Modeling Culture: Using Predictive Thinking to Break Assumptions and Boost Innovation with Jake Brintzenhofe.

Speaker A: Foreign.

Speaker B: Welcome to Vibe Makers conversations with culture Activators where we dive deep into the heart of workplace culture and people ecosystems with some of the most exciting minds in the business. And today I am, um, excited to invite Jake Fritzenhoff to Vibe Makers. Jake graduated from the University of Maryland in 2015 with a Bachelor's degree in math and economics. He spent the first 10 years of his career doing analytical and communications consulting, specialized in applied predictive analytics, analytics and technical presentation to non technical audiences like this gal right here. And recently founded his own firm, Air Term Consulting, where he is doing both educational and applied work around strategy and predictive modeling. And Jake and I actually work together briefly and you know, it was actually in a recent catch up conversation, Jake reached out, he said, hey, I'm thinking about something that's kind of cool is it applies to predictive modeling and decision making around talent acquisition. So we connected and he went over this philosophy and I was like scribbling like a mad woman of like, oh my gosh, this is brilliant and insightful and I wish more people could hear it. And so immediately in our conversation I was like, Jake, you gotta come on and be on Vibe Makers and uh, tell us what you're thinking about lately. I am excited he finally made it after some scheduling issues on my end. And welcome Jake, to Vibe Makers. We're so glad to have you.

Speaker C: Thank you for having me, Marnie. I'm flattered to be here.

Speaker B: Awesome. So starting from the top, could you tell us a little bit about yourself and what brought you to this philosophy around people and culture and talent acquisition?

Speaker C: Yeah. So as you said, I started my career doing applied predictive modeling. I've done consulting in a lot of different spaces. I've spent the bulk of my time focused on building predictive models, helping people who aren't necessarily experts in predictive modeling understand what those models are doing, what, when we can trust them, when we shouldn't be so sure, and then thinking about a lot of things in the rest of my life. So I've been in a workplace, I haven't been my own boss until pretty recently. And culture has had a major effect on my day to day life. That's been maybe eight, nine hours of my day that I was reporting to somebody else's culture in some other company. And it was a really, really important thing to me to pay attention to that. And because I've spent so much time building models and explaining models and thinking about models and trying to solve these problems for myself, there was some intermingling between Those two sets of ideas.

Speaker B: Yeah, it's funny how your mind shift changes when all of a sudden you're, you're having to take onus over your, your own thoughts and approach on things like culture and environment. So Jake, how do you actually see predictive intelligence supporting and shaping and sustaining great cultures and people, ecosystems?

Speaker C: So I'll tell you, it's not through like the high level technical application of mathematical ideas. It's more of a step back in a qualitative. But the idea when you're building a predictive model and all the math that goes into it, you're just trying to predict, I want this good outcome, how do I get to this good outcome? Or I don't want this bad outcome, how do I not get to this bad outcome? And that could be for building a site and trying to make money. Maybe you want to build a new store, maybe you just want to make your spouse feel nice at dinner and you want to serve them something that they enjoy. Maybe you want your daughter to, you know, enjoy your time with you. And so that happens in any part of our life. And that could be in culture, it could be in hiring. I want to hire somebody who's going to do a good job. I want to feel good when I go into work. I want my employees to feel good. All of the things we're doing in a more technical space with predictive modeling are saying we want to build a model that predicts this effectively. If you can do that and bring those ideas, maybe not the math, into the rest of your life, into building a culture, into hiring effectively, you're going to be able to do things more effectively. Because these models are all about making good predictions at the end of the day. And that's what we want to do

Speaker B: in hiring and culture, you know, And I think what really spoke to me with that, because it's brilliant, because when we talk about culture specifically selfishly on my end is, uh, we talk about how it's a little hooey sometimes or that we can't. It's what is actually a culture. Well, a lot of people say it's a thousand different decisions every single day that an organization makes. And so if we can get really intelligent about the way that we go about decision making in our organization and how it pertains to the culture that we strive to have as an organization, that's a great leading indicator of success in our organization. So I think that's where I was like, yes, like it is really taking that technical approach to a very non technical Way of doing business.

Speaker C: Yeah. And you can apply a lot of those technical ideas. Einstein said, if you can't explain it to a child, you don't understand it well enough. You don't need a map.

