Analytics Business Partnering at Adobe & the Value of Comms - Sasha Arjannikova - #186

Speaker A: Hello, friends of the podcast and welcome to Directionally Correct, a ah, People Intelligence podcast with your host, Cole Knapper and today's guest, Sasha Arjun Kova, Director of Analytics Business Partnering at Adobe. In this episode we will cover how to build a people analytics function that genuinely partners with the business, bringing the,

Speaker B: um, stuff that people analytics team works on bringing back to the business.

Speaker A: How to earn the trust as ah, a data SME and make your insights drive decisions. How do you navigate the subject matter expertise Part of it. How Sasha went from herding cows and modeling on TV to becoming one of the most people centered analytics leaders in the industry. You've been kind of famous in your own right, correct? You were on TV in the past, weren't you?

Speaker B: Uh, a little bit.

Speaker A: And now let's get down to business. Hey, Directionally Correct fans. This podcast is dedicated to you to help democratize people intelligence for the world of work. If you're looking to support the podcast, please make sure to listen weekly. Subscribe to the Directionally Correct Substack newsletter. Sign up for the Data Driven HR Academy at data drivenhracademy.com Purchase Cole's Book People analytics or check out everything else@colenaber.com before we get into it, a quick word about HR Bench, the company powering this podcast. You know, when we all started in people analytics, we wanted to do strategic work building predictive models, workforce planning, advising the C suite, and most of all, quantifying the impact for the business. Instead, we spend months building dashboards and reports that should already exist. HR binge eliminates that entire phase. Your HRIS connects your metrics, calculate your benchmarks, populate. This is not novel. This is day one, not quarter two. That means skipping straight to prescriptive analysis, storytelling, and taking action for the business. Want to learn more? Book a demo@hrbench.com Directionally correct. Find out more about the company powering this podcast and building the future of people intelligence. As always, all opinions are our own. And thanks for being a listener. Well, Sasha, we've known each other for a minute now, right? Yeah, yeah. It's been fun to see your career kind of blossom. I saw you doing some work with the HR Strategy Forum. Can you tell me about that? What is it? And, uh, how did it come to be?

Speaker B: Yeah, absolutely. So I joined the board of, uh, HR Strategy Forum. It's nonprofit professional association that's been around for about 30 years. It's all volunteer led. And what really got me excited is the opportunity to create something that doesn't currently exist. So a lot of practitioners today in HR and people analytics, just across the board are trying to figure out what is happening, how we respond, how do we get ahead. And so really this opportunity to bring together peers from um, different companies, bring together academics, practitioners to figure out how we answer the questions that are emerging. It's been a pleasure working with Shreya Sarkar, who is the president and board chair at hrsf, a fellow IO for those in people Analytics. So um, definitely been great to focus on AI, um, how AI comes along and how do we as people practitioners really make sure that it's a people centered implementation. And really HR is leading a lot of the decisions as opposed to coming along or lagging behind. So it's been really exciting. And then uh, we have folks like Alexis Fink and Alex Levinson, who are also part of the task force with HR Strategy Forum, um, really coming together, sharing practitioner perspectives, academic perspective. So it's been really fun.

Speaker A: Yeah, both of them are friends in the podcast. Previous guests. Love, love that you shared that. Well, uh, you mentioned it. It kind of exists to help people get ahead. How do we get ahead right now?

Speaker B: Yeah, that's a, you know, an excellent question and a million dollar question. I think certainly, um, a lot of things are evolving. It's understanding what are the questions at hand and um, what is it that we, as, whether it's people analytics or HR practitioners, um, bring to the table, to the conversation. Um, one of the things that I feel very passionate about as a people analytics practitioner is how do we use people data for good? With AI, all of this gets amplified, biases get amplified, um, bad data. You put AI on top of it and we know what happens next. Right? And so, so that's one lens. Um, then from HR standpoint, like if you are a, um, chief HR officer or HR business partner, it's really an unprecedented time for influencing the organization to get it right from the people standpoint, not for just today, but for the organization in even like medium term, even if we're not talking long term. So a lot of business leaders are currently facing a lot of pressure with just competing, um, implementing AI, transforming their workforce and it's sort of short term decisions all over, uh, regardless of the industry. It's just that business pressure, the time of change. And so as hr, as people analytics, we have tremendous opportunity and tremendous responsibility to help make sure the business is set up for success long term. Because if you automate all of the, you know, roles that uh, more junior folks were taking in the organization, what happens to Your pipeline, what happens to your business along the way, um, a few years from now. And so, yeah, so it's exciting to be in the position, uh, of being able to influence, being able to raise those hard questions. That's something that brings me a lot of energy.

Speaker A: Well, I know you were saying that rhetorically and I've heard many people say, and I've probably said it myself about like what happens if you kind of pull up the bottom rung of the ladder. But what happens like in your perspective, like what does happen if, if that actually happens to those junior people and to the organizations that, you know, don't have that talent, um, pipeline for them anymore?

Speaker B: Yeah, I mean I think it's um, a lot, a lot to figure out a lot of unknowns. And I think also too, I am very cautious about uh, or hesitant to kind of project what happens because there's just literally so many things changing. But I think um, one of the things that we've relied on for years. Right. Is building strong internal pipelines, making sure that we have that institutional knowledge in the organization, uh, to build up the teams. Right. Like that's what a lot of the workforce planning built by borrow, um, is based on. And so, so if you continue to have, continue to be on that trajectory of having fewer and fewer people who are your internal pipeline, it presents all sorts of different risks to the organization. Um, I know you have a lot of thoughts on that topic as well, reading the manifesto uh, that you shared uh, earlier. And I don't know if I'm out of turn, but I like to break rules. Ah, at times there's no rules.

Speaker A: We can talk about whatever you want to talk about.

