this discussion, right, and to everything we do these days, right? It's very much a part of how we approach things, too. But we still believe it's the strategic advice, the guidance that drives a lot of the support that we provide to organizations.

[00:04:39] My background is as a theoretical mathematician who became an actuary, so now a recovering actuary, and focused still on all things math. Recovering actuary, I like that. It was interesting, because when we posted, you know, when I said, you know, we're going to get Helen on here. I'm like, I worked with Helen. I'm like, oh, crap. I did not know that until we learned. But I am excited. I would tell people, I am excited that you were on here.

[00:05:08] When I went to Mercer as a developer, I went and I'm like, I'm just going to build this benefits database globally, the tech side. I won't lie, I was from TA. I'm like, HR is generally stupid, and you want to stay away from them. And there's no math or science. You know, it's just this human thing. And whether you're to blame or get credit, Helen, part of the reason I fell in love with the HR space and the science and the math behind it was the time I spent with you.

[00:05:36] Because you're like, actually, this is kind of super cool. And I had been, you know, if you went back prior to development was, you know, in DiviQ and whatever. I'm like, oh, my gosh, this is like a highly mathematical scientific. Like, I can get my head around this and started nerding out about it. And I have not left since then. But you are part of the, again, whether people want to give credit or blame, that I've been around for almost 25 years now. You are a big part of that. So I appreciate it. Those are some great kind. That was some great fun.

[00:06:06] I remember it was a Washington, WRG, whatever it was, right? Yeah. And I remember also that you always push the envelope to have challenged, you know, sort of the norms and the current thinking. And I think that's part of, you know, your secret sauce, if you will, of how you drive us all forward. So you have a lot to... I think mostly... There's a credit or blame for that, too. I just didn't know what the hell was going on. So I asked questions. I think that's a quick story on that front, too.

[00:06:33] Just to interject on it, Helen, you interviewed me when I was coming into Mercer in 2000. And you made quite an impression. And so I accepted that offer and then I came to work with you again. But you might not remember my first day coming to work at Mercer in 2000. I came into the elevator at the same time as you. And I greeted you, hi, Helen. And you had no idea who I was. And that was quite crushing. That was a crushing moment for me.

[00:07:02] Did she push you out on that floor? What have I made up for it since then? Perhaps. I will say the end of the love fest, too, Helen, for you was... If you remember and still am, I am the... Well, don't feel your need to add to this. Yeah. Well, as the general least professional person in any given room, at Mercer, I perpetually felt like I was swimming upstream.

[00:07:29] I remember, actually, when I went to Broadway Street in New York, one of the guys had a suit jacket for me because he knew good and well I didn't know to show up with a suit jacket. Right. But you definitely treated people with kind... Like you didn't care that I was in a Chuck Taylor's t-shirt. You were like, can the dude do math? Like that was like the bar for cool was like, can you do some maths? You're like, okay, cool. Don't care.

[00:07:53] So I am stoked that we dig in on this stuff and I'm really excited that you guys, the business you guys have built. Do you want to pile on, Paul, to see if we can turn around a record? I kind of wish I could. But like, while Helen and I shared a logo for a couple years, I don't remember actually how long precisely. I think we were both so busy in that era. I mean, we've met, but we never really worked together at all. Yeah.

[00:08:18] The thing I share the most is we, I think we share a common sort of mentor, sort of manager in Joanne Dom. And it was always the, you've got to meet with Helen. You've got to spend time with Helen. Like, so it was more the myth of Helen was sort of my experience because, I mean, you were busy. I was busy. It just never happened, right? The project that made sense for us both to be on never occurred. So less love, even though we shared a, you know, a business card. That's even better. That's the myth.

[00:08:48] That's like just a myth and a legend. I actually knew more about Paul because one of the things that you were known for is your ability to take complex topics and kind of put them in real world lingo and help clients truly understand so they could do something about it, right? And that's, to me, a very rare skill and a really important skill. So I know of you maybe more than you think. Maybe. Yeah. Well, hopefully we're not going to take that soundbite out of context.

[00:09:19] I push the record button so I can use whatever soundbites I want. That's right. Well, let's spin that out and kind of start diving down into where do you see your projects now with pay transparency and pay equity, both from what companies are doing, how you turn that into normal language and have companies deal with it, both in the U.S.

