SDS 631: Data Analytics Career Orientation

This is episode number 631 with Luke Barousse of the eponymous data analytics YouTube channel. This episode is brought to you by Iterative, your mission control center for machine learning. 

 

Welcome to the Super Data Science Podcast, the most listened-to podcast in the data science industry. Each week, we bring you inspiring people and ideas to help you build a successful career in data science. I’m your host, Jon Krohn. Thanks for joining me today. And now, let’s make the complex simple.
Welcome back to the Super Data Science Podcast. We’ve got the tremendous Luke Barousse on the show today. Luke is a full-time YouTuber creating highly educational, but nevertheless, often hilarious videos on data analytics for his enormous audience of over a quarter million subscribers. Previously, Luke worked as a lead data analyst and data engineer at BASF, the global chemical manufacturing giant. Before kicking off his career in industry, he worked for seven years in the US Navy on nuclear powered submarines, and he holds a degree in mechanical engineering, a graduate qualification in nuclear engineering, and an MBA in business analytics. 

 

Today’s episode is a terrifically interesting one because of the terrifically interesting guest, so it might appeal to just about any listener, but it will be particularly appealing to folks in the early stages of a data career, such as those considering getting into data analytics. In this episode, Luke details the must-have skills for entry-level data analyst roles as demonstrated by an original data analysis project he carried out, the data analyst skills mistakenly and erroneously pursued by many folks considering the career, how his experience as a submariner prepared him well for a data analytics career, his favorite tools for creating interactive data dashboards, his favorite scraping libraries for collecting data from the web, this skills people can amass now to be prepared for the data careers of the future, and the benefits of CrossFit beyond just the fitness improvements. All right, you ready for this especially engaging episode? Let’s go. 

 

Luke Barousse, welcome to the Super Data Science Podcast. I have wanted to have you on the show for ages. I’ve been following your brilliant YouTube channel for months and months and now to have you not in the flesh I guess, but interacting with me on screen is such a thrill. Thanks for coming on. Luke, how you doing? Where are you calling in from? 

Luke Barousse:

Heck yeah. Doing well. Thank you so much for having me on. First off, when I first got into data analytics, I found the Super Data Science podcast and was just like, oh my god, there’s such a wealth of knowledge, so it’s awesome to come sort full circle and be on here, so thank you. 

Jon Krohn:

Wow, I’m delighted to hear that. 

Luke Barousse:

And calling from Arkansas in the United States. Sort of the middle of nowhere. I’m near the Walmart headquarters, no affiliation, but they have some mountain biking here. I like to do mountain biking and so that’s what I’m doing here. 

Jon Krohn:

The mountain biking isn’t affiliated with the Waltons in any way. 

Luke Barousse:

Oh, it’s very much affiliated with the Waltons. 

Jon Krohn:

It is? The mountain biking is? 

Luke Barousse:

They’ve poured millions of dollars into building the trails around here. The Waltons are big to mountain biking. 

Jon Krohn:

Really? 

Luke Barousse:

Yeah. 

Jon Krohn:

No kidding. 

Luke Barousse:

It’s their way to basically build up the city to entice the vendors of Walmart and people, employees to move to Bentonville, Arkansas. It’s building up the trails because now you have this great little ecosystem and so people are like, I want to go mountain bike there, and they move here and they have something to do while working for Walmart. 

Jon Krohn:

Wow. I had no idea. I was saying that as a sarcastic joke. 

Luke Barousse:

Yeah, not sarcastic at all. I appreciate there are millions of dollars they put into the trail and I get to ride for free. 

Jon Krohn:

Amazing. Your YouTube channel, as I’ve already mentioned, is incredible and lots of people think so. At the time of recording, you have over 260,000 subscribers. By the time this episode is live, I’m sure it will be many, many thousands more. Your target market with your videos is primarily aspiring and entry-level data analysts. In the channel, you creatively combine amazing different kinds of videos. Some of them are step-by-step instructions for doing things. Some of them are hot takes. You’ve got engaging visuals and my favorites are the humorous skits. Sometimes you do them with other well-known YouTubers and I just get such a kick out of them. I love them. So Luke, you have an innovative channel. I’ve never seen anything like it in the space. Where do you draw inspiration on for your next video? 

Luke Barousse:

For inspiration, well, it comes away with anybody, but mainly I try to go back to whenever I first got into data science and more specifically, data analytics. I try to think of the questions and the issues that I had when I was first starting and what resources were not there, because there’s a lot of great YouTube videos out there, especially on tutorials. I used to do a lot of tutorials. And that’s great, it’s like, hey, I created Tableau tutorial, but when I first got into data science, data analytics, I didn’t know what Tableau was. I didn’t know what visualization softwares were, what the need was, at the time I was going to school for business and I knew I wanted to apply analytics and somebody came in and gave a presentation at the time on Tableau and really helped open my eyes to it. But that presentation, that’s not normal to understand what this software is used for. It’s really hard to just get a concise understanding of what different tools are used to accomplish jobs. 

Jon Krohn:

So you’re providing context. You’re kind of thinking if somebody came up to you and was like, hey, I’ve heard about this data analytics career, where do I get started? What are the different kinds of things you do? Those are the kinds of videos you think about, you’re going to provide some context for people. 