Speaker B: Great. Yeah. So I mean I'd love to turn that question back to you uh, because I, I'm, I'm sure you have a hot take and I am happy uh, to indulge you in that.

Speaker A: Yeah, I mean I've covered um, the manifesto a few times on here, so I'll try to cover some new ground if that's okay. Say something I haven't said, which is, um, especially on the bottom rung of the ladder. There was a reason why organizations were pyramidal shaped and it's because the bottom part of the pyramid is the widest. And it's the widest because it's the cheapest and it's like organization because we keep hearing about like diamond shaped organizations. Now a diamond shaped organization is pretty expensive and I think organizations are, because they're for whatever reason are temporarily forgetting that younger workers have like you know, engagement and, and they want to learn and they, and, and, and frank and maybe they need extra training and things like that. But they also are relatively cheap. And I think that is going to, in a very cost constrained environment. I think organizations are going to rediscover that because it's always like the pendulum is swinging back and forth and notice I haven't said anything at all about AI or people intelligence or anything like that. It's like we have to rehash these uh, arguments every few years, like every decade or so. We have to forget, oh yeah, in 2009 we just won't hire younger people anymore and then what will happen as a consequence? And then you go a few years and then organizations try to overcorrect because they didn't hire enough young people. And it's like this pendulum keeps swinging back and forth and it's fundamentally a workforce planning problem. And so I don't know if that, that really gets at what you're talking about, but I just have a feeling we're going to have to learn that lesson again. And I think it's silly.

Speaker B: Yeah. I mean, to build on that. Uh, completely agree with. We're over rotating on a lot of things. Right. And I think it's, it's kind of typical right at the time of change when we're trying to figure out how to do things differently. Uh, there tends to be going to extremes. Right. You read about organizations, you know, whether it's massive layoffs that are disguised under like, you know, AI is improving things. So, um, but I think there was one, one episode not on this podcast that you were speaking to that really resonated was the uh, meme you were describing, I think about um, the software and the junior software engineers and automating them with something.

Speaker A: Yeah.

Speaker B: That, you know, back to the cost picture. I think what's really interesting right now, a lot of organizations, especially in the early stages of AI adoption, they are not really looking at that cost as a comparison. Right. Like the cost that you're spending on like how much it takes you with tokens to get this stuff done with AI versus how much it, like whether it really should be, uh, continued to be done by humans. Uh, so I absolutely agree with you. I think it will hit a lot of the organizations and that's where I feel strongly that it's an amazing opportunity for HR to really like voice and articulate what will happen if they're not very intentional long uh, term decisions.

Speaker A: Absolutely. But really, really great points there, Sasha. I am Curious. And this may seem tangential, but I think it's actually really related to what your work and why I think you're really good at what you do. You're writing a book on something not related to people analytics at all. Can we talk about that for a second? Yeah.

Speaker B: Thank you, Cole. Um, that's been really fun. I think. I certainly am one of the folks with was a very non traditional career path to people analytics. And that's part of why it makes it fun. But for my master's and my doctorate, um, I studied communication. And the way I look at it is communication is the flip side of IO in the sense that how is reality constructed once it's communicated? And really a lot of theories about um, reality being subjective, reality in the kind of eye of the beholder. And your communication choices matter every step of the way. What you choose, how you choose to respond, how you choose to construct your narrative, um, it gets into influence and all sorts of things. So the book that I'm working on with, um, my co author, um, who also went uh, to the same school, um, for where I went for my master's, um, it's really focusing on how do you understand communication principles if you didn't go to school in communication. Um, so many of us communicate every day, whether you know, business leaders, HR leaders, um, and a lot of folks didn't go to school in communication, didn't have those principles. And with AI, once you use AI for everything, there is this notion of how do you evaluate quality if you don't have the experience, expertise in it. Right. And that's really what made me think about focusing on that. Uh, the idea here is like each chapter is designed to introduce uh, academic principles about communication. So I definitely didn't want to focus on something that you just can put an AI and never learn about it. This kind of, um, component of actually teaching folks in a very succinct way, but drawing on research, um, because what good looks like there's so many opinions now and there's very little credibility to it. Right. AI can come across as very convincing and credible. And if you don't have the academic backing or background to evaluate, you're going to be fooled very easily. And so the idea here is that each chapter is, summarizes the key principles that are academic driven and are research based about communication. And then it provides guidance on how you would use that effectively with AI. So, so there is a whole section of what you would include in AI instructions so that you can partner with AI to ask you questions on things like audience analysis expose the curse of knowledge and things like that. So it's been really fun, uh, to focus on. It's hard to find time on top of day job. But, um, thank you for asking.

Speaker A: Yeah. Well, first, I don't know, do you have a title and a release date?

Speaker B: Uh, we're debating the title. Um, don't have the release date yet, but more to come. Yeah.

Speaker A: Okay, well, when Sasha comes out, make sure you go check it out, everybody. But, uh, you mentioned the concept of audience analysis. What is that?

Speaker B: Yeah, it's, um, I think this is, like, super important. Uh, now more than ever. I see that a lot of times in my work in People analytics, it's very easy to assume that your audience is like one Persona or maybe two Personas, and that they're the same and they're not. Right. Like, even when you think about, like, I work a lot with HR business partners, and it's such a wide spectrum about, uh, who your HR business partner Persona is. And so if I am making assumptions about what people need to hear, what they need to know, um, that are based on this limited understanding on the audience, my message is not going to land well. And also in the time with AI, I see lengthy things, a lot of communication, so there's just a lot of noise. So being able to cut through the noise and land your message effectively by understanding your audience, it goes back to empathy, it goes back to being heard and seen, understanding who the audience is, what, what they care about. Um, and I think in people analytics and also in a lot of the AI implementation in general, we see sometimes this notion of, like, I've communicated it clearly, it's on them and it's not. Communication is a two way street. And if the message is not landing with your audience or particular segments of your audience, you have a lot to do with it. Right. The ownership is not just on them, and the ownership is on you if you want to have that effective outcome.

