What’s Holding Back People Analytics? A Conversation With Josh Bersin and Paul Rubenstein

What’s holding back people analytics?

In this special guest episode, originally featured on The Josh Bersin Podcast, Josh Bersin and Paul Rubenstein, Chief Customer Officer at Visier, join forces to dive into the rapidly evolving world of people analytics.

Josh and Paul share their unique perspectives on how high-performing organizations are leveraging AI, predictive analytics, and systemic data to solve business challenges and achieve measurable outcomes. They also discuss the obstacles HR leaders face in adopting analytics, the transformative role of AI in decision-making, and how systemic HR is shaping the future of work.

This episode includes insights from the 2024 People Analytics Best Practices Survey, exploring maturity models, best practices, and the business impact of effective analytics strategies.

Josh (00:00:05 -> 00:00:38)

All right, everybody. Today I'm very excited to have Paul Rubenstein with me, the Chief Customer Officer of Visier, one of the leaders, if not the leader in people analytics. We can debate that. I know Paul's opinion and Paul's gonna tell us about some stuff going on in people analytics and some research that we're just about to launch that you may have read by the time you listen to this about the maturity model and the evolution of people analytics. Paul, tell us a little bit about your job and just quickly, your work experience. 'cause you've had a lot of interesting career experience leading to this. Thanks,

Paul (00:00:38 -> 00:01:14)

Josh. Well, first of all, hi, happy Friday afternoon or whenever this is being recorded. Yeah, it's been a crazy road for me. I, I worked at, uh, two of the large consulting firms that specialized in human capital. You know, I've always consulted to the HR function. You know, I started out doing those eight big HR transformation projects, but then later on got into, uh, much more how do we understand talent strategy? Like how do you actually write a talent strategy that's connected to business strategy. I ran a leadership practice. I ran a bunch of IO psychologists who were doing assessments in leadership who reminded me every day that I wasn't a doctor. It was, you know, I've, I've had some really good stuff.

Josh (00:01:14 -> 00:01:16)

I, I've had that feedback too.

Paul (00:01:16 -> 00:02:24)

But I would say there was a pivotal moment where I was sort of obsessed with, how do you actually take, like what the CHRO or the business wants to do around talent and get lots of distributed decisions aligned to a collective outcome, right? How do you write a talent strategy that's granular and specific enough to make better people decisions? And I, at the same time, I found myself on my journey, like learning all about people analytics in the early days. And I met the Visier team and as I was working on how you write talent strategy, the question was, are you measuring the execution of it? Okay. We all tend to find that our organization is out of shape, right out of the shape that we want it to be in from cost or, or skills or geography or, or spans and layers, whatever it is too late. So how do we start to interrupt that? And I was like, man, if only we could get all the data across all the different parts of HR and get HR better to be better with data and make it easy. 'cause it's always been hard, disaggregated systems, et cetera. Anyway, I was on this journey, uh, in the early days. Remember big data, um, sure, yeah.

Josh (00:02:24 -> 00:02:27)

<laugh>. Yeah. I wrote a lot of papers on it. <laugh>. Exactly.

Paul (00:02:27 -> 00:04:47)

And, and during that time, I met a company that was taking a different approach. 'cause I had been involved in those big data warehousing projects, you know, that took forever and, and, you know, building things with hand-built tools. And man, when I saw Visier and I was like, wow, they didn't start with the data. They started with the questions. I looked at it and they, they had some technological innovations that made it easy and beautiful. And I looked at this and I said, for the first time, this was an investment in HR that wasn't about the efficiency of hr, which is how HR has always spent money, but it was about elevating the pattern recognition and storytelling capability of hr. So that it was actually insights to me the first time this was about recommending to HR spending money on tech that wasn't labor arbitrage.

And quite frankly, I saw this as the beginning of a change in the market that said, okay, record keeping. You've reached your limit in the return on investment. Now it's time to tell stories with data and help everyone make better decisions. And quite frankly, I looked at this and I said, this doesn't even belong in hr. It belongs to the people managers. It belongs to, you know, outside of hr. It had tremendous value. Anyway, Josh, long story short, I fell in love with the company so much. I quit my job. I joined first in what they called value engineering.