[00:09:41] And then with all the change, lack of change, lack of deadline, whatever you want to call the EUPTD, the thing that's perpetually moving. So I don't know, since you're the myth and the legend, Helen, I'll start with you. How do you want to start with that? So I believe there's a broader context to all this. And, you know, Brian's been really focused on our EU piece, so I really want him to lead that discussion.

[00:10:09] But just as context for it to say that in my experience, right, equity is about more than pay. Pay is one key component, and there's a lot of attention played there. But I'd like us to be able to also take a step back and make sure people are thinking about all the things that contribute to where you end up, right, that job you actually end up in, the career you actually have within an organization, what makes you successful in your role, right?

[00:10:34] Those things, there's a lot of data, math, right, to your point, Sean, in that work more broadly. But specific to this piece, Brian, I know you're really tight in on all of these regulations globally. I'd like to pass it to you. All right. Well, honestly, I don't know quite where to start, Sean, because your question was just so broad and covered so much.

[00:10:59] But, yeah, if we start with the EU pay transparency directive, which, you know, certainly is on a lot of minds, where companies are at is they are looking to get ahead of it, right? So there is still a lack of clarity, right? Member states haven't all put through even draft legislation, let alone legislation, to respond to that directive.

[00:11:27] However, there is a lot that will ultimately be the same, right? There are going to be some nuanced differences in terms of required timing, perhaps, definitions of pay that, you know, they're looking for. But ultimately, with regard to the reporting, for example, you're going to have to put out gender pay gaps. You're going to have to provide those by worker category.

[00:11:55] And the way the regulation is written, right, the way the directive is written, you can explain those gaps, right? And so regression, which has been used as really the core methodology for pay equity analysis, is going to be relevant to responding to the EUPTD.

[00:12:18] And so, you know, what our clients are doing, what organizations are doing on this front is generally, and where they can, in concert with their global analysis, because there's a lot that's in common, we can talk about that too, is, you know, they are looking at this country by country, particularly where their head counsel is such that they know they're going to have to be reporting.

[00:12:44] And they are running analyses, they are looking to see where there are issues, they are experimenting with worker categories, and let's talk about that too a bit, because that's kind of a really interesting area. How do you break out that data? And then where they see differences, they are looking to make proactive changes, because ultimately, and in the very near future, right, they're going to have to be reporting.

[00:13:10] And so you can consider different remediation strategies, think about different actions that you can take, different pay adjustments that you might make, so that ultimately, before you have to go out, you've fixed these issues. Yeah. I wanted to follow up specifically on the, you know, what is like work, right? That's the hardest part, I think. Like, you know, and I work with, personally, I work with more kind of smaller enterprises, typically,

[00:13:38] that only have, like, you know, maybe 30 people in a country. So like work is difficult to define. So I'd love to hear, Helen and Brian, your perspective on sort of the challenges of figuring out what like work is. You know, experimentation, I think was the word you used, Brian, around how companies are testing. You know, like, well, what's the right way to do this? But I guess I'd love to hear your perspective on what's the sort of proper way and what's the real way that, you know, folks are working their way through it.

[00:14:06] I mean, I think there's, you know, to your point, there's a proper way, right? You want to look at it in a manner that's consistent with how you pay people. And so you have jobs, and maybe that's too narrow, and maybe those aren't defined in a way that really makes sense for the EUPTD. We know that the EUPTD really is about looking more broadly, right?

[00:14:31] They want to look at people who are really equivalent as opposed to in terms of how they should be paid versus people who are in the exact same role, right? Equal valued work. And so the starting point is going to be the pay structure, right? The job architecture. And, you know, I think you guys can talk around us on job architecture.

[00:15:00] I would suggest that, you know, if you haven't, if you're not comfortable with your job architecture, that is a really important place to focus on right now. But what's really interesting about how you set the worker categories is setting narrow categories can really bite you, which is different than kind of how in a traditional pay equity analysis small categories might work out, right?

[00:15:28] If you're looking at things by job because sample sizes are small and because statistical significance was the criteria, you know, generally you wouldn't have a lot of issues by job. And in fact, maybe you're hiding those issues in terms of cutting those populations too small.

