David Langer_mixdown_01.mp3

David: [00:00:00] There's no script ability, there's no reproducibility, there's no robust version control. I mean, all of that is absolutely true. I would never argue against that. Well, I would say is, yes, that's all true. And effectively it doesn't matter, right? It doesn't matter. As we all know in the technology field, it's not always the best technology that is dominating the market. So you end up working with a dominant technology.

Harpreet: [00:00:32] What's up, everybody? Welcome to the Artists of Data Science Podcast, the only self-development podcast for data scientists you're going to learn from and be inspired by the people, ideas and conversations that will encourage creativity and innovation in yourself so that you can do the same for others. I also host Open Office Hours. You can register to attend by going to Bitly.com/adsoh forward slash a. Ds0h. I look forward to seeing you all there. Let's ride this beat out into another awesome episode and don't forget to subscribe to the show and leave a five star review. Our guest today is an analytics consultant, educator, YouTuber and LinkedIn enthusiast. He's well known for crafting and delivering training designed to help any organization or professional develop those valuable and extremely necessary data analysis skills that are needed to thrive in our digital age. Over the years, he's educated hundreds of professionals in a live classroom setting and thousands more in his online courses and tutorials via hands on sessions that teach the artistry of data analysis through a combination of Excel, SQL and AR. He holds [00:02:00] a vision of the world where data analysis skills are as common as skills with Microsoft Office and evangelize the fact that the most important part of data is identifying the 20% of analytics that drive 80% of the return on investment. So please help me welcoming our guest today, the founder of Dave Von Data, David Langer. Dave, thanks so much for taking time out of your schedule to come on to the show today. Man, I appreciate having you here.

David: [00:02:27] Thanks for having me. Harpreet. I hope your audience will enjoy the content.

Harpreet: [00:02:31] You are no stranger to my audience. People know you quite well from all the wonderful input and all the wonderful advice you've given to the data science. Happy hours. Super, super grateful for you taking time out of your data to come to those man. Those are some of the best sessions having you around. Hopefully have you back in near future. But I just want to say, Dave, thank you so much for coming to those things. Really appreciated having you there, man. A lot of good advice and I'm hoping we get into some of your background because we've chatted during the happy hours, but we've never talked one on one the first time we ever talked or talking one on one. So I'm excited to get to know you a little bit better. Speaking of getting to know you better, let's talk about where you grew up and what it was like there.

David: [00:03:11] Oc Yeah, so up until the time I was about 16 years old, I lived in a number of different places, so I lived in the U.S. my entire life and I lived in Washington state for a while in Texas for a while in Idaho for a while. So in general, I tend to think about where I grew up as being the Seattle area in Washington State in the northwest part of the US, because I lived there for about 30 years, starting when I was about 16 and as you might imagine, Seattle. So if you're familiar with any of the Seattle stuff I lived through the nineties, through the grand jury, I wore flannel, I went to the Paramount. I saw grunge bands. I drank coffee at Starbucks. I did all the things that you would think, including working at Microsoft for a while as well. So totally steeped in the whole Seattle thing. It was definitely interesting because one of the things [00:04:00] that I experienced growing up there was the effect of becoming a technology hub on the local environment. So the economy, housing, traffic, all that kind of stuff. So that is that is something that I've kind of reliving now that I live in Bozeman, Montana, which is kind of exploding as well. We're starting to see some of the same things that I lived through when I lived in Seattle.

Harpreet: [00:04:23] Seattle is a cool place, man. I went there a lot in the early to mid nineties. At that point I was like ten years old or something, whatever. I went there a lot visit some family, but how about grunge? Dude grunge is like the type of music I loved the most when I was like in my teenage years. All that stuff Pearl Jam, Nirvana, Alice in Chains, all that Seattle grunge scene. So that's that's awesome. And like, did you get to see any of those bands live? Like, like, did you get to see Nirvana or.

David: [00:04:51] No, never. Never got to saw any of those guys when they were small because back then they played in clubs and you had to be 21 or over to get into. And then I wasn't old enough at that point when they guys were still small, but I did see some semi-famous bands live. For example, I saw Grunge Truck, which is like one of my favorite Seattle grunge bands, which most people have never heard of before. But yeah, we saw lots of bands, but not any of the famous stuff because those guys blew up and before I was old enough to actually see them in the clubs, unfortunately.

Harpreet: [00:05:19] So coming up, I guess as as teenagers, late teens, what did you think your future would look like?

David: [00:05:27] Oh, okay. I think you're going to love this story. So as you can tell from the gray and the beard, I'm a little bit on the right end of the age spectrum here. So I'm a child of the eighties. And I remember when a Brief History of Time came out by Stephen Hawking, and I went to a Waldenbooks at the mall and I bought it and I was reading it like, Oh, I read almost the entire book just sitting in the mall. I didn't even leave the mall because I was so fascinated with it. So for a time when I was a teenager and I was in high school, I thought I was going to be a theoretical physicist. That's what I was going [00:06:00] to do. I wanted to be like Stephen Hawking. I didn't want to research black holes and gravity and the nature of the universe and the gut, the Grand Unified Theory, all that kind of stuff. So I was super, super stoked about that. And then a gentleman by the name of Oliver Stone made a movie called Wall Street, which came out in the late eighties, and I saw that.

David: [00:06:20] And so I lived through the the leveraged buyout type of KKR and those guys and Michael Milken and all that kind of financial stuff. And I became absolutely fascinated with that. And I wanted to be Gordon Gekko, like from Wall Street. So I totally switched and then I became super interested in finance and all that kind of stuff. Then I'm like, Hey, I don't want to be a, you know, I don't want to be a poor theoretical physicist. I want to be like Gordon Gekko. I want to be a hood, have $100 Million and have my own private jet and all that stuff. And so that's what I finished up high school thinking I was going to do, which is why I majored in economics. And then of course, I didn't end up doing any of that because I didn't want to leave Seattle. I didn't want to go move to New York or Chicago and do that back in the early nineties. So I ended up working in I.T., working in technology, and the rest is history.