Speaker A: What happens when employees behave badly? Boy, we could, we could do an entire tv, uh, show, uh, maybe, maybe a Netflix special on that. Well, Ryan and I sat down and recorded episodes for Fama and we asked practitioners, give us your most outrageous story. You know, the sales leader that brings cocaine to work, you know, whatever, just bring us the outrageous. And it is funny. So if you need a laugh, which we all do from time to time, search for workplace misconduct, um, wherever you get your podcast and you'll find it. And trust me, you will laugh and cry, but you'll definitely laugh. All right, thank you.

Speaker C: I agree. To understand this stuff, you can apply highly technical ideas and highly non technical ways.

Speaker B: Yeah, so let's talk a little bit about that. Uh, you know, as far as. Because the way you explain that, the way that you go about modeling. Let's talk about that. So does this happen in other places? You know, help me, help me understand that process a little bit.

Speaker C: Yeah. So what you're doing, when you build a model, you've got some sample of things that have happened. And let's use an example for hiring. I think that's a really easy one where like this person worked out, this person maybe didn't work out what you're doing in modeling and saying, I've had all of these examples in the past, these people that I've hired, and some have done really well and some haven't done so well. What were the conditions under which they succeeded? What were the conditions under which people didn't succeed? Who did I hire? What were their degrees? What was their experience, whatever it is. And you're trying to say in the past, what's going on and what can I learn from that? And so at the most basic level, what I see people doing is saying, this is good, we want to hire people with more education. And what that's assigning is that education has a positive effect on who you're going to hire. Maybe in culture that's time off has a positive effect on how people feel. Maybe it's that more time in the office has a positive effect on how people relate to each other. I'm not sure. But what we get to in predictive modeling is we get out of that basic level of things. And as you build more accurate models, models that do a better job of predicting on things they haven't seen before. We call those a holdout sample or a test set. What goes on after that is that you start to break out of this is good and this is bad, and you start to say, maybe, well, this is kind of good a lot of the time in this specific context, or this is actually not good in these contexts. And it's very good in this other context. And so it breaks down the rigidity of thinking that says, I should only hire somebody with a master's degree because more education is good. And it takes you to. I want a really dynamic thinker that might have a lot of overlap with people who have master's degrees, but I might be able to pick somebody out in a blue collar area who's really, really good at problem solving. I think that plumbers would be really good at a lot of white collar work because they're doing so much kind of dynamic problem solving on the fly, and then they're having to explain it to somebody who knows nothing about plumbing. I don't see any white collar firms reaching out to plumbers and saying, hey, help me solve my problems. And so that's a kind of rigidity of thinking, right? As we're saying, I want this amount of education because this amount of education is good. Models that are built like that aren't so good at predicting outside of very, very specific sets. And so if we learn where our models good at predicting, it's when they break down rigid thinking and when they go into things that maybe don't make any sense to us. And we take those and we say, well, I don't know why exactly, but this is working. This is what's going on. I want to do these things because these are the hires that are working out. I want to do these things because this is when my culture feels good. We can take that advanced modeling concept and just kind of say, you know, I want to test these things, I want to learn from these things and then continue to cycle and say, all right, I've tested them. I had this hypothesis. I thought this was really good. Did it work and not be so married to the ideas, but be more engaged with the outcomes than the ideas that lead us to the outcomes?

Speaker B: Yes. You know, when you talk about hypothesis, I've been bringing that word up a lot as it. As it pertains to culture, design, people, ecosystem design, things like that. And so when I hear hypothesis, I think historically we think of it as assumptions. You know, we make assumptions whether it's the college degree or the location or the experience. And you're breaking that back to almost that childlike mindset of, hey, I have an assumption, I have a hypothesis, but I'm not going to be married to it. And that's just great. How can you help people test those assumptions and not just accept them as truth?