Obsessed with the moment people see new data, do they make a better decision? How should HR think about investing differently than they used to? Not thinking of total cost of ownership, but thinking about business impact through some strange set of events. I spent five years as head of hr, which was, you know, my sort of person non-commercial at at Visier. Yeah, right? Which was, you know, learn to practice what you preach, experimentation, but also still helping a lot of customers. And then a year and a half ago, I think it's been, I went back to run, um, what we call customer success here, which is all of our consulting, all of our support, all of our making the software actually deliver value to our customers. So it's been a really privileged journey to be inside people analytics from the moment 10, 12 years ago when it was just really coming to scale and getting to operate in a living laboratory and seeing it mature. Well,

Josh (00:04:47 -> 00:05:11)

Thank you, Paul, because I've been involved in this space for a similar long time, but I'm still seeing from the research we just finished with you guys, a fairly low level of maturity in a lot of companies in learning how to use data to make better people decisions and business decisions. What, and you and I have talked about this before, why don't you just share with people why do you think it's still such a challenge and what are companies going through here? Well,

Paul (00:05:11 -> 00:05:30)

I, I, I think there are a couple of things. Uh, first of all, how important is it? Like Josh, it really starts with that. If it's important to the head of HR or somebody in the business to have good insights, not just like a report or a birthday list or a, or, you know, like use insights in a meaningful way to,

Josh (00:05:30 -> 00:05:35)

But do you really think there's companies where it's not important? Or do they not understand what important means?

Paul (00:05:36 -> 00:06:57)

I think it's a quite, I think there's a couple of factors in there. Important to who, right? If you think about the HR function, it may not be important to them because it may not be what they've been historically good at in some HR functions. It's like a lot of people didn't say, oh, I didn't get into HR to do math. You know, math in business is inevitable. And what's funny is nature of horrors a vacuum. I think even five, 10 years ago, there was great people analytics being done outside of hr. Any good general manager would take that HR data that they would scrape in a spreadsheet and they'd mash it up with their production data or their financial data, et cetera, to do some sort of productivity or workforce optimization, right?

Not great, you know, from a security or scalability perspective, but the people who want good answers will get them, period. Look, hr, it's a question of priorities and how much of the money is consumed and a little bit of learn helplessness, right? I think part of it is the zeitgeist, I dunno if this is the right word, around HR data or, or their relationship. For anybody who's in hr, you, you've experienced this, you get a question from the business, you go to your comp person or your HRIS person, you get a report because you wanna talk about a trend. You go in front of that business leader, and the first thing they do, if they don't like the data, if they don't like the trend, they say, I question the data.

Josh (00:06:57 -> 00:06:59)

I don't believe you. Yeah. That thing is wrong, right? I don't believe you.

Paul (00:06:59 -> 00:07:32)

And then, you know, you have this whole going back and forth to, you know, to shepherd the data and prove it out. And quite frankly, by the time you can convince them that your data is better than their assumption, the moment of impact has passed. Okay. I'll tell you who doesn't put up with that crap. The CFO, nobody from fp and a or the CFO shows up with a trend line and the business leader says, I think you have all the general ledger, uh, information wrong. You know, so I'm not going to take your observation on cutting costs. Come on, Josh. That doesn't fly.

Josh (00:07:32 -> 00:07:51)

You know, there's another, there's another dimension I Paul, you and I've talked about, which is that some HR functions define themselves as service delivery functions, not strategic consulting functions. So they don't try hard enough to correlate the, the data to the business. What you, we just talked about that. What's your perspective on that?

Paul (00:07:51 -> 00:08:25)

I think, obviously I, I believe it's changing. I, you know, the leading functions do do that. But, you know, you think about the role of the HR business partner, I actually think that the great ones are good at pattern recognition. Mm-Hmm. <affirmative>, the great ones are good at seeing around corners. They're actually naturally good at people analytics. They just don't know it. So I think there's a lot of hope. And I think on the whole, the organizations, I mean, it's where the study starts to pan out, right? The people who focus on, on data-driven decisions, the people who invest in people analytics are outperforming those who don't.