Harpreet: [00:07:11] Dude. Nope, nope. I'm not bullshitting you here. No, no bullshit. But that literally describing, like, my journey throughout high school, I checked out a brief history of time. I was probably a freshman, and I remember reading that book like just in a day or two. And this I was in, I was a freshman back in like 97, right? So Internet had just been kind of coming around and I would go on the Internet and research black holes and like all this shit about the universe all night long through the wee hours of the night. And it wasn't until I watched this movie. It wasn't the Gordon Gekko movie. There's another one called Boiler Room. Yeah, yeah, I saw the Boiler Room. And at that point, I was probably like a junior, maybe something like that. And that's when I said, I want to be a stockbroker. And [00:08:00] I went and majored in economics many years later because I took a detour in life. Bad stuff. I feel crazy, man. That's that's such a parallel to my interest. Like, how did you get into it? Stuff. Like, did you just, like, decide that this is what's going to happen? This is like the next trend. Did you see that or did you just like pick a book and start learning? How did that happen?

David: [00:08:23] So I'd always kind of been into computers. So I saw Tron, the Disney movie Tron the original one, when it first came out in the movie theater, and I was totally fascinated with it. And then a few years after that, my dad bought a Commodore 64 and my dad and I actually went to this is when we lived in Idaho, in Boise, Idaho. We went to the local community center and we took a basic programing class together, my dad and I, which is cool. So it was kind of like this bonding experience with me and my dad. So I've always been kind of fascinated with computers. And when I was in high school, oddly enough, while I was doing this flip flop from theoretical physics to being wanting to be Gordon Gekko, I still took AP computer science and took the AP computer science exam, so I was still into programing and that sort of thing. And when I graduated with my degree in economics, I was like, okay, well, I don't really want to work at a bank. I would rather if I'm going to get a job, I might as well do computers because one, I like it.

David: [00:09:14] And two, it's a pretty good career path. It pays pretty well. So I actually landed a job at an insurance company where I worked with actuary, the actuarial team for a while, by the way, as supporting them. And I know you've got some actuarial experience and I got into a program where I already knew how to do C and C++ and they trained me at the assurance company to be a COBOL programmer on the mainframe. So my very first job coding was actually on a mainframe, so I coded COBOL and JKL and Dial to 80. Like all the old school stuff, worked with IBM's and DB2 and all these old school technologies that most people probably don't haven't heard of these days. And that was really my start because I was like, I want to get into writing software. I want to be a coder. That's what I want to do because [00:10:00] I thought it was cool and I thought it writing code was cool for a very long time. So I just went that route because it was a it was a good career path and it's it's treated me well ever since.

Harpreet: [00:10:10] So what exactly is a mainframe like? Is it like in my mind, is this a giant computer? Like just a whole computer? Is that what it is?

David: [00:10:19] Right. So it's not what they talk about in the movies. Right. So you watch The Matrix, they talk about the machine mainframe like it's this. No, all the mainframe is is a very big computer. It's essentially just like a server. Think of a multi track intel based server running Unix or Windows or Linux or something like that. Same concept. But what made a mainframe special, if you will, is that it was built. They were originally built back in the days when processors were relatively unpowered and memory was relatively expensive and rare. And so they focused a lot on creating a multitasking environment that was very, very efficient in terms of using data, memory and storage and compute, and also extremely secure as well, because they were typically used by the finance departments of corporations to handle finance data. So they were they were also built to be very, very secure. But it's just basically a big computer as all it is.

Harpreet: [00:11:15] And so how did that interest in in computers and in programing like how did that lead you towards more towards that the data stuff, was that kind of like a natural path or how does that transition happen? If you can call it a transition, I'm not sure.

David: [00:11:30] Yeah, well, it all begins probably with three letters SQL. Yeah. So very early on in my career I had to learn SQL because before that I didn't need SQL in my Commodore 64. They didn't teach that in my AP computer science class in high school, and they certainly didn't teach it at university. When I went to university, they taught just straight up programing and C, C++, things like that. Assembler. But when you go work in the business world, there [00:12:00] are databases and there's data. So you actually spend a lot of your time when you're working on business type applications, business systems, working with data, worrying about data, understanding data, what's in the data? What are the tables? What do they mean? What are the columns? What are the values? All that sort of thing. And SQL. Sql is really where it kind of started for me to be kind of data centric because I would write code, I would write object oriented code in C++ or VB or later on, C-sharp, Java, whatever. But in the end I'll use usually talking to a relational database. So I was working with data, so that was basically the start of it.

David: [00:12:37] And then later on, as I morphed my career from working in technology to being a systems integration consultant, we started working on projects that were particularly data centric. So, for example, as a consultant, when I was a consultant, I worked on a Fortune 500 company where we built a customer data master because what we had was the client was a financial company and they had the same representation of Dave Langer in different databases all over the place. But in some he would be Dave Langer, some here would be David L Langer, sometimes you'd be David Langer, but they were all the same person. And if you didn't have some sort of unique identifier, like a Social Security number, to link those two together, what are you going to do? Because you you want all of those representations of Dave Langer to be married up. So that was really my first introduction into the power of data and how data is central to corporations, companies, organizations of all kinds. And then from there I moved into traditional buy and data warehousing and then from there onto advanced analytics.

Harpreet: [00:13:40] So could you have imagined that it would have been what it is it being like? You know, data science, data analytics, the community, what it's become. Do you think you would have imagined that it has gotten to the point where it has now back then?

David: [00:13:57] No, no, absolutely not now. And [00:14:00] you're going to relate to this, I imagine so. I took a course in econometrics in my undergrad and I just suffered through it. Thank God for the curve. That's the only reason why I ended up with with an A, not because I was awesome at econometrics. I was just better than the other folks in the class. That was it. And I scraped through that because I couldn't understand what useful, what useless was going to be for me. And now, oh, jeez, more than 20 years later, I'm like, I really wish I had paid more attention in econometrics, and I wish I had taken more econometrics courses because that stuff is actually really super useful. So no, I had no idea what, what, how important data would become. And more importantly, I didn't realize how much I would be fascinated by topics that I was exposed to when I was younger and I thought were useless, quite frankly. And I was so wrong back then.

Harpreet: [00:14:57] It's interesting, man. Like a third of who was J. James here maybe was talking about this and some other folks as well that the stuff that you were most interested in during those during that age of 13 to 15 tends to be the stuff that you gravitate towards the most later in in life. So does that kind of hold true for you here when you kind of look back at your your path and your career?

David: [00:15:23] Yeah. Yeah. And it's certainly true for music.

Harpreet: [00:15:27] Yeah.