Luke Barousse:

Yeah, and with that, probably a bad analogy, sort of a gateway drug, but basically make it as easy as possible for people to understand things they need to get into first for this field so that way they can decide, hey, maybe this isn’t for me vice, they don’t even know data analytics and then they spend all this time learning Tableau once again, spend all this time learning Tableau and then they start implementing and they’re like, I don’t really like this. Instead, look at it holistically. Take a step back and provide a big picture look at these things so that way they can get more of this big picture view to figure out if they want to go into it and then what they need to actually focus on. As you know, there’s so much you can focus on and so much you can look at. I think you need somebody to just provide that basic understanding and big picture of what’s happening there. 

Jon Krohn:

I love that. What’s the most gratifying outcome you get out of having this channel and these enormously popular videos? 

Luke Barousse:

Yeah, so I was actually just thinking about that the other day. I’ve been doing this I guess two years now, which I mean I guess isn’t that long. But man, I can tell you usually, I’ll finish up a video and then I’ll go to upload it and then I’ll upload it for the next morning and the night before I upload it’s like a kid whenever Christmas is the next day and you’re all excited or whatever. That’s how I get with these videos because I’m like, I don’t know what people are going to say. I don’t know, but I’m looking forward to, I just spent two, three weeks making this video. I’m pumped to release this thing and I’m excited. Sometimes I can’t even sleep. I’m ready to wake up the next morning and just check the comments and see what people are saying. That’s really what invigorates me is that cycle of getting to that point of releasing it and then seeing everybody, seeing the benefit that it has. Yeah, that’s mainly just what gives me energy. 

Jon Krohn:

Love it. And then in particular, your comedy skits. Where do those ideas come from? They’re really innovative. 

Luke Barousse:

Yeah, well so I mean, I’ll be completely honest, I didn’t originate those, I mean they’re obviously on YouTube, but I follow this other YouTuber called Julie Nolke and she has a famous YouTube video, it’s called Explaining the Pandemic to my Past Self. Basically the pandemic’s happening and you explain it to your past self before and it was sort of a skit talking to each other. Anyway, it’s a funny little skit and I’m like, oh she doesn’t have any people, she’s just using herself and talking to herself. I’m like, I could do that but with data science and that’s what I try to do a lot with my YouTube videos, just go about a different way of explaining concepts to people.
It’s actually pretty interesting. I don’t do the skits as much because sometimes they don’t resonate as much. But it’s very interesting, I found that this was my data analyst versus data scientist one. I received a lot of comments around one, new people like hey, I don’t understand this but I’m looking forward to whenever I can understand it. Then I’ve had people that have been like, hey, I’ve been studying for six months now and now I can understand the jokes in this video. This is awesome. It’s funny because it can provide I guess a goal for people that are new to this field of maybe some inside jokes that they should be able to get. 

Jon Krohn:

Yeah, I get that. I suspect that there are some listeners of the Super Data Science podcast out there for example, that are interested in the field. They’re interested in data science and AI and sometimes we have episodes that get quite technical but they listen to those episodes anyway because they’re just kind of understanding the way that people use the language, what are the trendy words. Even if you can’t really dig onto the meaning of it, just hearing those conversations happen. I guess if you were in an office and there were some people who you were thinking about moving into the kind of role that they have and you kind of hear them talking over lunch, that might be an interesting conversation to hear even if you can’t totally understand everything. But then as you explore the field more and more, you get the inside jokes, you get all the context. Yeah, that’s cool. 

Luke Barousse:

Yeah, I mean you got to fake it till you make it. I like how people can use it for that and basically just like, hey get exposed to these different topics and maybe in a fun way, podcast way, whatever. Over time, I mean if you’re exposing to it enough, you’re going to get it. 

Jon Krohn:

And that, especially I think if you listen to a lot of episodes of this show or you watch a lot of your videos, people will start to see, oh SQL that comes up all the time. 

Luke Barousse:

Right, I probably should focus, pay attention to this. 

Jon Krohn:

One of your most popular videos outlines a learning roadmap to become a data analyst. So, sounds kind of the perfect video you’re describing for providing the context to people. Do you think that a lot of people get lost or overwhelmed as they start getting into this data analyst career? Do you think it’s overwhelming looking at… You’re kind of describing earlier the situation where people might not know what Tableau is and they look up what they need to know as a data analyst and all these videos come up about Tableau and there’s Tableau tutorials and they’re like, I don’t even know what this is. They press play and they don’t understand what’s going on, and so it’s kind of overwhelming. I guess that kind of situation must happen a lot with people, right? 

Luke Barousse:

Yeah, I think it’s a lot. I actually did a bunch of interviews with my subscribers and for those that are entering the field of data analytics and what their biggest problem is in it. There was two main things. One of them was being overwhelmed and the other one was not really having a lot of social interaction, basically because we’re in the pandemic, not in the pandemic, we’re through the pandemic, but during the pandemic you didn’t really have a lot of people and a lot of interaction.
But the overwhelming aspect, yeah, just exactly to your point, there’s so many different things, where do you focus? You go to these job descriptions and you they’re like, you want to be an inspiring data analyst? All of them are saying different skills that you need to notice. You’re like, what do I need to learn? And so that was really the thought of the process of that video was to basically narrow down what you should focus on because I had a lot of problems. I learned tools that I think I’ve wasted my time on and I was like, I wish I would have just had somebody who had been like, hey focus on these core tools. 

Jon Krohn:

Do you have any examples of tools that you wasted your time on, off the cuff? 