[00:15:49] But what's interesting about the EUPTD is in a small group, because the standard is not statistical significance, and we have math errors here, so let's get into the math a little bit. Because it's an absolute criteria of 5%. In a small group where you have significant variance, it's very likely you're going to go above or below 5%. Sure. And so everything is going to be marked, every area, what do you do about it?

[00:16:17] And by the way, a lot of those are going to be against men, not just against women. And so for that reason, and we've done some simulation on this, and so if you have no issue at all, you're going to have areas that flag plus or minus 5%.

[00:16:35] And so to that point, it's really important to think, not just because of the purpose of the EUPTD, thinking broadly, but just in terms of the math, looking at broader groups for this purpose is sensible. So looking at grades, for example, as opposed to jobs. And Paul, in the example that you posed where you have 30 people, it's going to be tough no matter what. But broader groupings really is going to be sensible.

[00:17:04] But do you – so one of the things I kind of get a one, if you've heard any of it, nice to hear you say people should do job architecture. It is my least favorite thing in all of HR, Brian, but it is the thing I think everyone has to have. I've been avoiding it my whole life. Yeah, yeah. So when I do what I do – I consume what's there. But do you suggest when people have those that they run them across multiple groups?

[00:17:26] Because to your point, there's – I think people in a small end size are like, I can get my head around these people being in like roles, doing like work thing. When you start to get that larger, the end size grows. But then the ability to talk about what people do and is it really like work gets harder for people to explain.

[00:17:49] So how have you seen that balancing when people set up their cohorts between end size being small enough and creating huge standard deviations over a small end size? Or like I can fix that, but I can't explain it because the end size is not too big. Yeah. I mean, look, I think it's a balance to your point. I mean, I think that's the key word. But ultimately, you have to be able to defend the worker categories that you define, right?

[00:18:15] You're going to be having conversations with worker representatives defending the definitions that you use. They have to be based in the architecture. All I'm saying is that the tendency to go narrow, which has existed for a long time in this space, is something that can easily work against you. And so, fine, look at that. You know, I think some organizations might choose to go in that direction.

[00:18:45] It might be that they have more explaining to do because they do that. But they have to be able to have the conversations. I would just suggest, we would just suggest looking more broadly to the degree that you've established grades, rely on those grades because they're already designed, right, to bring together people who are supposed to be equally valued. I like the supposed to word, but yeah. Go ahead, Helen. Looked like you had a point to add.

[00:19:10] No, I was just going to say, to this job architecture point, I mean, part of the challenge we also have is with job profiles, right, and the proliferation of job profiles, right, and the desire we still have innately within HR and within the business oftentimes to make people feel unique, right, which work against each other, right?

[00:19:31] So, to Brian's point, the grades is the grade equalizer, right, where we can say, okay, fine, you know, I can only fix so much of that, so I can use grade as sort of a backstop, if you will, for some of these differences. One thing I've been thinking about lately, I appreciate both your perspectives on this, like, the math is going to produce some, like, false flags, I would call it, right?

[00:19:55] Like, this is going to be messy, kind of no matter what, and when I think about other circumstances where the intent was one thing, but the math was messy enough, people decided to ignore it. So, for example, CEO pay ratio, like, sounded great, but then everybody pointed out, like, the math just doesn't actually sort of support what you're really after, or kind of the UK, original UK gender pay gap reporting. Like, my early experience with that was like, oh, my gosh, how can we make this look good?

[00:20:23] And then it was like, it's never going to look, like, it's always going to be weird and hard to explain, the math gets in the way, so we're just going to post it and tell people don't worry about it. Like, is this headed down that path? Like, I don't know, but I just worry a little bit that because there is math, it's going to land in a spot where we're all going to say, yeah, we did it, but it just, it's like a big sad trombone if nothing happens, because it's basically, you know, too complex to understand.

[00:20:54] Like, am I worried about the wrong things? I mean, what are your thoughts on that? It just feels like it's a risk with this. I agree. I think it was, you know, there are other unintended consequences here, I think. But the critical thing for me is that there will be all these false positives because of the way in which you have to cut this data. And, you know, what does that mean?

