SDS 369: Real Data Analytics for Economics, HR, and COVID-19

Podcast Guest: John Johnson

May 27, 2020

In this great chat we discussed taking a bet on yourself, business culture, data science vs data analytics, how to select a consultant as a business owner, and the importance of data in analytics.

About John Johnson
Dr. John Johnson is a PhD economist who specializes in applying economic
analysis to large and complex data sets. He is the founder and CEO of Edgeworth Economics,
a world-class economic consulting firm, and Edgeworth Analytics, a company
focused on using data analysis in human resources and other business services. In his work with Fortune 100 companies, he has analyzed large sample sets and internal survey data in high-stakes, “bet-the-company” litigation. As an expert on interpreting data, John has taught at Georgetown University and
has been published in numerous books, journals, and magazines.
Overview
John Johnson has lived and worked in Washington D.C. for almost 20 years and is one of the few people who live and work in the city but don’t touch the political world or government jobs. He came to D.C. for the economist job market after having a run of academia. John studied economics at MIT and worked and learned at companies before deciding to set out on his own with his own company. Edgeworth was the result of John’s stress for core values for his business in both how they treat their customers and their employees. The core value of the firm is the growth of employees and assets to the economy followed by pushing for the best quality of work. Commitment to the culture is extremely important for even the best economist in the world. 
Edgeworth Economics works in litigation where there is a large dataset. For example, they had a case regarding the prices of candy bars nationwide. On the analytics side they work in HR and general business analytics to assist companies with structuring data problems and working through it. They’ve also worked in cases involving grocery products—data on every midwestern grocery product sold or shipped to determine the possibility of economic collusion between competitors. Through a lot of their work, it is about what the question is they’re trying to answer, what is the data available, and how can you rigorously test it?
John defines data science as the host of technology, software, and data capabilities while analytics is focused on what question you need to answer in order to function as a business. It’s not about selling someone on a tool, it’s about data and what that data ultimately tells you. What the company sells is a process. John also practices gaining domain knowledge to continue to function as a professional economist and make himself smart about data problems. For example, he toured chocolate factories to understand the big scope of the data while working on his case around chocolate bar sales. 
John discussed the response to COVID-19. The lack of federal response and the focus on state-level programs creates anecdotal stories in the media about the consequences of the economy “reopening”. The simplistic narratives, as John says, drives economists nuts. What Edgeworth is doing now are natural experiments, where they collect data during the reopening and track the variations between states and between counties. This is a great example of “lip service” analytics vs. truly thinking through the issues and factoring in variables and existing data. 
Beyond this, John’s company also works in analytics in HR for businesses. They enter into a business with an underlying labor problem, such as retention issues. They then bring in framing questions to build the reality of the business to start to see patterns that happen, through data analysis and the study of a business and its culture. He does note that data and technology cannot replace soft skills and the human aspect of companies, particularly HR. What will happen is a move towards more effective HR systems and more intelligent businesses. 
In this episode you will learn: 
  • Living and working in Washington D.C. [4:11]
  • John’s initial jobs before Edgeworth [8:41]
  • Edgeworth core values [12:01]
  • Edgeworth Economics and Edgeworth Analytics case studies [16:57]
  • Data analytics vs. data science [29:50]
  • Parachuting into industries [36:06]
  • Real analytics vs. “lip service” [42:11]
  • HR business analytics [51:13]
  • How much, as a business owner, should you rely on a consultant? [56:26]
  • John’s advice to worried business owners [59:24] 
Items mentioned in this podcast:
Follow John:
Episode Transcript 

Podcast Transcript

Kirill Eremenko: This is episode number 369 with co-founder and CEO of Edgeworth Analytics and Edgeworth Economics, John Johnson. 