Luke Barousse:

Yeah, I mean I’m going to bad talk it right now, but Microsoft Access, I can’t stand. You’ll have to bleep the cuss words that are going to come out of my mouth and I trash talk it anytime. I cannot stand that program. I mean, it’s an outdated program. Microsoft’s trying to phase it out but companies are still using it and they can’t get rid of it. I’m like, don’t learn this. Anyway, I need to stop right now. I’m so sorry.
But I’ve also learned, I don’t even want to give a shout out to the name of this company because they upset me so bad, an automation tool to extract data, and their whole premise was basically locking businesses in to paying a subscription fee to use this tool to web scrape data. I found out later that I could just use Python even more easily because it was supposed to be a low to no code solution and I actually was having to use code and so it was like, oh man, I wish I had just known Python, I would’ve used that. Yeah, those two situations, my blood’s boiling right now, would come to mind and I’m just like, I’m so glad. I wish that somebody would’ve been like, hey, focus on these things. 

Jon Krohn:

Yeah. What Python libraries do you like to use for web scraping? Beautiful Soup? 

Luke Barousse:

Yeah, Beautiful Soup, Selenium. 

Jon Krohn:

Pros and cons between those two? 

Luke Barousse:

I couldn’t give you that. I can just say I like Selenium mainly because usually I’m doing clicking or something or need to scroll on the page, so I need that sort of browser atmosphere and yeah, I’m just used to it so I just stick with it. 

Jon Krohn:

This episode is brought to you by Iterative, the open-source company behind DVC, one of the most popular data and experiment versioning solutions out there. Iterative is on a mission to offer machine learning teams the best developer experience, with solutions across model registry, experimentation, CI/CD training automation and the first unstructured data catalog in query language built for machine learning. Find out more at www.iterative.ai, that’s www.I-T-E-R-A-T-I-V-E.ai.
Cool. Well, we’re going to talk more about your favorite tools later in the episode, but let’s kind of get back to where we were in the conversation about people getting started in data analytics. What is the most significant assumption you think aspiring data analysts make when planning out their careers? I guess a kind of interesting twist on that question, what are the assumptions they make that are misplaced about a data analytics career? 

Luke Barousse:

Alex The Analyst, I think was talking about this, so I don’t want to steal his idea, but I’m basically going to steal his idea. I saw he made a LinkedIn post about it recently. I think people think that you have to go, especially for data analytics, data analysts, they want, an employer needs basic things done with data analytics, whether it’s Excel, SQL, Python, R, whatever, just basic tasks done to analyze data. I think one of his recent LinkedIn posts, he was talking about people present this portfolio for an entry level data analyst and it has all this machine learning algorithms and advanced methods of using gradient descent and stuff like that.
I mean, that’s great, but I need you to analyze this budget that I have right here and maybe provide some deficiencies. I don’t think we’re going to use a machine learning algorithm on it. I think people, one, they get a little too overwhelmed and they can think that they need to know stuff that’s way more advanced than they actually need to and that they just need to focus on the basics and understand that and display those skills in order to land an entry level job. 

Jon Krohn:

Cool. Yeah, that’s pretty similar. We recently had Shashank Kalanithi on the show. You probably know that. 

Luke Barousse:

Love me some Shank. 

Jon Krohn:

Yeah, so he was in episode number 623 and that was something that he said too. He was like, for data analytics, for an entry level data analyst role… I might not be able to remember totally off the cuff here, but he was saying things like, you’ve got to have some understanding of BI tools, maybe something like Tableau. He was like, you can learn Tableau in a weekend and then spend a couple of weeks learning simple SQL queries, be able to do some relatively advanced Excel stuff like VLOOKUP and you’re set. On the job, then you can maybe learn some Python that’ll help you automate some things. But he was like, you can be prepared for a data analyst career in a couple of months if you focus on it full time. 

Luke Barousse:

Yes, so true. If you could just focus. I like how you talked about those tools specifically from him because that’s exactly what I would recommend. My first job, I just knew mainly Excel, but then I got into it, got the job, and then I started to learn Power BI and implementing Power BI in my job. Then my next job, I started implementing SQL and also had to do Tableau, so I started learning that. You can sort of snowball and learn more tools as you progress in your career, but you have to start first at those core tools to even get the job. 

Jon Krohn:

Nice. Tell us about Power BI. That’s also a Microsoft tool, right? So you mentioned Microsoft Access earlier. I don’t want to get you fuming again. It sounds like Microsoft can build tools that are useful. I mean obviously Excel famous, probably the most widely used data analysis tool on the planet. What’s the difference? Why would somebody start using Power BI instead of Excel? What does it do differently? 

Luke Barousse:

I think so the ability to actually build succinct dashboards that people can’t mess with. When I first started, I made one of my first roles as a data analyst, I made this dashboard in Excel and you can make dashboards in Excel but one, they’re sort of clunky. Two, but then how do you share this Excel dashboard? You have to send the file and then what happens whenever you have to update it, you have to send the new file. I guess you could host it, you could host it on Excel online, but then you can mess with your cells. You could mess with stuff. That’s why I like Power BI. Power BI is a dashboard solution, provides visualizations, and I can use this within Microsoft, this Power BI service make this dashboard that people can’t mess with and well, that they can play with but they can’t- 

Jon Krohn:

You can give them knobs and drop down menus. 

Luke Barousse:

Slicers yeah, clicks, whatever they want to do. I can set up this controlled environment for them to go in, use the data, and then this thing, this dashboard can connect to a variety of sources, which may be Excel files, it may be a SQL database or whatever. That’s why I’m like, oh I’m really bullish on Power BI. I just love it. I mean, Tableau’s just as good. I don’t think either one’s different, it depends on what the company’s using, but I’m a fan of Power BI and I’ve made a couple of courses on it, so I’m just like, yeah, I’m very biased towards it. 