Josh (00:08:25 -> 00:08:41)

Let me get to another topic. Sure. One of the objections you hear all the time from companies about analytics is, well, we just put in workday, we just put in SAP, we just put in Oracle. We can't really do that kind of work till we get all the data cleaned up. Right? Is it fair to say there's no excuse for that anymore.

Paul (00:08:41 -> 00:09:25)

There, there isn't. It is. Your data will never be perfect. Anybody who's looking for zero noise data doesn't live in the real world. I mean, that's just the reality of it. And by the way, HR data, by its very nature, is not, it doesn't have FSB, it doesn't have gap, right? It will never be as clean. You're, you, you live in a world of retroactive restatements, et cetera. Oh my God. Have you ever seen recruiters record, uh, a TS data, somebody who was interviewed and hired on the same day? That not doesn't reflect reality. You have to understand, um, you know, how to interpret the data, not fight it for this, you know, ideal perfection. You can still use it to set a compass and hold people accountable for the decisions they make.

Josh (00:09:25 -> 00:09:44)

Well, you know, given, given that maybe it is much, much easier to get access to the data, especially if you have visier, one of the other symptoms of the challenge that I've seen is the people analytics group. Sometimes it reports to it, sometimes it reports to hr. it, let's talk about systemic HR and how you think companies should organize this whole initiative. Well,

Paul (00:09:44 -> 00:14:27)

I think before we go there, judge, actually, yeah. Can we zoom out on that? 'cause it's really interesting where, where you were going there, I we're seeing a change. If you roll back the time machine, you know, 10 years, the first people analytics functions, I think were two things. They were science projects, right? We all learned about the legend of Google and answer factories, Hey, we're gonna come to you with a question. Give us an answer. Okay? Mm-Hmm. And a lot of those were populated by builders, people who spent 80% of their time cleaning data, constructing data, and, you know, creating something rather than interpreting it, applying it, evangelizing it, right?

Because the insights, how much do you want to pay for the time to create the insights versus disseminate, explore all the, all the good stuff, make decisions, applying them, right? Yeah. And like I watched people hire data scientists who would spend 80% of their time just cleaning data. And so I think a lot of that for people who have used modern analytics applications, they've changed the labor distribution. And they've gone from just looking for functional efficiency as an answer factory to two different graduations in what I'll call the value curve of getting a workforce edge through AI and analytics.

The first one is impacting talent. So actually, so what do I mean by functional efficiency? Why don't I start there? These are companies who are like, Hey man, everybody, we got a, we got a single source of truth. And everybody can serve their own insights either by going in and getting them, or they're actively getting pushed or trigger. What do I mean by that? Nobody asks for the p and l each month. hr, those companies who are able to have functional efficiency in the delivery of those insights. They're pumping out meaningful content each month, whether people want it or not, just the way finance does so that people can hold a mirror up to the decisions they've made to that point and redirect, um, based on that, um, on where they're going in the future. Or they're using, you know, machine learning and AI to say, Hey, a certain threshold has been achieved, uh, and you should pay attention to X, Y, and Z. So they're serving up insights in the surface of work, be it through teams or email or copilot. Now, whatever it might be, they're helping managers find the signal in the noise of that daily work to know what human capital issue to focus on. So that's what I mean by functional efficiency.

Yep. And I'll tell you, all of visier customers quickly get to that, but the next group graduates to really impacting talent outcomes. And what are those people doing? They're not an answer factory. They have a point of view on the content that people should be looking at. And it's really looking at, it could be something as simple as turnover or, uh, performance risk or exit risk, or whatever it might be. And they're taking those basics in their distributing beyond hr. Those are people who recognize that talent decisions aren't made inside of hr. They're made outside of hr, and they're using the data to do that. And even within those people who are impacting talent outcomes, the more advanced ones, not only the distributing beyond hr, but they're using some light performance data around some sort of financial indicator or financial metric, uh, revenue, high level ones to understand people performance.