David: [00:15:28] I listen to the same kind of music now that I did when I was 13. Yeah, absolutely. So what was I interested in? When I was 13? I was interested in computers. I was interested in science in general. I watched Cosmos the first time it came out with MySpace, with my parents on PBS. So I kind of was kind of steeped and I was much younger than 13 when Cosmos first came out, but it kind of set the stage right, and I like science fiction and all that kind of stuff. So all of those kinds of things absolutely are just as applicable today. Like, I love [00:16:00] watching The Expanse and I look at my fiancee and I tell her I'm like, You know what? The single most important character in The Expanse is? It's not Holden, it's gravity. Everything about that show is about gravity. So, yeah, it's still the same stuff.

Harpreet: [00:16:16] Thanks, man. Thanks for sharing that, your background and all that stuff. I appreciate that. Loved hearing that, those stories. So talk to us a bit about maybe some some hot takes that you can share with us about where you think the field is headed. Maybe a secret contrarian viewpoint that you hold from the rest of the data community.

David: [00:16:35] Not surprisingly, if you if anyone follows my content and you're familiar with it, I believe, and I might be wrong, but I believe that we're going to see a contraction in what is typically referred to as data science within for profit organizations in particular. And mainly it's just the hype cycle, right, where you use the Gartner hype cycle. I think data science is starting to approach the trough of disillusionment because it's been overhyped and you can see tons of means and things on LinkedIn with like the the Scooby Doo thing where they pull the go sheet off and it's AI and they pull it off and if else statements, that whole kind of idea, that machine learning and AI has been overhyped at this point. I think like everything else, it'll eventually come back per the Gartner hype curve, but I think we're going to be hitting a trough of disillusionment. So what I think is really going to be the future is exactly what you described in the intro. I think what we're going to see is over time that organizations are going to realize that they certainly should have focused teams of data specialists. But just like in the old days when you had a word processing department and you don't anymore because everybody uses word and does their own word, processing a certain level of data analysis is just going to be ubiquitous across the organization. It's just going to be an expected skill set, maybe not to the point of doing actual machine learning, but probably far more sophisticated than people are [00:18:00] accustomed to today. And I think that's really going to be the future. I think that's and that's where the best opportunity is, I think, for the vast majority of organizations to really reap the benefits of their data.

Harpreet: [00:18:10] So I think this might tie in nicely here. While the content creation around Excel, it seems so counter to what you would see if you're falling like hashtag data science or scrolling through LinkedIn and you're following data science that there's not many people talking about Excel. So? So why double down on that?

David: [00:18:31] Right. And if, for example, if you go back and you look at what I was posting on LinkedIn back in 2017, I wasn't talking about Excel much, if at all. Almost no, nothing. So like most like most folks with a technical background and in particular a coding or software engineering background, I have suffered from technical snobbery over the years and especially vis a vis Excel and in particular like my very first exposure to it and I think this will resonate with you was when I worked at an insurance company, as I mentioned earlier, and I went to go help out the actuarial department with some of their data and it needs and they're single. Most important system that they used was this monstrous Excel workbook with VBA code and macros all over the place. And as a software engineer, I was like, What are you doing? Oh my God, this is this is wrong. This is this is wrong in a moral sense. That was the indignity that I felt about it. And of course, I was completely wrong, as I know now, because Excel is an excellent tool for so many things. There's a reason why it is as entrenched now as it was 30 years ago, and maybe in some cases even more so is because not only is Microsoft been putting more features into it, of course, but also just at base, it serves its purpose very well.

David: [00:19:56] Which is given its technical technical constraints, [00:20:00] does it allow you to work with data conduct, analyzes, create data visualizations, build applications to a certain extent? Yeah, it does. So if you believe, as I do, that the future of business is going to be where data analysis goes, a certain level of data analysis skills are going to be as expected as skills, with Microsoft Office then targeting Excel makes total sense because it is the world's most popular analytical tool and programing environment. By the way, by far and away, estimates are 700 million users or more worldwide. I mean, it dominates everything in terms of that type of usage count. So it kind of just makes sense to say, look, if you want to affect change in terms of making more people data literate, more professionals of all kinds, data literate, you've got to start with Excel. You can't start with Python, you can't start with or you can't even start with SQL. As much as I love SQL, you can't start there. You've got to start with Excel.

Harpreet: [00:20:55] 100%, 100% agree with you on that. Like I came up using R in undergrad and grad school, but I wasn't like programing with R, I was just fitting linear models or just doing some basic ANOVA. Just I was just doing statistics which, you know, you just call a couple lines of code and that's it. Read the results of a table done. But it wasn't until I started working as an actuary where, you know, I had started using Excel because like you mentioned, everything that we did was in Excel. And that's the first time I ever felt like I was actually coding, was doing those nested if statements and free lookups and doing all this crazy stuff. And I was hooked, man. I was like, This is this is amazing. I think part of it was that going back to, you know, back when I was 13 to 15, I was really into computers. I was doing like cool stuff out of dos. My mom got me this book, Dos for Dummies. And, you know, it just it took me back to that that phase. And I just became hooked. And I really felt like, oh, shit, I can actually write code. This is pretty cool. And then from there it led to, led to doing like VBA, writing macros and stuff in Excel. And then it [00:22:00] then made my ability to code in R and SAS a lot better, like I was treating it more like programing language. So 100% of it that you get good at Excel, then you realize, oh, actually I can help you program in anything.

David: [00:22:16] Yeah. And one of the things that struck me later on was, for example, the actuarial department at the insurance company I worked at. I mean, it was a $4 billion insurance company. So it was a big company, Fortune 500. They made $100 million plus reserving decisions based on what came out of Excel. Yeah, it's crazy. It's absolutely crazy. Or think of mergers and acquisitions or corporate finance. Like I couldn't even imagine when I worked at Microsoft the scale of dollar decisions that were made based on analyzes and tables in Excel. It's if you think about it, it's like, Wow, Excel kind of. It's kind of the center of the universe in a lot of ways.

Harpreet: [00:22:59] And there's nothing necessarily wrong with that, is there? Like that's just kind of how business is done like that.

David: [00:23:06] It is, it is. It is. And I thought when you look at it from a purely software engineering perspective, inevitably you say, yes, it's wrong. There's no script ability, there's no reproducibility, there's no robust version control. I mean, all of that is absolutely true. I would never argue against that. Well, I would say is yes, that's all true. And effectively it doesn't matter, right? It doesn't matter. As we all know in the technology field, it's not always the best technology that ends up dominating the market. So you end up working with a dominant technology. So Excel is good enough. It's good enough. And with all the features of Microsoft has added into it over the years, not only is it good enough, it can actually do a heck of a lot. So if it fits your technical constraints, size of data, those sorts of things, it's actually quite a very powerful tool and probably shouldn't get as much hate on social media as it does.