Jon Krohn:

Super cool. We haven’t talked about Power BI very much in the show at all, so I’m delighted that you can provide us with a bit of an introduction to it. Thanks. 

Luke Barousse:

Yeah, and I guess the only other thing that I want to add to that is the one problem that I have with Power BI still… So Tableau’s great for visualization softwares, you usually have Tableau and Power BI. Tableau has this thing called Tableau Public and you can go and share your dashboards very easily on Tableau public. People can go in and look at your dashboard, play with whatever data it’s connected to, and it’s really great. Microsoft Power BI doesn’t have anything public facing like that, so if you’re within an organization, you can use your organization’s Microsoft Suite and share it internally. But if I wanted to build a dashboard, me without an organization and to be able to share that dashboard, I can’t really do that easy and so that’s what one major drawback and I want to just capture on. 

Jon Krohn:

Nice. Super cool. All right, so we’ve talked about some assumptions that people might make early in their data analyst career that are maybe misled. How about you Luke? What are the biggest mistakes that you made early on in your data analyst career that you wish you could go back and do differently? 

Luke Barousse:

Yeah, so I think that one of the biggest mistakes besides not picking the right tools is then once getting into my job itself or any project if you will, I think one of the biggest problems I had was focusing too much energy on a specific task or a specific project before getting input from other stakeholders. I guess an example of that was in my first job, I’m building this product and I built out this, talked about this Excel dashboard, built out this Excel dashboard and I was like, oh, we’re going to do so many different metrics on it. We’re going to be doing standard deviation, variance, all these different things. I spent weeks on this thing and then I give it to my boss and it did have the information that he wanted, but unfortunately there was a lot of information and a lot of time that I spent adding other things that wasn’t any value add to the project and didn’t provide any insights.
I wish I could tell you that I only did that mistake once, but I did that in following projects. 

That’s why I want to bring it up now because it happened more than once and it’s mainly, hey, you work on something, get insights, get quick feedback. Even if it’s not a final product, it’s always good to have those stakeholders or whoever you’re building this for, to look at it and maybe provide that feedback to make sure you’re going on the right track so you’re not going to waste unnecessary time. I did learn some more skills while doing it, but I didn’t get to where I needed to as efficiently as I could have. 

Jon Krohn:

Nice. So, get feedback early and often is a tip that you wish you could give to your past self, just like the woman that inspired you with that Covid video talking to her past self. 

Luke Barousse:

Right, exactly. 

Jon Krohn:

In addition to your data analytics career, you have had some tangential entrepreneurial undertaking. In 2019, you founded a lifestyle startup called Macrofit to improve health through meal prep. This is something that resonates deeply with me. I am a big meal prep subscriber, I subscribe to two different meal prep services. I guess I might as well mention them on air so that people are aware if that’s something you’re thinking of. To me, this is a really obvious thing.
Before we even started recording this episode, Luke and I talked for about half an hour about CrossFit. He and I have very much been drinking the Kool-Aid. I don’t think it’s an unfounded drinking of Kool-Aid. If you are interested in forging elite fitness, there’s no easier way than being disciplined about going to CrossFit. It’s a key thing. It’s easy if you’re disciplined about doing it.
The thing that makes it so easy is that you have a community that propels you when you know that you’re going to be doing that workout against someone who’s maybe been doing about as long as you, they come to the same class time as you, you want to beat them a bit. 

But you also want to celebrate your progress with other people. You want to cheer on other people’s progress, and so just having this community, and it’s also been an amazing place for me personally to meet tons of great professional connections because unlike a lot of gyms out there, a lot of other non-CrossFit gym experiences I’ve had, people put their headphones in, they just kind of watch TVs that are on as they sit at a machine.
But when you go to a CrossFit gym, a lot of CrossFit gyms you’re not even allowed to wear headphones. You certainly can’t in class time. You’ve got to be paying attention, everyone’s in sync. You get to meet a lot of people. You show up at the same class time every day, you’re going to make a lot of friends, you meet a lot of professional connections. It’s just such a wonderful community and then once you start doing that for a little bit, you’ll start to realize that people in the CrossFit community, it’s not just about the fitness, it’s about so many other aspects of your life that support fitness.
Getting a good night’s sleep, not drinking too much, stretching and things like diet. Diet is super important to being able to grow muscles, to be able to expand your cardiovascular capacity, to have energy throughout the day, especially if you are working out in part of your day. So super, super important to have great nutrition. 

For me, it’s been a no brainer for many years that the easiest way for me to get the macronutrients that I need… So there’s these kinds of rules of thumb, you should eat as many grams of protein per day as you have pounds of body weight in order to sustain or grow muscle mass.
Then, depending on how much you’re working out, you’re also going to need carbs. Sometimes people will be like keto, it’s hard to be a serious athlete and be keto because you need carbs as fuel for your workouts. Depending on how much you’re working out, and you might want to work with a nutritionist or a CrossFit coach or somebody to give you some guidelines on how many grams of carbs, fat, and protein you might want to get in a day. But you can do your own meal prep.
One of the most time consuming ways of getting all your macros is to be weighing all your food. Maybe you spend your Sunday doing meal prep for the week. Some people even they’ll get groups together where you’re like, okay, every fourth week you’ve got to do meal prep for four people including yourself, so people will do things like that. For me, I was like, there’s meal delivery services. They tell you exactly the macros on each meal. I get this pre-weighed for three meals a day, six days a week. I’ve got pre-weighed meal containers that come from two services I use. These are both US-based services, but Territory and MegaFit Meals. Territory has a bit more vegetables, so I like that, and they’re really, really yummy. 