[00:21:18] I think companies are going to be smart enough not to put dollars into fixed problems that don't really exist, right, given that. But it just means there are going to be a whole lot of painful conversations. And ultimately, you know, I think that the way in which the directive is enforced, which is really another yet-to-be-determined thing, will be based on what actually comes out of it. So I think, you know, companies are looking to get ahead of it.

[00:21:48] But, you know, it's more follow the spirit of the directive. And, you know, look, when you're calculating those adjusted gaps, right, ensure that you're doing it in a way that ultimately you're getting to something that you feel comfortable defending versus the aggregate. I mean, it's, I think that's similar with the, you know, what you saw in the UK, right? That aggregate gap is what it is. Once you account for relevant factors, it goes to X generally much smaller. And you can explain the differential. Wow.

[00:22:18] Yeah. I mean, there's probably two things to follow up there. One, I think the UK in particular, I think it'll change because the raw pay gap. I'm sorry if anyone on the UK is on here. It's just stupid. Because you can be a good payer and be outside the gap. And you can be a horrible payer and be inside the gap. Like, it's just, to our point, it's just math. It's the wrong math. How do you think, my question though, Brian, like with these wide gaps, like how do you think it's going to play out in Italy?

[00:22:45] Because Italy to me is kind of the interesting one as a use case of like four bands. Like, you know, there's four categories. Like, okay, we'll go from being too narrow to like, here's four. How do you think that's going to play out just out of curiosity? Because that to me is the other extreme and actually, you know, maybe helps companies because you're like, there's just four. I think that, I think that, I think it'll be, it'll be more straightforward. It'll be easier to explain those differences.

[00:23:12] I think, you know, you might not get a lot out of it, but I think that broader focus just means when you get to the point of looking at adjusted gaps, you're less likely to see something that falls in excess of that 5% threshold. In the spirit of time, let's go from like... But on the broad gap, just to follow up there, you know, I think there, this notion of what Helen was getting to,

[00:23:40] if it's more than just pay, really is relevant, right? So the broad gap, kind of a fallacy of how we're going about this is that, you know, we look at the broad gap and we try to solve it by looking at narrow groups and making pay adjustments. But the bigger issue is much more, it's not so much that there's unfairness in pay within jobs or roles. It's much more that there is unfairness in terms of access to the more highly paid roles. Yeah, selection. Opportunity equity.

[00:24:10] Helen, is that where, I was going to ask earlier, like give a couple examples about where is this sort of bigger than pay, right? Selection equity is the way I frame it, but I'd love to hear your description of what that means to you. And to your point, it's both selection equity in terms of the person selecting, but it's also self-selection equity where people have, like there's a long history, right, of women selecting jobs, let's say on the tech side, right, going back to Sean, your, you know, legacy, right,

[00:24:36] of women kind of gravitating toward those project management roles which have lesser upward potential, right? They don't necessarily know when they do that that they've stunted their career opportunity, but you'll see it in the data, right, where the vertical progression typically is not going to be in that path but rather the technical expert path, the one who chooses the other route, which is more traditionally men were at the time, right? That's super interesting. I honestly had never thought about it that way.

[00:25:05] I always talk a lot about opportunity equity because the very evil side of pay equity would be we'll just never promote Helen. She'll continue to make more in this range and I'll solve some of my problems when you create opportunity equity. But it's an interesting other dynamic to opportunity equity is by default they're not giving themselves the larger picture kind of career framework, you know, opportunity equity for whatever reason. And I had never, like, that's actually a really cool point for people to think on because I honestly hadn't thought of that.

[00:25:35] Well, and you can see that in the job architecture, right? If you think about how the progression of roles within the job architecture and you start to look at, you know, how do the different genders work their way through those roles, right? You can start to see the probability that women are moving down certain paths and the probability of promotion in those paths, right? So you can kind of build out that storyline. And to Brian's point, you can do the same kind of regression modeling that you do on the pay side and pay equity. You can look at it for promotion equity.

[00:26:05] You can look at it for performance equity. We've even looked at it for a span of control equity, right? So there's a lot of things you can do to see how are we managing people, how are we developing people, and to make sure that, you know, as best we can, we can, you know, like give more opportunity across the board, right, so that we're not inadvertently creating pay equity issues for sure.