Harpreet: [00:24:00] You've [00:24:00] mentioned this in a couple of other interviews I've heard with you that Excel is complete. I'll just real quick about what that means. Like what implications does that have for Excel users? Can they start like doing crazy stuff in Excel Taking Over the world? Yeah.

David: [00:24:16] So Alan Turing was a famous, super smart computer scientist guy, an early pioneer in the field. And what he did was he created a bunch of different hypothetical tests to check things out in terms of computer science. And he also created the Turing test, by the way, which is a benchmark for artificial general intelligence as well, and in particular Turing completeness. What it says is essentially it is a check to see if a computing environment, a computing programing language can approximate a universal computing chain. So it's just a test to say, look, is a programing language Turing complete? If it is, it's essentially a general purpose computing technology and Excel was. Even before they expanded it recently with what's known as the lambda function. So you can now write functions inside of Excel, your own hand coded Excel functions without using VBA. But even before that, that addition to Excel, it was technically Turing, complete with just the combination of all of the kind of code that you type into Excel cells, as well as the fact that you can use worksheets to store large amounts of data and then cross-reference them from different worksheets that actually made Excel. Technically, Turing complete a programing environment, even if people don't think of it that way.

Harpreet: [00:25:30] So somebody wanted to do, let's say like a data or an analytics project and they wanted to do it entirely based in Excel. Like how would they go about doing that?

David: [00:25:41] So it's a great question. First and foremost is what is the business question or questions that you want to answer? That's the first thing you decide that. And then what that tells you is what are the types of analyzes? What are the types of things that you might need to do? And then you can map them to see whether or not Excel is appropriate. [00:26:00] So for example, if you want to do really sophisticated time series forecasting, Excel is in your tool, right? If you want to use LTCM or something like that to do Type three, forecasting Excel is not your tool unless you buy some sort of commercial add in to pop in there because it's just not set up for that. So that will certain scenarios are going to rule, excel out right away. So if you say, okay, cool, I can do what I need to do in Excel, the next thing you've got to check and say, Look, what is the scale of the data that I'm working with? So for example, you can do linear regression in Excel, you can do logistic regression in Excel, you can do market basket analysis, you can use clustering, you can do you can build movie recommenders about that.

David: [00:26:40] I would advise doing that, but you can even though I have a tutorial. So but all those things are possible in Excel. The only, the only fundamental problem is, is that Excel will break down at a certain level of scale, either in terms of just the number of rows that you're talking about in your dataset or more importantly, the number of columns that you have. So as long as you got like, let's say 50 or less columns and maybe you're dealing with a couple of hundred thousand rows, Excel can actually do quite a bit. You could actually do a linear regression analysis using the solver or logistic regression analysis during the solver. Or you can do market basket analysis. It's just going to be more time consuming than if you use something like R Python, but you can certainly do it. So business question Do you have the right technique to answer the business question available to you in Excel? Does your scale of data match Excel's capabilities? If all three of those things are a yes, then you're good to go.

Harpreet: [00:27:33] It's like a something equivalent to a GitHub or Tableau Public where people can showcase projects and excel like let's say somebody wanted to put it on the resume, that I've got Excel skills, like how would they how would they communicate that?

David: [00:27:48] So there is so as far as I know, there is no equivalent of Office 365, which is like the server web version of Office that Microsoft sells to large companies [00:28:00] predominantly. There's no public version of that that I'm aware of where you can just like throw your Excel workbook up there and it renders inside of your browser, but you could certainly use GitHub. Absolutely. So GitHub is a actually, I'm a big proponent of the combination of GitHub and YouTube. So you can use GitHub to actually store your actual Excel workbooks and share them publicly. You can put in nice multimedia documents to talk about everything that you did and what's going on in the worksheet. And then of course, ideally you combine that with a YouTube channel where you also create a video that describes what you did, and then you can embed the YouTube video in your GitHub. And that right there is more than sufficient, I would argue, for most people to share that level of work. And by the way, if you're in any sort of business domain where Excel is like the de facto standard like finance, let's say, you're probably going to be able to differentiate your differentiate yourself because you're not going to see a lot of other people doing something like that. But if they, like, communicate with the hiring manager, hey, check out my GitHub, check on my Excel project and my GitHub over here. It's going to be you're going to stand out, I would bet.

Harpreet: [00:29:05] Absolutely, man. Actually, I love that. I love that bit of advice here. That's a great way to go about showcasing those Excel skills and just showcasing the your ability to communicate, think through a problem and come to a solution when it comes. Like interviews, man. Like, I've never sat in an interview where I had to get grilled on Excel Skills or let's say there's some folks listening and they're going into a role that is heavily based in Excel. What should they know? What are like the must, the must know things in Excel?

David: [00:29:36] Yeah, and to be honest with you, I am not necessarily qualified to answer that question because like you, I have never been in an interview where they ask me any questions about Excel. But what I would suggest doing is, you know, there's lots of folks, for example, on LinkedIn that are Excel careers. So you can certainly start, I would say, go find those people out on. Linkedin. Start following their content, maybe connect with them, and [00:30:00] then just ask them what are the kinds of things that you would need to be done, need to be done. You can also search YouTube. By the way, there are plenty of videos on YouTube because I've seen them recommended to me by the YouTube algorithm for things like Excel interview question tips. So you can literally just go to YouTube and do that. And I would also imagine that some of that might be domain specific, right? Where if you're in finance, there probably be some differences between Excel, HR or marketing or something like that.

Harpreet: [00:30:27] So I've heard you mention this before, that Excel really helps make the transition into SQL much easier. So what is it about being proficient in Excel or just knowing Excel that makes that transition so much easier?