Luke Barousse:

I might check that out. 

Jon Krohn:

Yeah, Territory’s good for that. But Territory also, there tends to be a bit more fat in the meals. It’s hard for me to hit my lower fat targets on the macro plan that I’m on. There’s this other company, Megafit Meals, which the meals aren’t as yummy and they definitely don’t have enough vegetables, but in terms of hitting your macros, they have tons of meals with five grams of fat, 50 grams of carbs, 50 grams of protein. So, a few meals like that in the day and I’m off to the races with hitting my macro targets for the day. All of that lead up was to say that you also, Luke, created Macrofit to improve health through meal prep, so could you elaborate on this particular business idea? What was it about? What motivated you and what did you learn? 

Luke Barousse:

Yeah, so actually this has a lot of synergies, I think with your first point about community with CrossFit, sort of like we’re doing here in data science, you want to have like-minded people to understand what everybody else is doing for data science, where you should be going and how to improve your career. Same thing with I think CrossFit and you have a community to know where to go. That was one of the problems that I noticed was that this meal prep and specifically this macros, it’s very difficult. Macronutrients, like you explained, this protein, carbs, fat, you have certain amounts that you need to do or you want to hit and they all make up how many calories you get in a day. If you’re below or above your certain calorie target for the day, you can either gain or lose weight.
These are numbers that I’m talking about. Grams of these macronutrients or numbers of calories, these are numbers. This is data, so I’m like, I could turn this into a data science project. Like what you’re talking about, oh I went over fats because I have too much fats in this one meal. I’m like, okay, I want to make a solution that can basically tell me what I need to eat. I know I want to eat three meals a day, tell me with each meal one, the macronutrients I need to hit and then also maybe make some recommendations of what I need to eat in order to hit those macro nutrients. 

 

That’s what my goal was. I built this in Microsoft Excel, which started as a project in college with a bunch of other people, and then I took it further after that class. I want to make sure I give credit to those people I worked with initially in my school project. We actually used macros like Excel macros, Microsoft Excel macro, so we called the program at the time Macros for Macros because we used Excel macros. One of the girls in my group came up with that title. I’m like, oh yeah, we got to use this.
Anyway, that’s all we really did. We built this Excel sheet where it was like, hey, you know what your current weight is, what weight you want to get to, so you could calculate, you do the calculation in Excel of how many calories you need to do in order to lose this weight over how many weeks. 

Then from there, we broke down the calories further with Excel into different macronutrients and then I downloaded, I went to the USDA or whatever the food place is that gave the data for foods and their caloric and their macronutrients. I got all that data and from there we built basically a simple little algorithm that basically would tell you, okay, this meal, I need to eat one cup of rice. You need to have five ounces of steak and you have a cup of veggies and maybe some sort of fat, like a half avocado.
It would tell you every single meal what you needed to do. That was the data science project. It was so crazy because I got to use Excel, use this data in sort of a different way. I wasn’t doing analytics, but it was using data and calculating what you needed to do. That got me into, after the project was done in that I wanted to, I was like, I can make this into an app. That once again gets into tools that I shouldn’t have learned as a data analyst that I did. But I got into learning a web framework with Python into learning Django. 

Jon Krohn:

Yeah. 

Luke Barousse:

I don’t know why I pronounced the D. Django. 

Jon Krohn:

I was like, is that how you pronounce Django? 

Luke Barousse:

No, it’s not. I mispronounced in one of my videos. What’s that? 

Jon Krohn:

Oh, really? You mispronounced it in your video? 

Luke Barousse:

I mispronounced it in one of my videos. I was like, oh D-jango, and everybody was like, this isn’t the Quentin Tarantino movie. It’s Django. And I’m like, oh, okay. But during the pandemic or whenever, when you don’t have people around you to talk to you say things that you don’t know because you’re not talking to people. 

Jon Krohn:

Yeah. What do you think about the Super Data Science Podcast? Every episode I strive to create the best possible experience for you, and I’d love to hear how I’m doing at that. For a limited time we have a survey up at www.superdatascience.com/survey, where you can provide me with your invaluable feedback on what you enjoy most about the show, and critically, about what we could be doing differently, what I could be improving for you. The survey will only take a few minutes of your time, but it could have a big impact on how I shape the show for you for years to come. So now’s your chance, the survey’s available at www.superdatascience.com/survey, that’s www.superdatascience.com/survey.
One big one for me for a long time was the word awry. 

Luke Barousse:

Yeah. 

Jon Krohn:

I didn’t map saying… I would say awry verbally to people. I didn’t know how to spell it. I’m talking until my late twenties. I didn’t really ever think about how you’d spell awry. Then whenever I read it in a book, I was like, “ori,” and I kind of knew what it meant. I didn’t really realize that they’re the same word. So, I get that but in terms of a podcast, you actually mispronouncing Django as D-jango helps people find that tool, although now you’re not even recommending it. 

Luke Barousse:

But that’s one of the tools I was learning, because I was like, I’m gonna build this app and that’s not what data analysts should be doing. That’s the data analyst that’s submitting their machine learning algorithms to get this, it’s not something I should have done. I wasted a lot of time and effort trying to build this app that eventually just like I was like, Django is a really hard and really difficult web framework and I could have been spending my time on other things. Anyway, so I eventually abandoned it. But during that time I was trying to promote this and I was using Instagram and also YouTube to try to promote this application. That’s really how I got into social media was through this data science trying to promote this application via social medias. 