[00:26:30] But even so, right to Brian's point, we are giving the most opportunity we can to people within the organization and their potential. That's super cool. Look, I'm not going to lie. I'm going to plagiarize that later, Helen. I'll give you guys credit. But that is a really interesting point and more math, which is always more fun too. But that is super cool.

[00:26:54] I never really thought of the career progression and like an early selection or someone helping someone select early in their life. What are you choosing? And that, I mean, it does have an effect, but I had never thought about it that way. Yeah, certainly not within a company, right? Like I've always told this story. Like my wife and I are equally smart, equally educated. Actually, she's more educated than I am, but she chose to be a teacher. I chose to do something else, right?

[00:27:21] So the pay gap in our household when she was still in the workforce was, you know, she made 50 cents on the dollar. Not because anybody made a bad choice, right? She was in a union contract that was specified. Like she was treated fairly, completely. It was a choice, right? And, you know, we'll talk some other day about whether teachers should be paid what they're paid. But like that was a selection issue on her side. Yes. You know, but then there is the other side, of course, of I need to promote somebody.

[00:27:50] Who am I going to look at? You know, if I'm an engineering leader, I'm likely to look at other engineers, not necessarily the technical program managers or somebody else. Yeah. And that creates a different challenge. So, yeah, I think it's a two-sided, which is, you know, good. Absolutely. And to your point, Paul, right? The opportunity is there also to educate women, right? So that because I don't think when they make those choices, in some cases, perhaps they do know the consequence of those choices. But in a lot of cases, they don't know the long-term consequence of that short-term choice, right? Yeah.

[00:28:20] So, you know, and they might look at other women in that role, right, because certain roles have more women than men and think, oh, maybe I'll be more comfortable there. Whatever it is, right? It becomes this perpetuating thing that we have to do something to change that dynamic, right? It's like everything else, right? If we want to change the outcome, we need to change the dynamic, at which point it's happening. Once that's happened, it's much harder to fix it down the road. Yeah. No, it's true that, you know, the old 70 cents on the dollar line, which closed for a while, right?

[00:28:49] And now I think it's stagnated if my following of the news is accurate. Stagnated. Yeah. I mean, like the easy stuff was legislated away. Like the 1% or 2% of it that was real, like, okay, we're doing something we shouldn't do is kind of gone now. I mean, there's examples. Don't get me wrong. But now it's the harder remainder of that just takes a long time for that ship to turn, right? Education at a young age that we are sort of equitable in how we advance that.

[00:29:18] And then, you know, I think you made a great point about advocacy for what are the choices you may make earlier in your career that will have a 20-year impact on you if you're not thoughtful. It doesn't mean it's wrong. It just means be thoughtful about it. Yeah. Think about the CAGR on that, too. Like the value difference in that over a lifetime. Yeah. Well, and you can measure that, right? That's math, right? You can start to pinpoint that.

[00:29:46] And at the point at which it flips, that's where you need to target some of this work. Yeah. Yeah. Yeah. We do think that, you know, on the pay equity side, you know, we're more in a maintenance mode now, which is to say, you know, organizations have been focused on this. The gaps are down. They accumulate over time because decisions are made in decentralized environments and they aren't perfectly governed by formula. And then we go at it every now and then, hopefully more than annually.

[00:30:16] But we go at it and we fix the problems that develop. We really do think that the next stage of this is to address that issue of progression. And to do that, I mean, why not look at it in the same way? Why not? You know, when you're evaluating individuals for advancement, you know, why not rely on the data to indicate who are the people that maybe are ready and should be considered and allow that to enter into decision making, right?

[00:30:46] Have the managers looking at those curated lists to help drive some change on that front as well. Because we all know certain managers are more better advocates than others, right? And if we, the data is, is the great equalizer in that, right? I put that, I mean, comes to mind with merit, like giving managers discretion and for, or forcing them to stay in bounds.

[00:31:12] Do you think, you know, the pushback is always managers want to, you know, have discretion. Do you think giving them the data, like there'll be a pushback to say, actually, you should promote Helen because the data says so, even though I like Paul and whatever it is thing. Like, how do you see that playing out in practicum? Because I think there's always this, someone told me mathematically, this is what I should do. And it's, it's, it's no different than pay equity, right?