Speaker A: Yeah. Well, tell me this. How does that help you as a people analytics business partner and you unique and different. And what can other people learn from that that would help them be more effective in that role as well?

Speaker B: Yeah, thank you for asking. Cole. Um, you know the conversation that you and Alexis had on Directionally Correct a few podcasts ago, uh, where she played the, um, host, um, and there was a discussion about the importance of OD+IO psychology, and I would also add communication and change management to it. And a lot of times OD has kind of those components in it but this particular element, whether it's AI implementation, whether it's changes to the job, right? Like back to the manifesto and how the functions are collapsing, your jobs are changing, what you used to do, like, all of that is tremendous change. And we need not pretend that it's a, uh, light switch. Right? And I think when employees, when people feel like they're already feeling like it's moving too fast. And so when we don't acknowledge the human element, we don't acknowledge how much we're asking of employees, essentially we're asking folks to perform at the level that they're performing or higher because we need to compete. And we're also asking them to learn new tools. We're asking them to think about how to change their job, where there might be even thinking like, is this going to then replace me? So that's kind of threat in the back of their mind. Um, and so certainly as people analytics professionals and also HR professionals in general, leaning into that change management, communication, empathy, slowing down, understanding that we're talking to people, I think sometimes it's just very challenging because we're also under the pressure to move fast and show impact. But I think it's one of those, um, move slow and with intention to get to the outcome. And also like, related to that. This is the whole reason I'm m in people analytics. It's all about the impact. It is not about doing cool stuff, although I love cool stuff. And, um, advanced analytics and being able to answer really challenging, hairy questions with data. But at the end of the day, if the business doesn't take it and apply didn't matter. It's an overhead for the organization. You might have done cool stuff. And I think communication and change management is a lot to do with it because it's audience analysis. It's, uh, moving at the speed of the business because sometimes you can have great analysis. But the ship have sailed. They made the decision with the information that they had. And with AI, we have tremendous opportunity to help speed up those stages to get to those things faster. But it doesn't replace, but only amplifies this human dimension and that judgment, that ability to understand, that ability to influence.

Speaker A: That's so wonderful. Um, I'm thinking about this in terms of like. Well, let me, let me talk about the, the manifesto for a second since you brought it up.

Speaker B: Yeah.

Speaker A: Mentioned communication being important. And just because you communicate something doesn't mean somebody actually interpreted it that way. And then you said, you know, it seems like things are moving fast. M probably faster than people are comfortable with. That was actually one of the motivations for the manifesto, is everybody's saying things are changing, things are changing, things are changing. But nobody was saying what they're changing from and what they're changing into. And here's the playbook from how to go from one to the other. And that's why I wrote the manifesto, because it was like, hey, here is what I'm seeing that's changing here. I'm putting some. A trail of breadcrumbs as to where I see things are going. I'm not perfect at predicting the future, but I'm trying to build it, especially in terms of the work that we're doing at HR Bench, and, and that's the best I can do. And then, uh, you know, I'm probably, I mean, I don't know if I should mention this on the podcast, but I'm probably writing a book on the topic. Uh, I know I'm a sucker for punishment as well. This is my second one. Um, but we'll see what happens. You know, and it's all under that motivation of, like, can you communicate this effectively to help people move from point A to point B so they can be effective in their jobs? And then if they're effective in their jobs, that actually makes our whole field shine because we look like the brilliant ones who are actually navigating the future head on rather than just reacting to it. Like many other disciplines are

Speaker B: great call outs. Cole and I will be looking forward to, uh, reading that book. I also, you know, reading the manifesto, what struck, uh, me is even you calling out how the concept from a year ago that you were very proud of and made a lot of sense is now being impacted by the change in the world. And I think that's also something that doesn't go unnoticed. And for me, it, again, amplifies the need to slow down in some of the places. Um, I also think I want to poke on something you mentioned that a lot of people, analytics, uh, folks are thinking about it but not voicing it. I think it's also a human element. Right. Like, and I think as, as people think about it, it's back to that, like, do we want to create a sense, um, of anxiety among the team, for example? Right. If we are completely consistently talk about, like, oh, you know, the functions will collapse. Like, there is an identity element there. So I think certainly that human element, I also think it's certainly not going to happen over time. So that's also another component. It will happen over time. I do, uh, want to say, um, on the record that I disagree with you, that we will not build another dashboard. So I had to sneak that in.

Speaker A: It's not that you won't have to. It's not that you never will. It's that you won't have to if you don't want to.

Speaker B: Yeah, I thought that there might be and part of it also to manifesto knowing, uh, you. It uh, also the style is like you want to be a little provocative so that you can have a conversation. But certainly I do think that, you know, the dashboards, uh, the ability to kind of use tools to expedite a

Speaker A: lot of the work.

Speaker B: Like um, this is, certainly is here to stay. I do want to emphasize the, the importance which I think this doesn't get talked enough, um, because it's not sexy and not very convenient for vendors to talk about because there's not like a, a big, you know, sell on. It is data quality data. You know, that just the challenges with the process, the business process of how data gets created in people, um, in people's space, that's going to be continuing a big focus because again to that point you put AI on top of data that is not clean data that's not good. And so I think that's also going to be a big focus for people analytics moving forward.