David: [00:30:43] Yeah, so it based it simply is. Excel is all about working with tables, right? What do you do? You create tables, you store data and tables. You have to worry about the data formats and tables. You need to filter tables you need to pivot, tables, you need to use VLOOKUP or x lookup or index match to join tables of data. That's basically what most people do when they're using SQL from a data analysis perspective. They're not they're not worried about writing super optimized, stored procedures that are using iterative logic like cursors or something like that, which, of course, is, is a bad idea. Cursors are bad, but I'm just using that as an example. You typically join tables of data. You basically do the same sorts of things when you're using SQL for data analysis as you do an Excel. So conceptually you're already halfway there or maybe even farther. So with that base knowledge, you can springboard into the concepts in Excel, right? Because concepts of Excel are all about, Hey, I've got my left table, I got my right table, I got my join conditions, which is just like what you do on a VLOOKUP, by the way. You just have to learn the syntax, that's all. And of course, pivoting is something that you do extensively in Excel, which is also what you do in SQL Group BI is [00:32:00] your friends in SQL when you're using SQL for data analysis, so there's a lot of overlap there.

Harpreet: [00:32:05] Yes, great point. There's just a lot of conceptual overlap. It's just the way we name the thing or what we call it is a little bit different. But conceptually we're doing the same manipulation, the same operation cursor is bad in SQL.

David: [00:32:18] Dba is hate them because they typically hog resources on the server. Right. Because typically most most relational database engines are typically a big single server usually and just a limited amount of RAM. There's a little amount of CPU, so you're essentially slicing up the server. They had all these requests and the cursor will often take up disproportionate amounts of server resources. So that's why DHBs are not fond of them.

Harpreet: [00:32:43] No. Now you know my friends, so you mentioned a little bit earlier talking about data literacy. What is it about the data literacy that that is so important for professionals to have, I guess. Why is it important that people become data literate professionals.

David: [00:33:00] If for no other reason? Because my experiences have taught me. My belief is that, as I said earlier, the future of business is going to be data driven and just as just as at one point in time professionals that entered the marketplace, that had Microsoft Office skills, had an edge on professionals who did not. That's probably hard for a lot of people to believe, but I'm old enough to remember that it'll be the same. I believe the data analysis skills and like I said, it's not going to be that you need to know stochastic gradient descent and all that kind of stuff. That's not what I'm talking about. Right. But do you have a base understanding of data analysis? Can you do can you do diagnostic analytics? Could you even just use Excel and use the data visualizations available in Excel to explore an interesting business question in the data and arrive at some sort of conclusion, some sort of explanation based on the data. I think that level of expertize is going to become quite ubiquitous, quite standardized over time for basically [00:34:00] all roles and positions and organizations.

Harpreet: [00:34:02] So not necessarily data literacy equaling proficiency with tools to manipulate data, but data literacy as just a way to think about, to think through data to to look at, for example, a spreadsheet and kind of understand what you would need to do to that entity, that thing, in order to answer some questions.

David: [00:34:25] Exactly. So let's take a very specific example. So in the framework of descriptive analytics, diagnostic analytics, predictive analytics, prescriptive analytics, true data literacy starts at diagnostic analytics. The minimum bar for that is what we in the what US data nerds call exploratory data analysis typically, which is also something that you commonly do when you're trying to build a predictive model, like a machine learning model, for example, as well. That level of thought processes. That level of skill, familiarity with things like histograms and box plots and multi dimensional bar charts and things like that, and how to use them scatter plots, that sort of thing allows you to actually perform diagnostics analytics. That skill set is independent of the tooling. You can do it in Excel. You can certainly do it in power. Bi Tableau, Python, ar sas, you name it. Sql To a certain extent you can do all of that sort of thing that in those different tools. So it's actually the knowledge, it's the thought process, it's the understanding of the various graphs and charts and statistics and those sorts of things. That's actually important, not the tools.

Harpreet: [00:35:36] Because those principles, they are essentially timeless principles. The tools may change with the passage of time, but the way you think about it, that those principles probably will not ever change.

David: [00:35:49] I've got a book from the Explorer Exploratory Data Analysis. That's what it's called. And that was back. Yeah, and that's back in the day when computers weren't very powerful and they didn't have graphical displays. [00:36:00] Right. So, so a lot of this was done by hand and those principles are equally applicable today. It's just that you can do them a lot faster because you've got more powerful computers and tooling.

Harpreet: [00:36:10] It was up to you when when would the right time be to start educating folks about data literacy?

David: [00:36:18] Oh, that's a good question. In the United States, I'm not quite I'm assuming that maybe Canada is similar in terms of the educational structure of public schools at high school level, I would assume. Yeah. Generally speaking, if you're if you're going to if you're going to be able to teach a high school student calculus or physics, you should be able to teach them a certain level of data literacy. I don't see why that would be not possible.

Harpreet: [00:36:39] Absolutely good that. Bismuth It's clear that you're on a mission and you want to help develop analytic professionals. So where do this drive and desire to help? Where did that come from? How did you start?

David: [00:36:55] It came from Payne HARPREET, Kapor Payne. So I'm a hands on analytics professional. I do consulting work these days. I do hands on consulting work. Even when I had leadership roles at places like Microsoft and Schedule the City, I still did hands on work and it was all born of this pain of being an analytics professional, taking in data requests, exploring data, myself, coming up with useful insights, and then trying to disseminate that to the organization, either at the executive level or even below that to to drive change, to drive action. And it was very, very difficult because at a certain point I was talking past people, right? So certainly techniques like data storytelling can help with that. But one of the first things you have to assess when you do a data storytelling exercise is what's the level of data literacy of the audience? Because that matters. Even with data storytelling, it's not a panacea. So I was like, okay, I could either the old adage or I could either be part of the problem or be part of the solution. So I chose to be part of the solution. [00:38:00] We'll see if I'm successful, but that's really it was about the pain that I felt as a hands on analytics professional. And based on the responses I've got over the years from comments on LinkedIn, other people in the analytics space have felt it as well, right? Raising the level of data literacy across the board and organization makes the job of the data professional, the dedicated data professional easier, and it benefits everybody.

Harpreet: [00:38:23] So how did you get started, though? Like you've been even at like looking at these numbers that are so impressive on your YouTube channel and your LinkedIn followership? What was that? That I guess the initial content that the initial moment like Amazon start posting on LinkedIn would start to put my ideas out there. How did that start?