Speaker C: Yeah. So the first thing that I ask folks when I have culture conversations is what are you assuming about your culture? What are you assuming within your business that you think absolutely needs to happen, that every single employee has to do for this to work? And do you have any evidence to say that that's true? So when I ended my time at the company you and I were working at together, I was working a 24 hour work week. I had gone down from 40 hours. I have a lot of other priorities in my life. You see speakers, a guitar, there's probably a microphone in frame. I love to play music. I wanted to spend more time doing it. And so I came to my bosses and I said, look, I want two more days to do this. I would love if my salary stayed the same. And I don't blame them for not wanting to do that. But what ended up happening was that, uh, at 24 hours, I was just as productive as I was at 40 hours. And there is a large assumption that goes on in the corporate world, which is that more hours means more productivity. And this basic case for me kind of disproved that. That was generalized truth. And so going back to your original question and not talking about the work week, anything that's happening for every single employee is implicitly an assumption. You're saying, I want this to go on. You need to come to the office. It's okay if you work remote, whatever it is, having performance reviews once a year is okay. However you set this up, whatever your structure is, there's necessarily going to be a few assumptions. And if you can see what's the same across the board, or what do I know to be true? I like to ask that question, what do you know is true? And then you go back and you say, how do you know that? And a lot of the time it's an ideological proof for. Well, it makes a lot of sense that this is true, doesn't it? Don't you see how this really ought to be true? That's kind of just not how the real world works. So if you can think and reflect with yourself, what do I know to be true about my business and my culture? And then you can take a step back, maybe write it down so you can review it. And it's not just you talking to yourself, it feels a little bit more like another person. You can assess then and say, what am I assuming? And this is not obvious. Right. Like I go through my own life, I have tons of assumptions that I impose upon myself that I'm constantly trying to break down throughout. Like my day is a lot of why do I think this is good? Am I sure this is going to work for me? And instead of assuming it's true, saying I think this is true, how would I test that? And going through with that test and seeing if I'm right and if I am, hooray, and if I'm not, well, then I've learned something. I can do a better job.

Speaker B: Yes, it's so true. And we get in this, you know, we talk about even just bias. I mean frankly, we all have it. So adopting this type of hypothesis and being curious and malleable to that, that the results and the outcome is so important, particularly when we're talking about people, because everyone's different. And that's why they even talk about like employee engagement surveys. Well, are they really a good indicator of uh, how you're doing an organization? Maybe not, because it, because it's, it's really based on a more holistic view of the organization and not, not actually leading indicators of what you're trying to accomplish. So I, I, I love this philosophy and how that's kind of supporting that particular hypothesis that employee engagement surveys may not be the thing, but they're okay to, to measure lagging indicators, so to speak.

Speaker C: Yeah. And I think I want to say a little bit about the engagement surveys. You might have an outcome that says four out of five stars on average. People said we were doing pretty good at their work hours or how the office feels or just their general satisfaction with the company. And four out of five stars could mean, let's say you have 100 employees. That could mean that all 100 of those folks said four out of five stars. I'm super satisfied. It could be that you had 80 fives and a few ones and twos and zeros. And those are two drastically different looking cultures. And you have 80 really, really happy people, 20 who are just like kind of having a bad time. May, that's not such a good culture. And we use these indicators at a kind of average level. And I love what you said, everyone's an individual. We don't look so well at the individual and we don't quantify these distributions very effectively. Modeling might be how do I get the average? But on the basic, it Might also be how do I avoid any one and two stars? I might rather have three and a half out of five if everyone was above a uh, three than four where 20 people are just really, really struggling to get through the day.

Speaker B: Yeah. Uh, no team member left behind.

Speaker C: Uh huh.

Speaker B: If that is absolutely the modeling of look. And you could do multiple models. Yes. We want the average of people to be a certain level of engaged in this particular or reach this outcome. But we also want to run another model. Where do things break for one person? Where, where are we missing that? Where are we losing that individual on the west coast that wants to play their gu, you know, two days a week, you know and I think that's a very worthy endeavor when it comes to programming is not just the average or the law of averages, but also for the individual.

Speaker C: Absolutely.

Speaker B: Yeah. And so if you're someone sitting in, in that people operations, I have a lot of people who are either in people operations, hr, you know, you and I work together. So if you're imagining me at uh, at that company we work together with. How do you get started with this?