Speaker A: I actually think you're so right on the data point. And it's a really hard problem. It's an extremely hard problem. And I don't think anybody, I can't remember where I talked about it, but there's like, I called it like the existential problem of data quality. And I can't remember if it was on here or somebody else. I think it was actually somebody else's podcast where they had me as a guest and I talked about the problem of zip codes. Like imagine somebody just puts in the wrong zip code for their, like where they live. Like how do you know it's wrong? Like and like you could prompt them and be like uh, and if it's not five digits, maybe it's six digits. Like, okay, you can create some kind of mechanism. You can even have like the drop down of zip codes that show it might be the right zip code, but somebody might not look at it closely enough and still click the wrong one. You know, like at a certain point you just have to accept that people are going to put wrong information in and, and that's an existential problem because it's, it's only we uh, like how many times are you supposed to prompt somebody or ping Somebody be like, hey, have you double checked your zip code? Have you double checked your zip code? Have you double checked your zip code? Like, at what point are you just pestering people, just trying to improve your data quality? And it really. And um, I know I'm beating on the zip code point, but like it goes for every single data field we have, right. Is they all suffer from the same problem, which is like what is ground truth? Right. And if there's a divergence between what people say is ground truth and actual ground truth, how do you adjudicate that? And that's a really big problem.

Speaker B: I agree with you. I would also build on it and expand because a lot of the data is coming from even a business process that has to do with how people operate. Now that we're trying to move fast, now that there is a lot of disruption, like keeping up with that is very challenging. And a lot of times who is responsible for data? Maybe people analytics. Are they responsible for the end to end process in creating each one data point? No. Right. And also too, that's another element. Element that I think will continue to be salient. The other point too is, um, our current kind of ways we look at data and processes too. A lot of the processes and mental models need to evolve. Right? Like your annual performance cycle no longer cuts it because work is not happening that way. Even your hierarchies, right. If the teams are doing work not based on who they report to, but they are more small and jail nimble and short term kind of we're going to galvanize around X and our HRIS has them either by cost center or by reporting hierarchy. We're already behind. So there's a lot of opportunities and I am certainly a data governance nerd, um, because I crave structure. Uh, but I um, think certainly that that continues to be a big focus for you, um, know for us.

Speaker A: Well, it's funny. Um, I love that and I love that like that's part of your identity. And you mentioned earlier one of the things that people are struggling with is like the identity changing in the field. And I actually had a guest a while back, Jay Van Babel, which most people and people analytics don't know him, but he actually might be one of the most famous guests I've ever had on the podcast. And I brought him on just to talk about social identities and how they're changing. And so it didn't actually perform all that well compared to many episodes. But, but because the topic seems like not people analytics and the guest seems like not people analytics, but I actually brought them on just for that reason is because I feel like our field is struggling with an identity problem. And I wanted to bring on a researcher who could talk about what does that look like and how do you kind of navigate that challenge. And I, uh, frankly I've gone through that myself. Like, it's, it's been a tough ride and like some of the, some of the things I write and I put out there and some of the provocative stuff is just like my own personal therapy of like, how am m. I dealing with the fact that the world's changing around me and I want to have control over it, but I don't necessarily have 100 control over my own destiny. And I want to kind of like, take back the reins if I can.

Speaker B: Yeah, I love that I'll have to catch up on that episode because that certainly resonates. Um, and I think for people analytics and for HR in general, um, there is certainly an opportunity to continue to zoom out and learn from other disciplines. Disciplines, especially as the world of work is evolving.

Speaker A: M. Absolutely. I want to bring it back to what you're doing. Like, and I almost. I hate to go down just like such a simple level, but I think it'd actually be helpful. What does a people analytics business partner do and what does good look like and how do they play a role in shaping, like how Adobe's people strategy is executed?

Speaker B: Yeah, um, absolutely. Love to talk to that. So my role at Adobe is in analytics business partnering with, restructuring the team, uh, maybe like a year and a half, couple of years ago, uh, to have more in like those domains of expertise. So we have um, BI team and reporting research, um, data engineering, data science, uh, workforce strategy. And then the analytics business partnering world is really more direct with HRBPs, with HR, COEs, with business, uh, leaders. And that's the space that I operate in. Um, it's been a wild ride, a very exciting ride because it gets me closer to understanding how people are experiencing what we're bringing, where it's meeting the mark and gets them excited, where maybe barrier to entry is, uh, a bit too high, where there's some friction, as well as also hearing what is emerging, what is the business model focused on right now. And that's um, and as an analytics business partner, being able to then bring back to the people analytics team to say, like this number one thing that the business cares about right now, we need to pivot and we need to find ways to answer it and do it fast because otherwise the ship will Sail I find that really exciting. And then vice versa bringing the um, stuff that People analytics team works on back to the business and figuring out how to drive that impact. So we do you know of course the uh, employee survey work. And so there is tremendous level of insights driven by data engineering, data science, research. And then how do you get leaders to not just engage to understand but engage to take action and to make it visible to employees. Because what at Adobe I've been very grateful that the culture is very data focused and very people centered. Um, there's been places in the past in my career where it maybe hasn't been like that. And I've also worked not only internally in house but also as a consultant. And so coming in and trying to explain why people data matters and explain why you should care about employees, that's a very different conversation to have versus like this is what we've seen in the data. And so uh, I get very energized from that work fact that there is a lot of interest in understanding whether it's employee sentiment and what, what do employees care about and where the hotspots are. Um, but I think there's also a lot of opportunities to make sure that the People analytics is operating at the speed of the business and that's you know, a century old program problem. Um, and so again it gives me a lot of energy I think what to learn from it. I observe the opportunities to speak the same language. That's why audience analysis is really important because the way you engage with an HR business partner who is like a savvy user of data and they don't need much support versus somebody who is just early starting with it and maybe is not as familiar. It's a very different dynamic. Right. Um, and some of the great learnings from my work um, at ah, Adobe has been really leveraging HR business partners who are more kind of excited and engaged and more technical and data savvy to not only give us not only like champions, I think the idea of champions is pretty traditional, right? Like you build something and then you engage folks a little bit earlier but to actually co create. So we've done that and that's been really successful because we're getting very direct user input. And what struck me is sometimes the things that people care about are actually very easy to fix in the tool. So it's just giving us very direct lens of what's important to both the user as well as the business versus us, assuming that we know it because we're also in HR and we Work with, with those folks every day.