David: [00:38:44] Yeah. So let's start with the YouTube channel that was on a lark, essentially. So this is way back in 2014. I was working with a guy and we were talking about data science because that was the hotness back then. Well, it still is, but it was really hot in 2014 and he said, Hey, you know what? I bet you there's a lot of people that could benefit from what you have learned the hard way through Kaggle competitions and just investing in yourself, what you learn during your master's degree, blah, blah, blah, blah, blah. And he's like, Why don't you start a YouTube channel and start throwing some videos up there? And I said, All right, why not? And I threw a video up there in 2014. And it's I don't know, it's got 1.3 1.4 million views so far. That's great. But but that video, there was no master plan. I just some guy said I should do it and I'm like, why not? So I just I had a MacBook Pro at that point in time and I just turned screen flow on and just started recording myself talking.

Harpreet: [00:39:43] That's awesome.

David: [00:39:44] So, so for over time, I recorded some more videos in that in that tutorial series, and then I stopped for a long time. Getting into LinkedIn was actually more deliberate, and that was that really started in about 2017 because I read a book called [00:40:00] Micro Domination, which was all about influencers, quite frankly, influencers. I know you and I both don't like that term, but I'm going to use it because that's what they talk about in the book. So Gary Vee right. Gary Vaynerchuk was one of the one of the people they talk about in this book and they were talking about how you can use the power of social media to build a brand, build a business. And at the very least, even if you never plan on going into business for yourself, build a brand that helps you find a job, right? Because like if you're I don't know if you're Ben Taylor, you probably don't have a problem finding a job, right? You, Ben Taylor, even if you don't ever plan on going into business for yourself. So I got fascinated with that. At the time, I was working for a startup called Data Science Dojo, where I was a full time instructor, and I also saw that as being synergistic with our startup because we went around North America, Canada and the United States predominantly, and we taught week long intensive bootcamps on data science. So it was just part of an overall marketing strategy. And I also thought, hey, be good for myself as well. And that's when I started posting quite regularly on LinkedIn was back in 2017, and that was always the plan was to build my personal brand. I didn't know if I was ever going to start a business myself or not, but I figured there's no downside to having a personal brand. So why not?

Harpreet: [00:41:18] This coming on the heels of of a post that you and I are both tagged in earlier today, that was somebody sharing somebody else's sentiment on Reddit about what's the deal with these LinkedIn influencers. And I mean, I was kind of a bit triggered by that post because like, I don't know why people are out here complaining, but I'm just I'm wondering what your thoughts on that are like what responsibility do we I guess as influencers, if you can call us that, what responsibility do we have to the people that are following us?

David: [00:41:49] Yeah, it's it's a great question. And if folks want to check either your profile by profile LinkedIn, they can see our activity and you can see my post on it. That book that I mentioned that I read back in 2017 [00:42:00] really framed up my philosophy on social media, which was, yeah, it's about personal branding because it's a lot of work. Like if I, if I have a post on LinkedIn that hits and I get lots of comments, I try really, really hard to respond to all, if not almost all of the comments in a thoughtful way. It's a nontrivial amount of work. So yeah, it's not purely altruistic and I don't think that's a problem. What the book said was, Yes, you can do that and people will be okay with it if you provide value. That's the kicker. You got to provide value. And of course, I have to let the community decide whether or not I provide value. But that is my aim. That is my goal. That's why I don't do, for example, on LinkedIn, that's why I don't do a lot of polls. That's why I don't do memes. It's okay if people want to do memes, it's totally fine. That's why I don't do them myself. Well, that and the fact that I just don't have very good graphic design skills, that probably is part of it too. But I'm not Danny Marr. I don't got his skills right. But but for me personally, I'm trying to add value and sometimes that's very explicit, like maybe a post where I put up the PDF with slides from one of my online courses teaching you about how to calculate impurity, for example, how a decision tree works. Or maybe it's about relating a personal story from my career around that, the pitfalls of self service by but always my focus is trying to provide content that will be helpful to the community in some way. And I think that when, when done correctly, folks don't begrudge you building your personal brand and they might not even begrudge you trying to sell something if you're going into business for yourself, as I have.

Harpreet: [00:43:41] I like that. Absolutely like that. That point about personal branding, right. That's putting yourself in a position where luck can. Mind you, you develop a unique brand unique reputation, a unique mindset, a unique perspective, voice, whatever, and you share it, you communicate it. And then all of a sudden, when people [00:44:00] need something that is adjacent to what it is that you do, they're going to come looking for you. Like if people need to get educated on Excel, if people have like, oh my God, I got like a group of new grads coming in. I need to teach them all how to use Excel. Who do I go to? Well, they've lingered on top of mind, like developed a brand perspective and reputation around that. And that lets you take advantage of opportunities that other people might characterize as lucky. But you know damn well it wasn't luck because he put in a ton of hard work to get there. You put in a ton of effort to get there. And I think that's the biggest thing is you can't just always rely on some corporation or organization to just pay you and give you money, right? Like your job is never secure anywhere. So if you're not doing something on the side to develop a brand or reputation for yourself so that you can capitalize on, on, on things that happen when bad things when they happen, then you're not you're not setting yourself up. Right? You know, you want to be I guess I've been I've been reading Antifragile by Nassim Taleb a lot over the last couple of weeks, and he talks about just being able to have options and optionality and being able to to benefit from randomness. And so the more things you do, such as building your brand, the more options you have available to the more you can capitalize on bad things when they happen. Like, I know I can lose my job tomorrow, I know it's possible, but at least I know how to go about setting myself up without having to have somebody else give me a paycheck. Right. Those type of skills, I think, are extremely valuable. Extremely necessary.

David: [00:45:40] Yeah, absolutely. I would want said agree. So for example, I've worked for four startups in my career and I've been laid off from three of them. Only one of them did I not get laid off from. Two of them were in the late nineties and they just basically ran out of capital and they couldn't secure more capital. That happens and that kind of experience teaches you [00:46:00] that you really are accountable for yourself. And so having a personal brand, there's really no downside to it. Like I said, you don't have to go be an influencer and try and sell an online course or coaching or whatever. You don't have to do that even if you're just an employee having. I shouldn't say just simply that sounds terrible. My apologies. If your goal is to be an employee, which is totally awesome, then it also benefits you. Because if nothing else, when you engage with a hiring manager, they're like, Show me. Tell me about yourself. Be like, Hey, check out all my stuff on GitHub or YouTube or social media or all three. There's no downside to it. And it provides you, like you said, options.