But the third group, the third group has fallen in love to use your words with the problems outside of hr. They're looking to understand how do I build an effective bank branch, and what should a bank branch manager understand? How do I make a luxury watch sales store? Right. You know, how do I, like, there are all these examples in our customer base where they're taking the data of work and they're blending it with the data of people, and they're no longer concerned just with how work impacts people, engagement, things like that. But how people impact work, how does attendance distance from the office, um, how your manager stability rate went, what we invested in learning in, in your learning impact, same store sales, team profitability, whatever that is, right? And I would say about 25% of our customers have achieved that, um, that final tier. And more and more are moving up. Um, as they do that, we have a customer that is loading their entire, all their general ledger data into the visier environment because they wanna search for the connections between people and financial outcomes in a more deep way. And that's a whole new world.

Josh (00:14:27 -> 00:14:31)

People, I, I think that's gonna become common with ai. I mean, that's actually a really good point.

Paul (00:14:31 -> 00:14:38)

Oh, yeah. But Josh, the question goes back to you like, do do a lot, do you think a lot of HR people feel they have permission to play in that space? No,

Josh (00:14:38 -> 00:15:14)

I think there's a little bit of a lack of imagination amongst HR people that they could do this because maybe they have been frustrated by the lack of data consistency, the v the various platforms and the tools they've been given. They've been given a copy of Tableau and Excel and said, go do analysis and say, well, this is gonna take me six months to do it, so I'm not gonna do anything very difficult. I mean, uh, this gets back to my next question. My perspective on this is that the AI based analytics, like what you guys are doing in others is so advanced that we have to ratchet up our level of imagination as to what we can look at.

Paul (00:15:14 -> 00:15:42)

Every, every people analytics leader I talk to in our customer base, and especially advanced ones, they're like, yeah, my job's gonna change. The function's gonna change. AI is changing it, it changes it in so many ways. I'll give you the easy ones first. And then we're seeing this with the early customers of the, and you know, this is not an ad, but if anyone hasn't seen it, please just call us. And you, you shouldn't need a decoder ring to understand HR data, you know, the secret decoder ring. You should just be able to go into a system and say, you gotta

Josh (00:15:42 -> 00:15:47)

Go through the metadata book and read the whole book first before you know what the data all means. Right?

Paul (00:15:47 -> 00:16:19)

And it's funny because HR people often have a skewed view on this because the language that they use every day, people outside of HR only use once in a while. So, hey, what is my turnover? What is my turnover? Who am I, what's the context of, of how I look at the organization? What am I privileged enough to see based on my security? What is that turnover? Break that down for me. Well, of that turnover, who is at risk of exit? Think of that back and forth that used to happen with a people analytics analyst that would've been two weeks of back and forth.

Josh (00:16:19 -> 00:16:33)

You know, exactly. I mean, I think the bigger scenario is the one you mentioned just a minute ago, we're losing money in this business unit. What could be the contributors to that that might have to do with overtime or errors or whatever? Which

Paul (00:16:33 -> 00:17:33)

Employees, am I getting the best bang for the buck on the dollars we spend? Right? Right. You know, there's a common language, but what's interesting is it's appreciative inquiry, right? It's built for the way and talked about collapsing speed to impact for hr and extending hrs reach so that everyone makes better decisions. AI allows the HR to change its delivery model in new ways and extend its reach, especially to the people who actually don't want to talk to hr. So the first one is this notion of inquiry. The second one, seeing the unseen anomaly detection, things like that. You look at a graph and chart, it might look fine, but how do you very quickly understand what is aberrant underneath it? Where should you spend your time and attention? The third is, who should I talk to when something is going wrong? Knowing who is responsible, who is the right person? And AI alerting you. And then of course, there's the entire authentic discussion, digital twins, et cetera, which, and this is the real irony for hr,

Josh (00:17:34 -> 00:17:40)

Why wouldn't we just wake up every day and say, here's some anomalies you should look into that I discovered on your behalf

Paul (00:17:40 -> 00:17:53)

That's what's happening. So, so you and I like, okay, so you and I did this seven, eight months ago. Yeah. Right? Remember? And we goofed together, Hey, here is, and we prototyped it and it works

Josh (00:17:53 -> 00:17:54)

Connected. It was like,

Paul (00:17:55 -> 00:18:08)

Hey, this person's at, this person's at risk of leaving or, or this person left. So these other people, here's the contagion effect. Right? Right. Here's how, and then it took it all the way through to how do you have a stay conversation with that person? That's the future, right?