Speaker A: Absolutely. How do you navigate. I feel like this is the hardest problem with being a business partner. How do you navigate the subject matter, expertise, part of it, which is you have to be, um, in the data and in the products that you guys have and in the, the, the just the analytical techniques, the sophistication, all of that to, to frankly deal with challenges that come from your customers. But you also have to know the business, have to have earned the trust, have to speak the language, have to communicate effectively and, and be, you know, kind of that, that true business partner enough that you know, you're actually can influence decision making. How do you do that? Like what. I mean, I know I'm probably asking a terribly form question, but I feel like that's a really big problem.

Speaker B: It's, it's really fun. I think this is one of the things that like gets me really excited. Again, it's understanding the. I, uh, think you talked about that. Did you call it translator in one of your writings?

Speaker A: I've used a variety of terms, yeah.

Speaker B: So it's kind of that like you, you essentially live in both worlds in order to be effective, right? Because you have the luxury of spending the time with the business to understand not only what keeps them up at night, but what hypothesis they have about what keeps them up at night, what assumptions they have. You also spend time asking questions, observing. Um, it's kind of almost like, um, an ethnography, but much quicker, very, very fast. Um, because you're making all of these like you're constructing the reality that the business has. And then you're going back to the people, analytics and distilling that into something that your data science team or your research team can help with. And then you're taking this like complex, robust, uh, analytical product and figuring out how to communicate a few things that will land really well. And in the language that the business is talking about, you leave out a lot of the sound stuff that is very important from the analytical rigor perspective because you have the trust of your business partners, of your business leaders that you know this stuff that they need and you can bring it up, right? Like if they ask a question, want to go into a rabbit hole on something, you, you happily do that. Um, but that's the magic of it, I think. To me, what energizes me is that it's a lot of complexity, it's a lot of judgment, it's a lot of reading the room. It's a lot of understanding what is unspoken and then being Able to then translate it into something very clear, take into account what data we have. Because a lot of questions we can't solve because the data is just not there. To be able to answer it in the way, um, that is required, um, those are some of the things that I find most energizing and kind of been exciting and impactful.

Speaker A: I mean, you. You've been kind of famous in your own right, correct? You were on TV in the past, weren't you?

Speaker B: Uh, a little bit.

Speaker A: Care to tell us about it?

Speaker B: Um, I think you're. You're talking more about, like, let me frame it this way. I'll share about my, um, untraditional career path. So my first job was a cowherd. So it's like a shepherd, but for a cow. I think I was 8 or 9 years old, and I had a responsibility for one cow because the cow was too young to be able to join the big herd. And so I had to get up early, I had to be on time, and I had to not lose the cow because that would be a big deal. And so shepherding or cow herding early on in my life taught, uh, me a lot of just responsibility. I, uh, was paid in milk. It was the best milk I've ever had in my life. You know, when you buy the whole milk here in the store, it's just nothing compared to that. Like, the layer of cream on the fresh cow milk from the cow that you herded is, um, very, very satisfying. Um, and then the other detour that is very unconventional was modeling, which, um, taught me a lot of confidence because you had to be confident no matter what. It. It also made me realize that, um, there's other things I could be doing in life. I had all these ideas and contributions, and they're like, no, just smile, please. So, um, that was, uh, um, another interesting experience. I think also too, what it taught me is what a career is from the perspective you have from the outside is very different a lot of times than the reality of it. The glamorous part is pretty short, and, um, there is a lot of just work that is not glamorous. And so definitely for those who are starting their careers, ask people who are in the role, what are some of the things that they do as part of their career that is maybe not as fun?

Speaker A: I have to ask which one's less glamorous, Cow herding or modeling?

Speaker B: I enjoyed cow herding a lot more, actually, because I guess with modeling, maybe there was a little more disillusioning with cow herding. It Just feels pretty straightforward. You're out there. Uh, but, yeah, you know, comparison we don't speak of very often cow herding or modeling. There might be a need for a research, uh, project there.

Speaker A: Yeah, absolutely. Well, you want to join me in Cole's Corner?

Speaker B: I am very excited about the articles. Um, so, yes, please.

Speaker A: Welcome to Cole's Corner. Well, before we get to the articles, we got some rapid fire.

Speaker B: Okay.

Speaker A: Um, and, uh, one of them is related, what we were just talking about. So if you weren't in people analytics or being a model or being a cow herder, what would you have done with your career?

Speaker B: You know, I'm really interested in just HR in general. Like, um, I think it's. It's probably. It's one of the things that is, like, very appealing. HR or business, um, in general, because again, to me. And it also speaks to kind of how the roles evolve. There m. May be more of the kind of, uh, merge between the domains of people analytics, and there are some, you know, HR business partners who come from more analytical background, um, hr, senior leaders. So that's, That's a track that is, um, of interest. I also have a lot of, um, hobbies. I love plants, so I, um. My latest has been to help orchids re bloom again. And that's been very satisfying because a lot of times you buy an orchid, they look great and then they. Yeah, so that could be another, you know, thing I could.

Speaker A: You've got a lot of.

Speaker B: A lot of curiosities, a lot of hobbies.

Speaker A: I love it.

Speaker B: Yeah.

Speaker A: Um, what's the place you've never been to that you most like to go and why?

Speaker B: Um, Dominican Republic. And we have a trip planned later this year. Just heard a lot of great things and very excited. Uh, I love the kind of cultural experiences. Um, my husband and I, and we have a daughter who is five. Um, so we travel quite a lot. My daughter, I think, has been in seven countries before she turned five. So it's, uh, it's always really exciting to see the world through her eyes because it's like all, all, all new. Um, Costa Rica is another country that's on my list there to visit.