Harpreet: [00:46:38] Yeah. Yeah. I think some people think that just because, you know, we're quote unquote influencers, that we can't we can't walk the walk or whatever. Right. Like like. Have you met? I guess I still still write code every single fucking day. Like, I'm still out here training models daily. So there's I don't know, I guess I was treated by by that is like. Right. Just because I may be a quote unquote influencer, just because I may be posting shit doesn't mean that I can't actually do the damn work when time comes. Right.

David: [00:47:05] And and I think that's I think that's a very, very important point, which is, as I often put in my posts on LinkedIn, for example, I let my content do the talking right, like here's, here's some slides out of one of my online courses on the decision tree algorithm. There you go or go check out my YouTube channel, check out my tutorials, and yeah, I've got skills. Am I a world class, python based machine learning engineer? Absolutely not. But I don't claim to be. I don't even talk about that sort of thing. What I do talk about in my posts I do have experience with and I try to prove that through my content. I think I think that is a big differentiator from maybe influencers on social media that are being influencers for influencers sake. And I think there is I think that Post did have a certain amount of validity insofar as there are certain folks, of course, on the platform on LinkedIn, for example, [00:48:00] that are in that camp legitimately not you, not myself, but there are folks that are like that. But painting all influencers with that broad brush, I think was not ideal. It was not good, it was bad, it was very bad.

Harpreet: [00:48:13] Speaking about content creation, I read a post from years. I think it was just maybe a week or so ago or just a few days ago talking about you. You were going through content creation, burnout. Talk to us a bit about that.

David: [00:48:24] So for a while now, I have been pretty serious about building out my YouTube channel with a focus on Excel, teaching various ways to do data analyzes and supporting skills with Excel. So like, I have a tutorial series that teaches SQL to Excel users how to use Power Query in particular for transforming and massaging data for Advanced Analytics. I got tutorials on Market Basket Analysis, canines, cluster analysis. Things like that. And creating those videos is a lot of work. It's it's a lot of work. So for me, 15 minute video, 20 minute video, it takes five or 6 hours of effort to create that. Just that one video between building all the materials, recording, editing, creating thumbnails and Canva, all the things. It takes a lot of time and effort, and LinkedIn's pretty good about this. But other social media platforms like Twitter and Facebook and Reddit and YouTube, there are a lot of trolls, there are a lot of haters out there. And just recently I was just getting a lot of gist, vitriol regarding not the technical content, necessarily, not the quality of the technical content, but attacks on me personally, the fact that I wave my hands too much or the fact that I use certain types of words or whatever it might be, they weren't actually focusing on the actual goal.

David: [00:49:44] They were just focusing on me and personally attacking me. And I just got I just got burned out. I know I shouldn't. It's something I'm working on. This idea that you should just ignore the haters because haters are going to hate. But it takes a toll because when you create stuff like that, you, you have a genuine interest in helping [00:50:00] the community. Because make no mistake, yes, I do have ad revenue that comes in, my channel is monetized, but it it's not a lot. It doesn't justify the amount of effort that it takes to build the video in the first place, not even close. So part of that is putting a lot of energy out there to try and help people. And when you get that hate coming in, it can be quite demoralizing. And I will admit that I'm not nearly as strong enough. I'm not nearly I haven't developed the mental fortitude yet that I need. So I'm taking a break. As a consequence.

Harpreet: [00:50:30] It's hard making content and it is extremely difficult. It's not just like it might be easy for someone to watch a ten minute video because it's only 10 minutes of your time to watch a video. But the effort that goes behind the scenes, like, like you mentioned, like scripting the video, making sure that when you're executing it, everything goes as expected and then the editing and then all that other stuff that goes along with it, it's, it's not easy. Like it easily could be at least like what would you say? Like 5 hours of effort for every one minute of video put out there? I feel like that's kind.

David: [00:51:05] Of my YouTube videos aren't that good of quality. Yeah. So I would I would say that creating, creating a, an online course that you sell that that is definitely a more likely ratio. Right. Based on my experience. So like the online courses I've created, it is a god awful ratio like that. Youtube, not so much. Probably more like like I said, for a 15 minute video might be 4 to 5 hours. But for online courses, it's even more of a multiple for one minute of high quality online course video. It takes could take an hour or two. So, yes, there is a lot of effort that goes into this. And a lot of folks, you know, until you've done it, you don't understand, which is why I am 100% convinced that I have never I have never received any sort of hateful or excessively negative comment in social media from anybody who has a substantial [00:52:00] content creation platform of their own, whether that's LinkedIn, YouTube, whatever it might be, because they know, they know they know how hard it is.

Harpreet: [00:52:07] So I guess when you how are you going to be able to tell that you're kind of ready to go back on on onto the horse? I guess you've noticed some of the early warning signs that burnout was settling in. Were there early warning signs or was it just kind of like, oh, shit, I'm just I just need a break? How did you kind of figure out you're burning out?

David: [00:52:25] So it was pretty easy. I was I was recording a video one Sunday night that I was going to release the following Monday morning, and I was in OBS doing my recording thing as normal. And I didn't realize that my Yeti wasn't actually turned on in OBS. So I did a bunch of takes. A bunch of takes. So maybe 45 minutes worth of takes with no sound that. And I won't lie. I let off a bunch of expletives and I just shut us down. I said, I'm done. I'm not doing videos anymore. For a while that was it. Just because I had reached that point where I had had I had no patience for the foibles of the process itself. I wasn't enjoying the process anymore. It was it was work. It was unpleasant work. I was like, forget it. It's not worth it at this point. So that's that's how I knew it was it was pretty easy.

Harpreet: [00:53:22] Like when you start stop enjoying the process. That's a good, good way to tell, I guess enjoying the process. I guess when it comes to career growth and career development, what would you say is the biggest lesson you've learned the hard way that you want to make sure no one else makes?

David: [00:53:39] Don't do it for the money. Don't do it. Don't do it for the money. If you're data science is a great example, but it could be software engineering, too, by the way, or any any discipline, being a lawyer, whatever it might be, don't do it for the money, because just take data scientists. The example, data science isn't necessarily the most highest paying technical career that you can take. You can you can actually [00:54:00] make quite a bit more money. But just being a software, just damn not that word again.

Harpreet: [00:54:04] Yeah.