Jon Krohn:

Oh wow. That ties in really nicely to my motto, to my favorite saying Luke, which is, it’s a Latin phrase, it’s qui audet adipiscitur. I did a whole episode on it, episode number 610. It translates into English as, who dares wins. Where, by trying your project, by trying the Macrofit project, that got you into doing social media stuff that has now turned into this incredible YouTuber career that you have where you’re inspiring hundreds of thousands of people, millions of people. This whole idea is that when you do something, when you take action, especially if that action is audacious, then even if you don’t achieve the initial outcome that you envisioned, which is creating this nutrition app, great things will come from it. Who dares, wins. 

Luke Barousse:

That’s crazy. That sounds interesting. Yeah, I think that relates because it was just so crazy because it was just like, yeah, the goal was to build this app and then I started… I was from an Instagram, I was taking pictures of food, as silly as I might sound, but it was actually sort of fun and I enjoyed the social media aspect and I’m like, I actually like this social media aspect better than, and sharing these data insights, but I like this better than actually trying to build this Django app. So yeah, that’s crazy. 

Jon Krohn:

Yeah. Cool career transition, Luke. It isn’t the only super cool career transition that you’ve had due to taking on audacious challenges. It’s interesting because that who dares win quote was popularized in recent centuries by the British Royal Air Force. One of the most interesting things about you is that for seven years, you were a submariner in the Navy and in order to run a nuclear submarine, you were trained in nuclear engineering. That’s wild to me. That’s such a cool specialization to have about the world. Wow. Can you tell us a bit about that experience and did nuclear physics help in your data analyst career later? 

Luke Barousse:

Yeah, so I served seven years in the United States Navy, specifically in the submarine force, which all of our submarines in the United States Navy use nuclear power. Yeah, well it was a crazy time, just so much time spent almost close to two years of my life underneath water. It’s just like, oh, my twenties completely gone because, or not, I wouldn’t say completely gone. It was a great time, great learning experience, but so much time away from family and friends out to sea running a submarine, going places. There’s parts that I miss, especially the camaraderie and everything like that. But then there’s parts of I’m looking out of a window right now. I actually have a window and I can go outside later. Whereas I would go on this submarine, I would go underwater for three, four months at a time. 

Jon Krohn:

Oh my God. 

Luke Barousse:

Yeah, we just go for three or four months at a time. The only thing we’re really limited by, because it’s nuclear power, we can make our own power, we can make our own water. We’re limited by food. That’s whenever we have to come back to land. Actually, our submarine was not one of the bigger ones and so you’d have to… There wasn’t a lot of food storage and so you’d actually have cans of food on the ground and we put plywood on top of it and then you would eat those cans of food and then you can remove the plywood, but sometimes some of the areas that you walk through the submarine, it was even shorter than usual because you had these cans of food that you had to walk on. 

Jon Krohn:

Wow, that is so interesting. I had no idea. I kind of thought it would be three or four days at most. I thought, and then my liberal estimate, once you started saying three to four, I thought at most you would say weeks. Three to four months, that’s wild. How do you make the water? You take salt water and then you use nuclear energy to desalinate it? 

Luke Barousse:

Yeah, so they have two different types. I hope this isn’t classified. I don’t think it is. It’s pretty open knowledge, I think. You can either boil the water… I’ll give general examples. You can either boil the water, they’re called an evaporator and so you use this hot water and boil the water and it cleans the water. Or you use a reverse osmosis unit, which is more just powered through electricity to shock it and get out the clean water. We had the evaporator option on mine and you can only make so much and you can only make it so fast, so water conservation, they call it, they have submarine showers. You’re supposed to get in, bathe real quick, wet yourself, shower down, turn off the water, soap up, and then rinse off and then get out. You’re not supposed to waste water. 

Jon Krohn:

I mean that’s amazing that there’s enough to be showering at all. That’s wild. 

Luke Barousse:

Yeah, well, I mean the submarines are pretty, they’re pretty… Oh, I mean they’re such fastening things. I mean, like a football field long and just tanks go lower with freshwater tanks, sanitary tanks to store your waste, ballast tanks, so large tanks to ballast the ship up and down and it’s just… But I guess this sort of relates to your question of how does this relate to my analytical career.
A lot of that stuff I didn’t realize at the time, but we’re dealing with data at real time on a submarine. You have to analyze this data all the time. I had to work for a little bit in the engineering power plant, so I would run the nuclear reactor, me and a team of 10 other individuals, and we have to look at all these different gauges, monitor pressure inside of the reactor, temperature inside the reactor, look at trends, you have to plot it. There’s a lot of look at chemistry, understand the trends that are going on with the chemistry inside of it. 

 

Anyway, there’s a lot of numbers and so there’s a lot of understanding behind what’s happening with the data and where it’s going. It was just sort of second nature whenever I transitioned to a data analyst coming from that role because it’s like I’m surrounded by numbers all the time. Just now I have… I don’t want to do talk bad about the government, but now I have just better tools to analyze the data now because we were doing paper logs and the Navy has improved the paper logs recording these numbers for all these things and we’ve improved in that way. But now, I have advanced tools that I can analyze the stuff and do some more trend analysis. So yeah, there’s I guess pros and cons. 

Jon Krohn:

Super cool. Really delighted that we had that question to ask you because wow, I just learned some amazing things. Another project, another kind of career twist that you’re working on right now though firmly rooted in now your data analysis and data science and even your web scraping experience specifically, you alluded before we started recording to me about a job data collection project. Do you want to tell us a bit about that? 