Speaker A: So, yeah, I got to speak at a conference down there earlier this year. It was awesome.

Speaker B: I've heard about that. Yeah, that's great.

Speaker A: That was super cool to be down there. Um, if you were a character in any book, TV show or movie, who would you be and why?

Speaker B: I would be a superhero because, uh, it would be really cool to have some skills and competencies that I don't have today. Um, and to do good things with it, like, you know, it's just like a very logical transition with superhero.

Speaker A: You have to choose one.

Speaker B: Um, you know, I'm not very versed in superheroes. I don't watch much TV because I always have projects that I could be doing myself.

Speaker A: So what's the superpower you would have?

Speaker B: Um, I would fly. I would be able to see things that I don't see. Um, so, yeah, those two.

Speaker A: All right, I love it. So I've got one big question for you before we move into the what am I reading? Section. And it's about the future of business partnering in people analytics, but in HR more broadly. What do you think? Like, if you looked three years into the future, how will your role change and what will it look like?

Speaker B: Um, as in people analytics or HR business partnering?

Speaker A: Um, well, the comment.

Speaker B: Both.

Speaker A: Let's.

Speaker B: Okay.

Speaker A: Or either.

Speaker B: I think, certainly, you know, we see a lot of the. Like, I agree with you on the manifesto side about, you know, the roles are merging. Right. We see that in the business. We see that in people analytics. Um, what you had to have a specialized, maybe engineering skill set or a technical skill set before. Now you could use AI for a lot of it. Um, so I think certainly the ability, and hopefully the ability to answer questions faster. Right? Because you just have tools at your fingertips and you're not relying as much on, you know, working with other teams to get you something that is like analysis, et cetera. Um, I think just ability to move faster. Um, I think there'll be more. Certainly AI native is an element. Right? Like being able to have AI in every part of your job, but, um, also integrating in such a way that you still have the human element there and you retain the excitement. Um, I have to quickly reference, um, one of the folks that, um, is a influencer, um, and teacher. Ah. Um, I'm gonna have to come back as a name. Um, it's Dave, I think. But anyway, um, he was talking. He came to do an AI day at Adobe, and it really resonated. For the first time, I heard someone in the AI training really talk about how to think about tasks to automate in a way to not take out what brings you joy. And I think that's really important. So I think that future of the profession, whether it's, you know, analytics, business partnering, people analytics in general, or HR roles, it's making sure that you are augmented with AI, but in a way that you continue to have deep satisfaction versus you're feeling Threatened that, like, what is it that I'm doing here? So I think that will be a big element there. And, and I think also too, we're also seeing this trajectory where AI is, um, bringing value. But that desire to have human connection, that desire to have human judgment, somebody who's actually an expert and not just read something that AI put together, I think that will be amplified. That's just my take.

Speaker A: Yeah, I know we've talked about it a few times, but Stacia and Danny from Red Thread came on and talked about the concept of hollowed out expert, um, being, ah, a problem of the future. Like, how do you even know if somebody's really an expert if everybody sounds like they're an expert? But the other thing you were saying, um, about I guess the guy named David or whatever.

Speaker B: Yeah, it's cool. Dave burst from, um, the Gen AI Academy and he's based in England. And I highly recommend because it's like that human centeredness, I keep saying, like,

Speaker A: don't build a future you don't want to live in. And for whatever reason, I find when people are experimenting with automation, for whatever reason, they always start with their favorite part of the job first.

Speaker B: Exactly.

Speaker A: And it's so dumb. Yeah, it's like m. Automate, the thing you hate doing.

Speaker B: Why exactly. Yeah, you talked about it for calendaring. Right. In one of your podcasts. Like, automate stuff, make it fun. I don't like doing status updates and when we introduce some automations in there, I'm like, oh, I actually could do that. Like, it's what I dreaded the most.

Speaker A: But it's like, it's almost like we, um, there's like this, uh, like, I, uh, don't know, Icarus type moment or Frankenstein type moment where we like have to as humans be like, can this do my job better than me? Can it do everything I can do? So let's see if I can try out the thing that I think I'm the best at first. And it's like this weird, like psychosocial thing that we have and like, we just have to try. It's like, stop it. Just go do the thing you hate doing, like, why?

Speaker B: And figure out how to make yourself faster. Right? Like take out parts of the work so that you can deliver more faster and retain that experience excitement.

Speaker A: Yep, yep, absolutely. Well, let's get into the. What am I reading? I have a feeling you're excited to talk about one of these articles.

Speaker B: I am excited to talk about all of them. Um, so, yeah, you go your intro and then, um.

Speaker A: Yeah, all right, let me get down to it. So one of the things that I've covered probably ever since the podcast started is just people talk about bias a lot. And one bias that I think is particularly pernicious and no one talks about. And I found an, uh, article from the Journal of Personality and Social Psychology that came out last year called Social Bias blind spots. Attractiveness bias is seemingly tolerated because people fail to notice the bias. And I'll say this about, it was a really well written abstract. Uh, and so I'm just going to read it as is, because most abstracts are not as, uh, you can't understand them most of the time. But this is a good one. It says, discrimination remains a key challenge for social equity. A prerequisite for effective individual and societal responses to discrimination is that instances of it are detected. Yet prejudice and discrimination are rarely directly observable. And the presence of discrimination has to be inferred by circumstantial evidence, such as overrepresentation of certain individuals. Here we study how people judge the outcomes of statistically bias samples along different dimensions and they use six different studies. So I'm going to skip down here a little bit. But basically they look at gender, uh, biases, race biases, and then attractiveness biases. And the funny thing that they find is, um, if people find bias for gender or race, they, uh, they, they say it's like, okay, this may exist, but it's bad. However, if people find bias associated with attractiveness, they recognize that it exists, but they're like, they say it's more likely to go undetected. Our findings suggest that seeming tolerance of attractiveness biased outcomes is partly explained by people's failure to spontaneously notice that the outcome is attractiveness biased in the first place. In other words, it is possible that people showed muted responses to biased outcome because they are not actually, uh, they actually approve of it because they fail to notice the bias in the first place.