Speaker C: So the first thing I want to do, honestly, personally I really believe in the hours thing and I want to mess with people's hours. I want to give my company a competitive advantage and I want to say here's what I need from someone throughout the week and we're going to test this. I want to be open about testing. I think that a lot of leadership teams have a tendency to say we're going to do this behind closed doors and we're going to do our test and then we'll present the results of the test and look really smart. I would much rather tell my entire company, hey, we're trying to do a good job and this is what we're testing and this is what we're going to look at. And I would say we're going to set general expectations throughout the week for what needs to happen. Maybe that's totally generalized. Maybe that's every employee has the same job and we just need this amount of productivity for them. Um, maybe it's one on one with your boss. And then I'm going to say work whatever hours you want to and I'm going to test this idea and say if you can accomplish this, I don't care how. Truly it does not matter to me how you do this. And I'm going to see maybe my business falls apart and I need to stop doing this. What I'm going to measure is do employees stay? Did I have Any resignations during this time period? Did I have an increase in enthusiasm? And that's often qualitative because I'm working the halls and I'm seeing what's going on. I would encourage people to be having qualitative conversations to see how folks feel, rather than operating a survey. Surveys can be valuable. That's kind of another conversation. But I would make a test and I would tell the employees, this is what we're testing. If it works, we're going to keep doing it. And so you've set up now an incentive structure where the guy who wants to play the guitar two days a week says, oh, this is really cool. I want to keep working at this company. They're telling me what we're doing. They're telling me we're testing a hypothesis, and they're telling me that I might be able to work one day a week if I'm really, really good at my job. And they don't mind because they're still getting the same outcomes. So we take a step back and we don't do my little obsession with ours. What I would want to do is I would tell the company and everyone in the company, all hands meeting, these are the things we want to improve on. We're going to start trying some stuff and we're going to maybe even solicit ideas from this group on what should we try? What's bad throughout your day? Can we just remove it? What's good throughout your day? How can we make more of that happen and then iterate on that process rather than just talking at all? Hands meeting, getting some feedback, saying what you've learned rather than trying to make yourself look really smart by saying how smart you were beforehand and then going forward and just repeating this process. I think there's nothing that can replace people watching you genuinely learn. And if you can explain it to somebody and you can say, look, we went down, we had these unlimited hours, and it was, you could have worked one day, we could work two. We had real struggles with productivity. People will respond to that. People will be able to say, oh, well, you do run a company. Like, I understand that you need to make money. And so I get why I have to come in eight hours a day for five days a week. Because we saw when we tried something else, it didn't work. But I think that what happens in these conversations, that leadership teams tend to say, we need to do this. It's really important. We can't test it. And then you have someone on the other side who doesn't really have any strong evidence that that's true. And so what I would want to do is give as much evidence as possible that what I'm doing makes sense. And if it doesn't, then I want to listen to that because I don't want to do something that doesn't make sense. I really, really don't want to do that.

Speaker B: Yes. And what that also does is gives your team members not just an incentive to test this out and iterate on it, but it gives them engagement. All of a sudden they are like, I have impact on the operations of this business. So not only am I incentivized about being SM efficient, about the way that I go about my role, maybe I am now triggered to find ways to more efficiently accomplish my role. Instead of counting the hours or time and seat and whoa. All of a sudden a, ah, company's innovation cycles are rapidly quicker, et cetera. They're not losing performance by trying these things out. If anything, they might be propelling their performance.

Speaker C: Yeah, uh, I love that word too. Incentives. That's big in economics. And so what I see with some companies is that there are a lot of rules or regulation, regulations or guidelines I don't often see. These are our incentives for good work. And so I think that if we can apply an incentive structure rather than a set of guidelines, you're just going to get what you want because it's better to do what you want. And making sure that doing good is rewarded by more good and that doing something that isn't so productive is rewarded by, you know, maybe some kind of a course change that's really meaningful rather than. These are the rules. Here we go. I think it's just two totally different directions to solve the same kind of problem. And one, to me, seems a whole lot easier and a whole lot more effective.

Speaker B: Exactly. A whole lot more fun. And frankly, you know, those are the type of cultures that, you know, we're trying to foster here. These type of curious, productive, you know, environments where people are doing things because they're engaged and they're, they're incentivized to do so. And that's not always monetarily either. Incentivized through freedom of work, flexibility, you name it, you know, whatever that looks like. But that's another hypothesis that you could take within your organization.

Speaker C: Exactly. You could test both. Right. You could spend three months doing. We have this incentive structure and maybe the incentive structure. I, uh, love what you said. It might not be monetary, it might literally be praise. In a meeting. When I was running a Team. My incentive structure was, I am going to tell you you're doing a great job in front of as many people as I can, and I found that to be far more effective than anything else that I could do. People want to hear that they're doing well, and so if that's your incentive structure, test it for three months. If it doesn't go well, set up some rules and regulations and see how that just be learning.