Luke Barousse:

Yeah, so I started that last year. Well, so I have all these… I’m a data analyst obviously, and you have all these people asking me like, hey, what are the top skills? What skills should I be focusing on? And I could tell them, I could tell them what it is, what I think it is, but what’s the data say, I’m like, I need to get the data. So you have stack overflow. If anybody doesn’t know stack overflow is like a coding website and it collects all these coding questions. You can go on and see it. Anyway, they provide a survey every year that will tell people what are the most popular languages, what are the most popular SQL databases. It’s actual data of what’s important, but it’s more focused for developers and web developers. I mean there is a sprinkle of data scientists and stuff in there, but mostly it’s I think good development.
Anyway, they have data, it’s like solid data, you can look at it. There’s nothing like that for data analysts. It’s like, where do you go to collect that? How do you know what jobs require what skills? So I was like, I need to collect this for job postings. 

So last year, I built this Python bot because I didn’t want to use that one solution that I talked about from work that was basically a money Ponzi scheme that I feel. I used Python instead to go in to LinkedIn, which not really allowed but did it anyway, go into LinkedIn and scrape job postings of data science roles, data analysts, data scientists, data engineers, scrape all that data and then from there pick out those courses, just go through and pick out what skills and then from there, aggregate it to find out what are the most popular skills?
I did that for three months last year, running this bot every single day. I got a ton of job postings, a ton of data, and actually was able to go in and list like Excel, SQL being the top two skills, and then from there, the visualization tools and programming languages so Power BI, Tableau, Python and R, they’re all about equal. Those are basically the top eight skills that I found as data analysts.
Sorry, funny side note real quick because I was thinking about that today. I shared this and Excel was the number one from the data because I finally had this and somebody, multiple people commented, Excel’s not a core skill of data analysts, and I’m like, I’m literally showing you what the data says. Don’t be mad at me. I’m showing you what the data is. The data says Excel is the most popular skill. I don’t like it, I don’t like it either. I would much rather it be Python, but don’t be mad at me. 

 

Anyway, I ran into a lot of issues with LinkedIn trying to block me doing the web scraping, so I sort of halted it after a while and I’m trying to get back into it. I’m working with the company right now to try and get this data and actually start collecting it on a daily basis. What I would love to do to continue this project is continue to collect it over time, and you know how they have those races of programming languages over time. What’s the most popular programming language? You could watch it, oh Python, in 1992 it was nothing, but now it’s like, oh, it’s high up there. That’s like the goal. Collect this on a daily basis just indefinitely and start mapping out what is the most popular skills so people know as they can go and they can focus on the top skills for data analysts. 

Jon Krohn:

I know that you don’t have this developed yet, but just based on your experience, do you have any predictions as to what tools someone who’s interested in getting into data analysis today should be learning? What are the tools of the future that people should be learning today? 

Luke Barousse:

Yeah, it’s not changing very quick for data analysts, unfortunately. I don’t know if you feel the same, but I feel like we’re sort of slow moving in regards to that, in regards to technology. I mean, yeah, specifically for data analytics, I feel. I’m still pretty Excel and SQL, I feel like first two main ones and then a programming language and a visualization tool, and that’s what I would stick with. 

Jon Krohn:

What about, are there anything Python libraries or anything that you’re really excited about right now that you think people should get into or a tool or approach that you’ve recently discovered that you’re like, oh man, I really need to dig into that and learn more about it? 

Luke Barousse:

Yeah, I’m really… Yes, I have an answer for that and it really relates to data engineering because as a data analyst, that’s the biggest problem that I’ve found. This problem right here, I need this data of job postings and it’s not easy to get. It’s a data engineering problem. A lot of my subscribers are getting, they’ll get this Google data analytics certificate and they’ll actually go into data engineering because it’s such strong demand. So, tools around data engineering I think are invaluable to learn whether that be around how to manage databases, like Snowflake, Databricks or how to do orchestration. I like Airflow, which is a library of Python. I really like that language and think that’s so cool. Those type of things, that would be sort the fun things that I want to learn, but I would not, unfortunately, I wouldn’t be like, you’re an entry level analyst, you should go learn Airflow. I would be like, you need to focus on Excel and SQL. But if you want to continue- 

Jon Krohn:

It’s a great answer though for the question that I asked of what can they be doing to prepare for the future. I think this is something as data sets continue to get exponentially larger every year, increasingly, data analysts to be effective are going to have to be able to do some of their data engineering. Otherwise, a company would have to hire a separate data engineer to be providing data to one or more data analysts, which is an added expense. If they could find somebody that could do both, it’d be ideal. Actually, Shashank, in his episode also talked about how he thinks that the best data engineers are people who used to be data analysts, which is a transition that he’s made. He says that when you know what the downstream user is going to need, when you appreciate their concerns, you’re going to be better at extracting the data and providing those data to them. 

Luke Barousse:

I think once again, that gateway drug data analysts are like that. Once you’re that entry level data analyst, you sort of peek into data engineers, data scientists, or maybe a more advanced data analyst and you could then choose where you feel like you should focus more. 