Speaker B: What do you have so many thoughts? Um, I mean, do you want to share your thoughts first or do you want me to go?

Speaker A: Because I just talk for a long time.

Speaker B: Right.

Speaker A: So what are your thoughts? And then I'll share mine.

Speaker B: I think, um, a couple of thoughts come to mind. Definitely that point of not being recognized. And I see a lot of parallels with different types of biases in the data, especially if you put AI on top of it. Um, so back to your zip code example, right? Like there's a lot of things that are not culturally known as, in the same way as like the Other protected characteristics bias. Right. They're not part of the vernacular. They're not part of the culturally known and kind of accepted set of biases that we as a culture globally think, uh, of as bad. And so when biases go unnoticed, there is a big danger of them being perpetuated because we don't test for them a lot of times. Right. Yet with AI, it can be amplified a lot more. And so I think that's one element. The other element that I think is very salient and very dangerous is that, um, if you look at how attractiveness bias is perpetuated by, um, large, um, like just AI models, right. When, um, because that bias exists, it goes unnoticed.

Speaker A: You can try to get AI to show you an ugly person when they generated from scratch. It does not exist.

Speaker B: Absolutely. And there's an article I was reading as I was nerding out on kind of additional things related to this. Like there was a test actually with, um, uh, with AI looking at pictures and trying to assign a characteristic like a profession or whether the person is confident or insecure. And they took a set of normal human pictures and then they put a beauty filter on top of them. And guess what? What AI coded. AI consistently, persistently coded the folks who were beautified as more confident, more rich, more in like, successful careers. So it's already here, it's everywhere. And it's also something that humans have not established as a kind of set bias that we test for. For. So I think, like, to me, there is a lot of hard, um, implications of that.

Speaker A: Yeah. You know, um, Mark Andreessen, uh, who's a famous kind of venture capitalist, um, he wrote something pretty provocative a few months ago, and it was like, there's only going to be four jobs in the future. And I can't remember two of them, but one of them was like, engineers and product people and like any kind of scientists are all going to be combined into one. And then he said the fourth category is hot people, and hot people are going to do all the interpersonal jobs. And I just keep thinking about this as, like, it's so out in the open, this attractiveness bias, that we're just like, going to call it a whole job category in the future and people are just going to accept it. Like that is a proxy for competence, which it is not. It's a huge bias, and I do not understand why people tolerate it. It makes no sense.

Speaker B: Yeah, I couldn't agree with you more. I think it's. It's certainly an area that needs attention and Especially with, um, kind of the evolution of things. And was AI exacerbating this?

Speaker A: I always try to bring on a research topic related to something that our guests are uniquely qualified to talk about. So as a prior model, I appreciate you suffering through this, um, with me and talking about my personal access to grind.

Speaker B: Talk about the second article.

Speaker A: Okay, so this is also an interesting one.

Speaker B: This one is. Yeah, I have a lot of personal

Speaker A: experience to share, and so, uh. And this is actually Jay Van Babel, of all people. Um, the, uh, he. He found this on LinkedIn, but it's. It's from a research article in Psychological Science called the Long reach of one's Spouse. Spouse's personality influences occupational success. And so he says, pick your relationships wise. Wisely. For both both men and women, their partners conscientiousness predicted their own future job satisfaction, income, and likelihood of promotion, even after accounting for their own conscientiousness. Why? Because conscientious partners, both men and women, perform more household tasks, exhibit more pragmatic behaviors that their spouses are likely to emulate and promote more satisfying home life, enabling their spouses to focus more on work. What do you think about this one, Sasha?

Speaker B: I loved reading this. So I got the full article and I nerded out on their method section. I absolutely love that they studied the idea, uh, of delegation, the whole notion of like, if your, uh, uh, spouse is conscientious, you could delegate more things. Right? You could feel like, I trust this to be done. I don't need to be worried about it. Um, I also really liked the gender, um, you know, balance on that because there were citations about like, behind every man there is a great woman, right behind every great man. So they flipped it over on the other side as like, uh, behind every person, uh, there's a great spouse. Uh, so I have a lot of personal experience. So my husband and I are both extremely conscientious to the point of being nerdy and very boring. And so, um, I just personally can relate, like, um, the times when he is on travel and I have to manage it all, you know, childcare responsibilities, et cetera. It's very stressful and I am not as productive. I get less sleep because there is just more to worry about. And the times when he is here and his normal self, there is, you know, this sense that you could trust to have this whole dimension of things be taken care of some by someone else who will do a good job. I think it removes a lot of stress. It allows you to focus on work. It makes you more productive at work because you're not mentally kind of checking off your list. So, uh, I loved reading this article and shout out to Edgar.

Speaker A: I love it. Yeah, I guess the only PSA I'll make because I agree with everything you said is there's been all this stuff coming out lately about how like people aren't coupling anymore and their people are having less babies and like the population is about to like burst and all this kind of stuff. And the thing I just say is go get married, go have some kids. But just make sure you find a conscientious mate before you do because then everything's going to be much easier on the back end.

Speaker B: That's, that's a good, good plug. And I think also too that's what the LinkedIn post was, was about. Like think about the priorities because you want somebody who's agreeable, et cetera. But it's the conscientious.