[00:31:39] You don't make a change on the pay side just based on the data alone because they're extenuating circumstances. They're not in a model that you might want to, you know, take into consideration, right? Because the model doesn't, even a great model, as we all know, does not predict every part of, you know, the equation. So, so we would want to have explanatory factors also in progression, right? But that still means that the squeaky wheel doesn't get to just say, I want my person without being able to back that up with evidence. Yeah.

[00:32:10] So you mentioned modeling. So talk about, which just kind of leads to, did we just go like 32 minutes without the two letters? So talk about how you're seeing AI, you know, from a modeling in other places where you're seeing its effect in the pay equity analysis over time. Because I'll say this in the nicest way, Brian, you've been doing this for a little while. Yeah.

[00:32:36] So like, how have you, how are you seeing it change with AI, you know, being able to do the analysis for better or worse? Yeah. I mean, look, I think AI is driving some very significant change in terms of how we do our work. So I know I'm speaking to Excel experts here, but I'm terrible with Excel, but now I can do incredible things in Excel very, very quickly.

[00:33:02] So, you know, from the perspective of using tools, productively coding, it's incredibly helpful for that part of our work. Often for initial investigation of issues, you know, here's some data, you know, how might you look at this to get some insight on X, Y, and Z? I think it's helpful for pay equity. I think it's maybe less helpful.

[00:33:32] And, and here's why, because I think a lot of it is fairly well automated. Okay. So we take in data of a particular form and we do this year on year. We run a set of models. And then we move to, based on those models, address where there are issues, look to see where there are issues, address those issues. And we plug in at certain points, right? We plug in when we process that data and make sure it looks sensible.

[00:33:58] We plug in when we run those models and make sure that the explanatory power of those models is sufficient, that the factors are coming in the way we expect that, you know, performance is a factor that we're accounting for, that it drives up pay, right? It doesn't drive down pay. Sometimes you see weird things in that regard. You know, what I would say is that the process is automated such that we have to look at it at these different steps.

[00:34:23] So I'm not sure the extent to which once the automation has occurred, you know, AI is additively helpful for people that really, you know, know what they're doing in this space. Now, if you have an individual who seems to be out of alignment and you want to say, why is that individual out of alignment? I think you can get a lot of insight out of an AI tool. But, you know, I think there's AI and there's automation and they're not quite the same. Yeah. That's how I would respond. That was going to be my big point. I'm glad you said it, right?

[00:34:53] Like most AI now that's sort of making people's heads blow up is generative and it's predicting sort of text. I do think the big game changer for all of the math was when it was generating code, right? So, but then you're just back to where we could have gone before, right? When Helen and I overlap, people were coding, you know, in our scripts, you know, modeling out statistics. Same thing is happening now, essentially. Right? Right. So maybe the gen AI helps you code it a little quicker. But the underlying math and models is there.

[00:35:22] So if you've got the math already, AI doesn't really advance much because it's deterministic math that has been built and it's already there. So it's – And you don't want a whole different methodology applied period after period in this space, right? You need it to be done consistently. And, you know, of course, there's the legal application of all of this. And so, you know, having it done in that specific way really matters. Yeah.

[00:35:47] I'll just close on that and say I actually think the AI piece is probably more helpful in the job architecture, taking multiple jobs that are the same things and saying this is the same crap, put it in the same bucket. And then I actually – I'll add now I'll pile on. I think doing some of the career pathing and where did everyone go and kind of what were those things.

[00:36:09] So they're kind of the things that, to your point, we could have done prior but you had to choose which ones you did because the consumption of time wasn't there. It can do it. But being a little cognizant of time, if you could tell people two things to focus on, one or two things, like as they start to think about going through a pay equity process, what are the kind of one or two highlights? And I don't know which one he wants to go first. It's always easier to go first.

[00:36:35] You know, one or two things that you would say make sure you focus on this to get it right. Well, I'll just start with one and then, Helen, I'll let you take the second. You know, there is fixing the problem, right? We see where the gap is. We consider individuals who maybe can get adjustments and process those adjustments or review those adjustments, ultimately process and test. But there's also root cause, right?