Jon Krohn:

Nice. Well, then I think you actually have kind of answered the question that we had. Prior to you being on the show, I posted on LinkedIn and Twitter and asked if people had questions for you. Mattias Baldino, who is a BI analyst at a company called Brain Technology, he wants to know what you expect of your own future. Maybe you can expand a little, you can try to look in your crystal ball a little further in your career. But he says, “Does Luke see himself moving to a leading role or moving to a role more focused in data science?” He says, “You’re always giving advice to aspiring or new data analysts,” but he’d love to know your own thoughts about how you’d like to move forward with your data analyst career. We kind of already got a sense that data engineering is in that vision, but I don’t know if you have anything else you’d like to add maybe looking further beyond, where would you like to be in a decade? 

Luke Barousse:

Yeah, so I love data science, but as we talked about already, I love the social media aspect and being able to bridge people that’s really… I don’t look to be a CEO or a lead data analyst or lead data scientist. I want to learn about new technologies and I want to be able to share that with others and I think social media provides that platform for me to grow mentally and then also share it. That’s where I want to grow is through YouTube and like you talked about at the beginning, I’m focusing on entry level analysts, but I’d love to expand this to help more than just entry level data analysts in the future. 

Jon Krohn:

I love that answer and I’m so excited that’s your answer because it means that I get to continue to enjoy more and more Luke Barousse YouTube content in the future. I love it. All right Luke, so those are all the questions I’ve got for you. Before we end the episode, I always ask for a book recommendation. You got one for us? 

Luke Barousse:

Yeah, so I guess top book, actually I made a video on it, but for our data analysts and also data scientists and maybe data engineers, I would recommend Storytelling With Data with Cole Knaflic. She was on the Super Data Science podcast and it’s such an insightful book into how to communicate with data. Actually, funny enough, I read this years ago and this book was actually a huge motivator, not only for my prior social media experience, but this was a huge motivator for me to try to tell stories with my YouTube channel with data. Anyway, besides that I think it’s must read and it’s so easy to read and it just provides such immense amounts of detail and insights that I’m like, I can’t recommend it enough. 

Jon Krohn:

That is a cool recommendation. Cole Nussbaumer Knaflic, she was on episode number 395 and somebody, I can’t remember off the top of my head right now, so some listener out there was like, “You bonehead, Jon Krohn.” I know who it was, in a very recent episode somebody else recommended that same book. 

Luke Barousse:

Oh, okay. 

Jon Krohn:

Yeah, so really popular recommendation. 

Luke Barousse:

I wonder if Shashank did, because he also interviewed her on his YouTube channel I think, and she has a new book out, too. 

Jon Krohn:

Yeah, there’s a good chance that it was Shashank. I can’t remember off the top of my head, but yeah, I’ve had this conversation recently. So yeah, super cool recommendation and that was a very strong recommendation. Must read from Luke Barousse, and also recommended by someone else, possibly Shashank. Check it out, listeners. It was episode number 395 of the Super Data Science podcast that she was on. Sounds like maybe we should be having her on again because of her next book. 

Luke Barousse:

Yeah, probably. 

Jon Krohn:

Cool. Well, there’s a thought. All right Luke, well obviously people know that if they want to keep tracking your career after this, your YouTube channel is probably the primary place to do it. Any other places that people should be following you?

Luke Barousse:

YouTube mainly, that’s where I want to focus most of my effort. I need to beef up my crew and mainly I need an editor, but that and then LinkedIn is the main two places. I’m also on Instagram and TikTok. 

Jon Krohn:

Nice. All right, we’ll sure to include your social media links in the show notes for this episode. Luke, I had such a great time chatting with you today. It’s been such a fun episode. Thank you for coming on. It’s been an honor to meet you and I look forward to catching up with you again sometime soon. 

Luke Barousse:

Heck yeah. Thank you for having me on. Wish we could have talked more about CrossFit, but I’m glad we talked about data. 

Jon Krohn:

Wow. I loved chatting with Luke today. What a guy. In today’s episode, Luke filled us in on how Selenium and Beautiful Soup are his favorite web scraping libraries, how data analysts don’t need machine learning projects in their portfolio to get an entry level role, instead, they should focus on the essentials, namely Excel, SQL, a dashboarding tool like Tableau or Power BI and maybe a programming language like Python or R. He told us how his biggest mistake in his early career was not getting regular enough feedback from stakeholders on consulting projects. He also told us how his experience with real-time submarine data was foundational for his data analyst career, while trying to build his Macrofit data product was instrumental in him becoming a full-time content creator. He told us how familiarity with data engineering tools like Snowflake, Databricks, and Airflow will prepare you for the data jobs of the future. 

 

As always, you can get all the show notes, including the transcript for this episode, the video recording, any materials mentioned on the show, the URLs for Luke’s social media profiles, as well as my own social media profiles at www.superdatascience.com/631. Every single episode, I strive to create the best possible experience for you and I’d love to hear how I’m doing at that. For a limited time, we have a survey up at www.superdatascience.com/survey where you can provide me with your invaluable feedback on the show. Again, our quick survey is available at www.superdatascience.com/survey. 

 

Thanks to my colleagues at Nebula for supporting me while I create content like this Super Data Science episode for you. And thanks of course to Ivana, Mario, Natalie, Serge, Sylvia, Zara, and Kiro on the Super Data Science team for producing another magnificent episode for us today. For enabling this super team to create this free podcast for you, we are deeply grateful to our sponsors. Please consider supporting the show by checking out our sponsors’ links, which you can find in the show notes. If you yourself are interested in sponsoring an episode, you can get the details on how by making your way to jonkrohn.com/podcast.
Last but not least, thanks to you for listening all the way to the end of the show. Until next time, my friend, keep on rocking it out there and I’m looking forward to enjoying another round of the Super Data Science podcast with you very soon.