Josh (00:18:08 -> 00:18:44)

Right. Exactly. No, that, that's the, the v Galileo integration. We've been doing a couple more quick questions, please. One, one of the, one of the tricky ones for you guys, I wanna, I wanna give you a chance to talk about this Workday, SAP SuccessFactor Oracle claim to be able to do what you guys do out of the box. They all have people analytics modules, they're building AI under the covers. I kind of know the answer to this question, but I want you to say, <laugh>, what, what, what, what do you do about those guys? Especially when the customer just spent hundreds of thousands of dollars on some tool that came with the HCM system?

Paul (00:18:44 -> 00:18:53)

Yeah. And, and, and the way a lot of those vendors sell some of the analytics is, um, we'll bundle it in for free. You won't even see it, and it will say it's good enough. You know,

Josh (00:18:53 -> 00:18:56)

Lay it on the line, Paul, this is your chance. Yeah.

Paul (00:18:56 -> 00:20:40)

Like, here, here it is. I don't know how to, I don't know how to say it politely. The 10, 14 year headstart that Visier has and the vision of where it's going and the ai, okay, let's break it down in a couple things. Number one, none of those customers are even close to doing what we do with ai. We are very blessed because what we have is not just an incredible data model, but we ha we are able to connect across all the customers and leverage, um, the understanding of the patterns of data that make the model work. So the training, the level of quality of answers and training and understanding that they're gonna have to do on a single tenant for any of those other vendors, right? It'll take years for them to get to where we are.

The second is this open platform. We're open to extend it beyond just the vision of what HR could do. Not solving hrs record keeping problem alone, but solving the problem of work. It's a philosophy. Do you want go with a vendor that is there that was built to just save you costs and make it easy to do record keeping? Or do you want to go with a vendor whose philosophy is, how are you gonna get a workforce AI edge? How are you gonna get a competitive edge by leveraging ai, by understanding your workforce? And remember, it's not just about using AI for fancy reporting, which is what a lot of the analytics that are coming out of the, the boxes we're talking about. Using it to understand the workforce, understand work that's a bigger vision than those platforms have. And then plan for the future and budget for the future. And make sure that you're,

Josh (00:20:40 -> 00:21:00)

Well, you could almost argue, I mean, Paul, you could almost argue that the HCM platform is more of the commodity really than the analytics platform. Because the reason you buy technology is to make better decisions and to automate things. So I've had customers tell me before, we're gonna replace all our HR systems so we can get better data. And that's kind of a backwards,

Paul (00:21:01 -> 00:21:17)

So what's what's book if all those vendors were amazing at this than the top brands at visier who have all those vendors wouldn't be buying us. Right? You know? Right. Like, I get it. Don't listen to your salespeople talk to customers. Is is my advice to everyone. But again, I don't wanna make this a commercial for

Josh (00:21:17 -> 00:21:20)

Visier alone. No. I just, I just wanna give you a chance to talk about it for a minute.

Paul (00:21:20 -> 00:23:32)

But I think there's a bigger point here, Josh, which is what is going to be the tech that you spend that's differentiating versus commodity? Which are you gonna spend as the head of HR or the head of it on this stuff that is about managing cost and compliance and risk versus unearthing the insights that drive business, right? That competitive edge, that workforce edge that you need from those insights and that, and, and the ai, I respect all those vendors.

They are adding features that are good enough for minimum viable products as quickly as possible. 'cause they have a financial incentive to squeeze out every other innovative vendor in the world, because a lot of software vendors aren't growing new logos. They have to just expand within the footprint they have, and they're desperate to retain it. I respect that. But sameness does not yield greatness in your tech stack. You have to leave room for innovation and agility. We saw this cycle a long time ago. If you get locked into one vendor and they aren't innovating fast enough, you get left behind and HR gets left behind the needs of the business. That tech stack agility becomes really important.