Speaker B: Exactly. Uh, that's a great way to end on this part of the podcast. So thank you for that, Jake. And now we get to move on to my favorite way to round out this podcast. And it's the rapid. Not really rapid fire questions, because they're. They're never quick, but they, you know, they're for fun. And so just whatever comes top of mind. So are you ready?

Speaker C: I am, um, ready. I'm excited. So.

Speaker B: Okay. Yay. Okay, good. So what is your favorite book, podcast or thinker that's shaping your current perspective?

Speaker C: Ooh, John Steinbeck. I was thinking about the Grapes of Wrath today. I love classic literature. I read a lot of stuff from, like, the late 1800s and the early 1900s, and I'm really interested in these people who had ideas, whose shapes still apply 100 years later. I think the Grapes of Wrath is just a beautiful story and that a lot of what it says about collectivism and people wanting to do well and do well for their families is incredibly relevant in a company in the country. And your personal life.

Speaker B: Brilliant. I am dusting that off. I'm going to have to read it. I think I read it many moons ago, and so I'll be dusting that off. Love it. Okay. If you could model anything about company culture, which you may have already talked about this a little bit in the past, actually, I think I already know your answer. But if you could model anything in a company culture, what's the first thing you would want to predict?

Speaker C: I would want to predict individual productivity, and I would want to tie it to ours, and I would want to see if there was a real relationship there. And my personal hypothesis is that over some barrier, there is going to be a positive relationship. You only work 30 minutes a week. It's going to be really hard to be productive. Maybe you're a genius. I don't know. But I would want to see who are our most productive employees. If we give them fewer hours, what happens? I love the idea of a hiring advantage. I think that setting myself up, um, if I'm running a company to hire people and say, look, you can go work five days somewhere else. You can work free for me. I'm, um, gonna pay you the same amount of money. Cause I'm getting the same amount out of it. That sounds really, really appealing. I'd love to get the best people. It'd be ours.

Speaker B: Yes. Awesome. And I really want you to run that somewhere and post the case study. Seriously, it's top of mind and that's actual data behind it. What is your main heads down work playlist these days?

Speaker C: Ooh, uh, I listen to full albums quite a bit. So I really believe in the artistic structure of the entire album and that the artist chose first, second, third, all the way to the last song. Pretty meaningfully. I love Blind Pilot's album. Three rounds in a sound. It's like soft folk music. A little bit sad. It's quiet enough that I'm not gonna be like amped up to do work, but it's still active. There's still a lot going on melodically where I can listen and be engaged and not get too distract. And it's just beautiful music.

Speaker B: That's fantastic. My album that I have to listen to from beginning to the end. I can't remember the album name, but Anathema, they go through a whole journey through. Through their album, so. I get it. I do. You're a purist.

Speaker C: I'm a little bit of a purist when it comes to the album.

Speaker A: Yes.

Speaker B: I should have known. Like, avoid the music questions when you are speaking to musicians or you're gonna.

Speaker C: Yeah, we could spend the next three hours talking about albums I care about, but I don't think that's really what you need on this podcast.

Speaker B: It's a different vibe, you know, like, it's fine. Just. We'll have a whole nother episode on that. Jake, this has been such a thought provoking conversation. Uh, I just knew we'd have a great light bulb moment conversation. I really hope that you know, my other listeners, all five of you out there, you know, get. Get some good information out of it, but would love your feedback. Jake, where can people find and learn more about your work?

Speaker C: So honestly, the best thing to do is just to email me@aerotermconsultingmail.com I would love to talk to you for free about any of this stuff. I love predictive modeling. I'll have a really good time. You can find me on LinkedIn under my name, Jake Brinsonhoff. I don't have a company website. This is all word of mouth. So shoot me an email or shoot me a direct message on LinkedIn, probably don't ask Marnie. I think she's got a lot of other stuff going on. But if you need a connection, that's a great way to go.

Speaker B: Yeah, no, please do. You know, I would love to introduce anyone that's in my, my network to Jake. And I mean, it's, if you want to get a lot out of a half an hour, one hour conversation, you absolutely will. And speaking to Jake. And like I have copious notes from my previous conversation. So if you're curious about how to bring modeling, how to bring even just that mindset, that's talk, leadership mindset and the way that they approach their philosophies and their cultures, Jake's the one to talk to. Thank you very much for coming on Jake. And I hope we can continue the conversation soon.

Speaker C: Thank you so much for having me, Marnie. I really appreciate it.