Speaker A: So yeah, absolutely. Let's, let's do the last one real quick. Um, this is from, and I made a post a while back asking people for like recommendations of smart people that they uh, follow. And somebody recommended Sanjeev who writes this substack called the Work Design Lab and it's been fascinating. I've been following it. It's doing a really great job. He made this post called Finance Got there first. Anthropic CFO Krishna Krishna Rao is no longer running backwards looking meetings. The architecture that made that possible isn't proprietary. HR just hasn't built it yet. And so he goes through all these examples uh, from Anthropic about how their finance team m actually built. I think it's something like 70 something skills in Claude that allow them to automate essentially all of the things that a ah, traditional like finance meeting would need to do kind of backwards looking analysis. And so they get to spend all their time looking at forward looking metrics rather than this, this backward looking, very manual way that most teams do things. And so he gives a few examples of how HR could adopt a similar mindset. Tool number one is the HRAS mcp which the Tool number two is the Comp Planning dashboard. Tool number three is the Talent Signal internal mobility layer. Tool number four is the HR variance analysis. And then he talks about how you can just do this all yourself. You don't necessarily need um, any external tools but you got to make sure you get your architecture correct and all that type of stuff. But I thought this was extremely provocative. Um, and I tried to imagine what it would be like to have An HR leadership team meeting. And it only be forward looking in nature and not backward looking and just that in itself, even technology aside, that in itself would be quite radical. And I know you probably are in a lot of those types of meetings, not just in HR, but with the uh, parts of the business you support. And so I was very curious, what did you think about this one, Sasha? And how could you see this, you know, manifesting itself in the next few years?

Speaker B: Yeah, I mean I think a lot of things about the architectural setup and the infrastructure, those are the focus for a lot of people, analytics professionals. Right. Like taking advantage of like even the data layer, the insight layer, how you deploy it. Right. Is also uh, a question. In addition to forward looking, I also think it's very important to continue to bring in the impact of people analytics. It's traditional, been very hard to measure because you give uh, leaders an analysis, it really resonates. They go in, they make decisions and very infrequently do you go back and check like what exactly difference that it made. So I think that's certainly one um, layer. I do see that in the business conversations there is a lot of focus on what, you know, what it will be in the future. Right. And what do we know from today? Like they don't know, they don't want to hear about what happened, they want to hear about what to do next. Right. So I think all of this certainly resonates and I also do feel like it builds on the curve of credibility. Before somebody listens to you about predictions of the future, you need to build credibility about um, kind of the work you're doing day to day. But it resonates.

Speaker A: Certainly resonates, absolutely. The one, uh, the one statement that I'll make is um, I think this is actually even more relevant for finance than even for hr. But it's putting a lot of trust into your own internal build applications and making sure there's no like AI drift that happens from when you build it to in the future. And I don't know if I'd want to do that with my company finances frankly yet. And so in, in your HR measures to, to boot. And so I don't know, it is pretty radical. I love the ideas and it kind of gives examples of like how you would build this yourself. I'd say you better test and double test and triple test and quadruple test and make sure all of this is correct before you put it out there. Because there is um, I can't remember where I saw it, but somebody uh, said something the other day which was like, um, if you're, if you're doing things um, that are determinist, deterministic in nature, like they're actually like a fact and the fact never changes, you probably should be using traditional software. And if you're doing things that are probabilistic and making projections and needing to uh, combine multiple ideas and it's okay if not being 100% correct every time is okay, you should probably be using some kind of AI tool. And so the most important thing to know is is your outcome deterministic or non deterministic? Not should you use, be, be using AI or not using AI. And I think some of these things, especially in finance and especially in HR when it's just a set of facts that you should probably be using more deterministic things in those cases.

Speaker B: I do feel like there is this caveat also. It's anthropic. Who's talking about their own tools? Right?

Speaker A: Yes.

Speaker B: Right. Versus really, you know, who out of any other organization really would, you know, it's using your own product. So it's kind of that element, um, versus being a universally. This is what uh, what you do. And I think it's also back to data quality, back to setting up all of the parameters in the way that actually is sound from like solid competency perspective. And as you said like the review uh, before it um, it lands. I think I find that a lot of times in these publications and articles that position something as like revolutionary or groundbreaking. There's a lot of context that's left out. Right. About the um, the parts that actually it takes to, to get there.

Speaker A: Well, and this is why actually I read a lot of this stuff is because it's fun to experiment, it's fun to build an HRS mcp, it's fun to try to see if you can automate your HR variance analysis. All this stuff is super fun. The question is, is it right and is it right every time? And those are different things like. And so I totally agree with what you're saying. But Sasha, you've been a fantastic guest to have on. I've been looking forward to this for a while. So thank you so much for bearing with me going through some of this stuff. But if anybody wants to reach out to you, where can they uh, where can they find you?

Speaker B: Yeah, um, LinkedIn is a good place. I also would love to. So Sasha Gennikova, there is not another one like that in people analytics. So uh, and then um, I would love to invite everybody to the next AI collab because it's, you know, it's. We have one session in September. We also have a session with Alec Levinson as a thought leader. So certainly it's HRSF AI collaboratory. And so definitely, um, would love to see more practitioners from People analytics space. From HR space, because the quality of conversation depends on who is in the room. So when.

Speaker A: And that's part of the HR Strategy forum, correct? Yes, M. And if people. Is that open to anybody or who can attend? Yeah.

Speaker B: So the target audience is really HR practitioners. Um, but we also have folks who are maybe in transformation role or in IT role that is part of, like, AI implementation in organizations, because in a lot of companies it doesn't necessarily sit in hr. And so. Yeah.

Speaker A: All right, well check out the HR Strategy forum and some of the upcoming sessions and look UP software on LinkedIn and all the good things. But you've been listening to Directionally Correct, a People analytics podcast with your host, Cole Knapper, and today's guest, Sasha. I'm not going to try to pronounce her last name. Thank you so much for joining me.

Speaker B: Thanks, Cole. Appreciate it.