[00:37:04] What are the factors that create those issues in the first place? And some of this is just obvious and some of it the data can reveal, right? Right. So often when we're doing this work, we also look to see where in the pay determination process do gaps develop. So, for example, does it occur through hiring?

[00:37:31] So when we look at new hires, for example, are women and men coming in at different rates? You can see that in the data. It's almost universal that that's true and that drives gaps up period after period. It's also the case that we see sometimes with off cycle adjustments that they are contributing to those kind of issues. Sometimes it relates to promotion. Sometimes it relates to performance.

[00:37:54] Maybe we have fairness in terms of how those performance are allocated, but the way in which performance is rewarded differs by gender or differs by race. So, I mean, look, doing this analysis period after period, you often get questions from executives. I'm confused. We did this analysis already. We did it last year. The gap is back. What's going on?

[00:38:17] Being able to show why it's being created and trying to address some of those issues in terms of practice really is – and you're not going to get ahead of all of it, but that really is, I think, going to be engaging to people and showing that they can understand what's happening and it's additional value you can get from the work. Yeah, and I also believe that AI plays a role after you've done the modeling, right?

[00:38:46] Like you have a wealth of information about individuals that are the outliers, right, that you could potentially access and leverage AI to say, you know, what is it about this person that makes them different, right, to the extent that, you know, I think that there's opportunity there that we haven't really fully explored or exploited, if you will. And I think over time maybe that also, you know, because, you know, what we're talking about, Brian, right, is the sort of macro, how do we solve the problem?

[00:39:13] This is sort of the micro, how do we look at each of those outliers and think about what we can do to help their case. That's awesome. So we are going to circle way back before I ask the last question because it was such a love fest at the beginning over Helen that you guys almost got out of doing the speed round and we can't, like I'm consistent of having the speed round of questions. And I noticed you guys didn't highlight that.

[00:39:42] So I'll go just, you know, Helen, Brian, Helen, Brian, just because now it moved you, Helen, to the top left. So are you tea or coffee, Helen? Tea. Okay. And Brian? Coffee. And we always follow up on the coffee. Are you a coffee snob or just coffee? I'm very picky about my K-cups. That's snob.

[00:40:09] Don, you don't want to say it, but we're just going to put you in the snob bucket for that. But picky about K-cups is a very different form of snobbery, Sean, than our form of snobbery. It's not a soy latte. Yeah. Or like I want this particular sort of region's beans roasted a certain way, right? Which is more where Sean comes from. So like that's a unique. That's a different. That is a unique. And not definitely like lipped in as basic as they can come. So not snob. Like that is definitely on the not snob spectrum for tea.

[00:40:39] Your sports team of choice, Helen? Oh, this is a sad one. It is today. Okay. Oh. No, no, not even. Okay. I'm not even that good. New York Mets all the way. Sorry. Yeah. Yeah. Okay. Brian? They're really good with their fans because they really don't win much. Yeah. Right. Well, easy for me to say the next these days. Oh, yeah. That's a good. Now probably the most important one as we go through here. Helen, you have a weekend away.

[00:41:11] You get to choose by yourself. No other human influences. Are you going to a posh resort? Are you going to go rough it and go camping? Oh, well, this is an easy one for me. I do not camp. I know. I know. I know. This was hard. It was one of the reasons I almost didn't invite you. I knew 100% where this was going. So quickly, we'll just pivot to Brian because that's. Can I do something in between?

[00:41:38] So, you know, not quite camping, but in a cool space, maybe doing some hiking. And then come back to a nice warm place where you sleep off the ground? Sure. Yes. Okay. I'm in. Glamping. Glamping. Helen, best place to get a New York slice of pizza? Are there a lot of great places to get slices of pizza? Explore them as many as you can and pick your own favorite.

[00:42:07] I'm a firm believer. And I don't go more than five blocks for a good slice of pizza is my rule. Which is so funny for me because how long does it take you to walk five blocks? Maybe 10 minutes. Yeah. I can't walk to another house in 10 minutes. It's so funny that you have choices of pizza, like oodles of choices of pizza. Oodles. And I'm like, I have to go find Ann Neighbor. Yeah. Brian? Yeah, I'm kind of the same.