The second thing is, all of those big HCM vendors were supposed to drive down your total cost of ownership over time. I wrote those business cases when I was a consultant, and that is, that's not actually proved to be true. The staff you have to run it has gone up in price and complexity and the teams have gotten bigger over time, et cetera. What's beautiful about the type of software that visier is, and Visier itself is it allows you platform agility. It allows you to turn around and say, Hey, I only need cheap and cheerful for my transaction systems, right? My, I need to rebalance my portfolio to spend more on the system systems work. And by the way, when I acquire a new company, I don't wanna have to change or integrate. I can leave them on their own cheap and cheerful systems. As long as the layer, these insight letters, visier becomes like the swan looking above the above the lake. Because people want consistency in their trends in data. People want to not have to restate things just because you move to a new system underneath you. Can

Josh (00:23:32 -> 00:23:36)

We see your hands making the swan move across the screen? Wait

Paul (00:23:36 -> 00:23:38)

A second. This is not a visual podcast. <laugh>.

Josh (00:23:39 -> 00:24:17)

All right. I, I, I got you. One more question. Uh, so one of the other challenges, of course, is that within hr we have all these silos. We have the TA group, we've got the l and d group, we've got the comp group, we've got the employee experience group, we've got the DEI group. They're all running their own reports, doing their own metrics, trying to figure out the impact of their thing. You know about systemic hr, I think, and we've talked about this, that that people analytics effectiveness also essentially requires a systemic look at the way the analytics is done. What's your experience and sort of perspectives on, on how to operate HR to really take advantage of these tools and opportunities?

Paul (00:24:17 -> 00:25:46)

HR in the pursuit of specialization on economies of scale over rotated to deep functional expertise. And each of those functions had to have systems specialized just for the record keeping and administration of that function. The problems of business and people together changing work and creating productivity and competitive edge is not a comp problem. It's not a learning problem, it's not a, you know, performance management system problem. It's not any of those individual problems. It happens across those. And I know I came from a world where we looked to find out who was an HR generalist and eliminate that job because it was too broad. But we could use some more general manager thinking and we could use more capacity for that in HR functions. There is always room for deep expertise, but HR leaders have to make sure that they are showing up not as HR people first, but as business people first and HR people second. And that's where that new HR business partner and CHRO exists. You have to be a business person first and HR person second, and that's the future. And so a good general manager embodies systemic HR because they understand all the different DI disciplines are like a chef with the misin plus they get to pick the different elements to create a, a magical dish. It's the combination and the recipe that makes it delicious.

Josh (00:25:47 -> 00:26:12)

Paul, this has been fantastic. Thank you so much for your time. Let me just tell everybody listening, not only is Visier a great company and a pioneer in AI and analytics, but they've also partnered with us on Galileo. Many of the benchmarks that Visier develops in real time are available to Galileo users. So you can actually experience kind of the power of Visier in some sense in Galileo as well. Thank you. Paul. Yes. Wait, Josh. Josh,

Paul (00:26:12 -> 00:26:20)

Wait. One last plug. Stella, Kathy, Andrea, the amazing work on the People Analytics best practices survey. Can you plug that?

Josh (00:26:21 -> 00:26:38)

Yes. By the time this podcast comes out, we will have that report available. We've been working for a year on a large study of the people analytics domain maturity model, best practices. And thank you, Paul, for your help with that. I, I won't tell you guys the answers, but there's a lot of interesting findings.

Paul (00:26:39 -> 00:26:40)

Visier customers do better,

Josh (00:26:41 -> 00:27:00)

Visier, customers do do better. And I, I do think this function is growing up a lot very fast. It's not completely mature yet, but there's lots of progress. And thank you again, Paul, for, for your help with that project also great talking to you and, uh, always Josh. We'll talk again soon. Thanks. Bye, everybody. More later on other topics.