David: [00:54:04] You can make more money being a software engineer. That's what I'm going to say. So do it for the love. Right? Data science is not that. It's all it's cracked up to be, right? There's the interview process is hard. There's a lot of things you need to know. You oftentimes get ignored. You don't make the change that you want. So if you just do data, if you get into data science simply because of the money, you're not necessarily going to have the wherewithal to withstand the process that comes along with it. So just don't do things for the money. Find something that you love to do that also matches your career goals simultaneously. And certainly a level of income is a legitimate career goal. I'd be the first one to tell you that. That's absolutely true. That's totally legit, but it can't be the only thing. It can't be necessarily the most important thing. Otherwise you will find oftentimes that you might be successful ostensibly. But inside, yeah, you're you're hating life.

Harpreet: [00:54:55] So let's talk about, I guess that a little bit of around artistry and creativity in our profession and in our field you feel like is if we're not doing it for the money, should we do it for that? Should we do it for kind of like the artistry and the creativity? What's your what's your take on that?

David: [00:55:11] Absolutely. And I have always thought that technical roles have a certain level of creativity. If you've ever met a dyed in the wool software engineer, they consider themselves artists. Obviously, they may not paint, they may not sculpt, but they consider themselves artists, that's for sure. Code is their medium. And I used to think of that same way to back when I was a software architect. If I created a nice, layered, object oriented design with all of the patterns that you would like to see, the separations of concerns and all that kind of stuff, I thought that is a very creative process myself, even though it was technical data science is the same way, data analysis is the same way. There's a certain craft of a certain artistry of sitting there and thinking, coming up with new hypotheses, coming up with new ways [00:56:00] of thinking about the business, new things to explore in the data to see if there are there are profound insights. There are opportunities for economic gain for the organization. That is a creative process. As I mentioned earlier, the rise of data storytelling is also a manifestation of a very creative process.

David: [00:56:18] You conduct an analysis and maybe you want to characterize that as uncreative. Okay, great. But it's going to fall on deaf ears unless you put some artistry and some creativity into the narrative to actually drive change. So yes, absolutely. You have to love data. A while back, I put up a post that was pretty popular where I said, If you're going to be a data scientist, you need to be like Will Hunting from the movie Good Will Hunting. He worked as a janitor at the beginning of the movie at MIT, not because he loved being a janitor, but because he wanted to be close to what he loved, which was mathematics and all of that sort of thing. Same thing with data science. Go into data science if you love data, if you love working with data, analyzing data, building predictive models, driving change in organization, using data to provide real value to the business. That's why you should be in data science. And if you happen to make good money.

Harpreet: [00:57:11] Love that, love that perspective 100% on board with that I was I interviewed released just recently with Andy Hunt, who wrote Pragmatic Programmer. And he was talking about how a great book, great book, excellent, excellent book. And he's talking about the same thing you were saying about how you are treated like an artisan. You know, it's a craft. Very well put, Dave. Dave, let's go ahead and wrap up a little bit here. And I mean a little bit because it got one formal question and then a quick random round of question is it is 100 years in the future, what do you want to be remembered for?

David: [00:57:43] Bringing in data literacy to the masses, making skills and data analysis as common as skills with Microsoft Word and Office?

Harpreet: [00:57:51] I think you are definitely well. On your way there with all the content you're pushing out with the great hosts and stuff that you're doing on on LinkedIn and [00:58:00] just so much, so much value, you're giving back a lot to community. So thank you on behalf of professionals everywhere. If I'm able to speak for them. Thanks. Go ahead and jump right into the random round. What are you currently reading?

David: [00:58:12] I am reading the Wall Street Journal Guide to Information Graphics because as I mentioned earlier, my graphic design skills aren't the greatest. So time to kind of get better at that sort of thing.

Harpreet: [00:58:23] I think I'd check that one out. Wall Street Journal Guide to Information Graphics.

David: [00:58:26] I believe that's the title, but if you search for it on Amazon, you'll find it, I'm sure.

Harpreet: [00:58:30] What song do you currently have on repeat?

David: [00:58:34] Believe it or not, All Out Life by Slipknot.

Harpreet: [00:58:37] Nice. I would not have put you as a Slipknot guy.

David: [00:58:40] Oh, look at the hair, man. I'm living the night I live in the 1980s hair band dream finally.

Harpreet: [00:58:45] Yeah I even Slipknot's crazy man Slipknot say that's the stuff I listen to and I was like in later high school years that stuff was I listen to really, really aggressive music back back in those days and that yeah.

David: [00:58:57] So I grew up on the big four Anthrax, Metallica, Megadeth and Slayer.

Harpreet: [00:59:01] So nice. So we're going to open up a random question generator here. All right, cool. First question out of here, is mountains or ocean?

David: [00:59:11] Mountains?

Harpreet: [00:59:11] What is your favorite candy?

David: [00:59:13] Kit Kat? American Kit Kat.

Harpreet: [00:59:16] Is different, man. I'm not. It's different. It's different in Canada. The Canada Canadian Kit Kat is not the same as American Kit Kat.

David: [00:59:23] It's good, don't get me wrong. But I prefer the American. That's American.

Harpreet: [00:59:26] Kit Kat. Yeah, 100%, man. I don't even eat Kit Kat in Canada. It takes a weird who's your favorite teacher and why.

David: [00:59:33] My favorite teacher? Oh, it has to be Mr. Fresh. He was my AP computer science teacher in high school.

Harpreet: [00:59:40] You lost all of your possessions, but won. What would you want it to be.

David: [00:59:46] If I lost all of my possessions? But why would you want it to be probably my laptop?

Harpreet: [00:59:51] That's definitely a come on. Yeah, yeah, yeah.

David: [00:59:55] I mean, I do have Dropbox, so technically all the stuff is cloud, but yeah.

Harpreet: [00:59:59] Yeah. [01:00:00] I mean, yeah. Now I feel you, Dave. How can people connect with you and where can they find you online?

David: [01:00:06] So the easiest way to connect with me is on LinkedIn. I am a very, very frequent contributor on LinkedIn. You can also find my content on YouTube. Just search Dave Langer and I am the first result on the YouTube search engine as well.

Harpreet: [01:00:18] And if you're interested in hearing more from Dave, you could tune in to like the first 25 or 26 of the happy hour sessions. Dave was a regular attendee there, so a lot of great advice and a lot of great conversations happening there. So that's another place you can check Dave out. Dave, thank you so much for taking time out of your schedule to be on the show today. Man, I really appreciate having you here.

David: [01:00:37] My pleasure. Thanks for having me on the show.

Harpreet: [01:00:39] My friends, remember, you've got one life on this planet. Why not try to do some big cheers, everyone. Okay.