[00:42:35] New York pizza is pretty much great everywhere. Okay. Crunchy or creamy peanut butter, Helen? Oh, got to be crunchy. Okay. Brian? Oh, creamy. And I eat PB&J pretty much every day. And it's quick now to let him buy the brand names again. For a while we were store brand over there. Oh, okay. I should have. Yeah. Wow.

[00:43:03] So you moved up to creamy. What do you do if you have a company meeting? That's now the problem. That's the problem. They eat pizza instead of peanut butter. Yeah, they eat pizza. That's what I'm hearing. That's what I'm hearing. So we end with, and I'll just mix it up. I'll start with you, Brian. If you could solve any one thing automagically in HR, anything, what would you solve? So I guess I'm going to, I'll answer this as somebody who's in the data maybe too much.

[00:43:33] I'd like the data to be better. So there are a whole lot of things that we talk about having access to, and we've talked about wanting access to going back decades, and we still don't have it. So, uh, actual experience, uh, code and education. Um, I mean, that might be something that AI can help us with kind of going to the applicant tracking system and looking at some of that data that is, um, you know, messier and turning it into something that's useful.

[00:44:00] But I think that access to better data can, can really help us in a lot of different ways. That's an awesome one. I won't lie. The maths that we can do are super cool, but if you plug the wrong numbers into the calculator, it just doesn't work. Um, what you got Helen? I was actually on the other end of that spectrum that we get HR. HR has become much more evidence-based than it was years, many years ago, but I still think

[00:44:26] we have a lot of opportunity to make more of what we do in HR evidence-based, right? That lead with evidence. And I think that would be a big sort of still to this day, transformative thing for HR. Yes. Make it provable. Let's actually test some things. I'm with you. I wrote a whole chapter in a book about it because I just can't stand out. Well, I'll add to that provable, but also, um, my pet peeve is we haven't gone back in, in theory. And a lot of times we're going to go back and check our hypothesis.

[00:44:57] Merit is the one we do a lot. We're like, this is how we're going to do a merit matrix because it's the right thing. And we never mostly because merit sucked and no one wants to talk about it for 10 or 11 months, but they never went back and said, did that have the effect on our people the way we thought it was going to have and be evidence-based to your point, Helen, to be like, and then have, you know, the wherewithal to change to be like, we thought this was good. That was fine. But that's not how it worked out. Let's adjust based on the evidence as opposed to the field.

[00:45:25] And we never do cohort analysis, right? To your point, like if all the group changes at the same time, where are we really analyzing, right? We need cohort analysis up front to say, take two like groups, do two different things. What happens, right? It's not a common thing to see in HR. We can do that now. The maths, you know, cause you can break the maths out. This has been great. I'm so excited. Sorry, Brian. I am excited to also talk to you. I don't want to make it sound like I wasn't, but hang out with Helen is super cool.

[00:45:55] People should check out Merit Analytics, some of the stuff you guys put out. Obviously, you know, I love the work you guys do. It's so math focused and evidence-based. So make sure people follow. We'll tag everybody, of course. And I guess I'll say it the usual, like follow us, tag us, whatever it is. That's right. Listen thing. I should write that down at some point. You'll figure it out at some point, Sean. And then as our new close beforehand, we asked who's going to pick sort of the genre of the exit music.

[00:46:26] And Helen, Brian said Helen should choose. And Helen's choice is that Brian gets to choose. So it works a lot like, I think my kids would do that to me. They kind of reflect it back. So Brian, since your name starts with a B, I'm going to give you four choices of music genres that all start with B. Broadway, boy band, bluegrass, or barbershop. Um, let's go with bluegrass. Bluegrass. Love it.

[00:46:56] All right. Thank you, everyone, for joining Brian and Helen. Thank you. Here we go. Bluegrass. One mission and a firm is full of takes. Comp plans, marriage cycles, and the bonus that it makes. Sean runs a crappy world at work. Community, his vibe. Paul's a nova practitioner who helps programs come to life. Crunchy or creamy, they'll litigate for days. Sweep on the ground.

[00:47:41] It's the rewarding comp show.