Kirill Eremenko: Welcome to the SuperDataScience Podcast. My name is Kirill Eremenko, Data Science Coach and Lifestyle Entrepreneur. And each week we bring you inspiring people and ideas to help you build your successful career in data science. Thanks for being here today and now let’s make the complex simple. 
Kirill Eremenko: Welcome back to the SuperDataScience podcast everybody, super pumped to have you back here on the show, just got off the phone with John Johnson just over an hour ago, and we had an amazing chat. I’m still under the great impressions from that conversation. A quick note, this episode is available in a video version and you can find it at www.superdatascience.coms/369. So check it out there if you’d like to see the video version of our chat. 
Kirill Eremenko: What you need to know about John Johnson is that he is the CEO and the co-founder of two companies, Edgeworth Economics which he founded in 2009 over 10 years ago and Edgeworth Analytics, which he founded in 2019. And they use data to do very cool things at both organizations. So at Edgeworth Economics, they use data to help with litigations, help with cases that go to court in front of judges or juries and extract insights, and then present them to the jury. So as you can imagine, it’s a very exciting space of analytics, and you will hear a few examples of cases that John has been involved of and learn how they work. And now onto Edgeworth Analytics, this is a consulting company in the space of data, in the space of analytics, where they focus on helping businesses overcome any data challenges. 
Kirill Eremenko: Again, there’s going to be a lot of valuable insights. Especially you will find this episode useful and valuable if you’re a business owner, a director, an executive, an entrepreneur, or you want to become one of those four categories. Because John is going to guide you and help understand what it means to use analytics to your business advantage. And this episode is actually going to be useful for absolutely anybody in the space of data science and analytics because of all the amazing case studies. 
Kirill Eremenko: And here are some examples of the things that we’re going to be talking about. Taking a bet on yourself, entrepreneurship and John’s background, how he got his PhD at MIT, what he did, when he went to Washington, how he started his businesses, a business culture, case examples, data science versus data analytics, domain knowledge, econometrics, and how it’s the combination of economics and statistics. How to pick a consultant if you’re a business owner, coronavirus, and the state’s reopening and what they’ve been tracking around that, some interesting insights there. 
Kirill Eremenko: Building an analytics function versus hiring external consultants as a business HR analytics, lip service analytics versus data doesn’t lie, and many, many more cool topics. I found myself sitting on the edge of the seat, listening to John. So I highly recommend checking out this episode. So without further ado, let’s get straight into it. And I bring to you co-founder and CEO at Edgeworth Analytics and Edgeworth Economics, John Johnson.
Kirill Eremenko: Welcome to the SuperDataScience Podcast everybody. Super excited to have you on board. And today we’ve got a super special guest. John Johnson calling in from Washington, DC. John, how are you doing today? 
John Johnson: Doing great. So fun to be here and excited about our discussion. 
Kirill Eremenko: Yeah, me too. I’m very excited about what we have to talk about. And yeah, so as we were chatting before the podcast, you live in Washington DC, and you get the opportunity to have a run or before the coronavirus, you could run towards the White House and all the monuments. Tell us a bit about that experience. How is it living in Washington? 
John Johnson: Yeah it’s really interesting to be in Washington. This is a great place to raise a family. I’ve lived here about 20 years and but, my job and my work has nothing to really do with the government at all. So I’m kind of a little bit rare and there is a group of us here that actually just are consultants or do other kinds of work that have nothing to do with the federal or state government. So for me, a lot of my attraction to the government is now I see all the big buildings and I can go every morning and go for a run and it’s a beautiful place to visit. And just I’m always struck by the monuments when I’m able to run and I go up the Lincoln Memorial steps every morning. 
John Johnson: So I definitely miss that. It’s one thing that I miss the most from the coronavirus shutdown is the ability to do that. But this is an interesting city, but there’s a lot more to it than the political world that you see. I kind of liken it to people who live in Los Angeles, there’s Hollywood, but there’s other stuff there too. And in DC, our Hollywood, unfortunately is the government, but there’s other stuff to do here as well. 
Kirill Eremenko: Yeah. Interesting. How long have you been in Washington?
 
John Johnson: I moved here in July of 2001. I’d been a college professor for a few years at the University of Illinois, which is in the Midwest. And I was very quickly realized I was going to be the most popular professor on campus, but I found being an academic a little bit too theoretical, a little too lonely for me. So I decided to try consulting and DC is a great city for economist job. So my wife and I kind of moved this coast and just tried it and figured I’d try to get a job. And if that job didn’t work out, we wouldn’t have to move again and been here ever since. 
Kirill Eremenko: Awesome. What do you mean by most popular professor? 
John Johnson: One of the things I would do with the undergraduates, I had had classes with about 75, 80 students and I would learn all their names. If they didn’t show up for class, I’d call them and find out where they were. And so I was really popular. People would come to my office hours. In fact, one of my closest friends and business partners actually was an undergraduate in my very first class, 20 years ago. 
Kirill Eremenko: Oh, wow. 
John Johnson: I know he and I stayed in touch over the years. And when I started my business, asked him if he would move here to help run our HR function. And so being a professor was a great experience in many respects, but being an academic is a lot about the research you do. And although I like research, I just found it to be a little bit too theoretical. And in my job now I get to work on real world problems and that’s what I really wanted to do with my career. 
Kirill Eremenko: Fantastic. That’s a great analogy. And we talk quite a bit about guests who’ve been in an academia background, the difference, and maybe we’ll touch on that as well between the difference between academic applications and real world. But I want to talk a bit about you got your PhD in economics from MIT. What is it like studying at MIT? 
John Johnson: It’s crazy. Look, I always joked that I felt like I was one of the dumbest people in the economics department, but being the dumbest person in MIT, isn’t terrible. I mean, in my class, there were four John Bates Clark Medal winners who are the best economist under the age of 40. Four of those people were in my graduating class alone. 
Kirill Eremenko: Wow. 
John Johnson: Esther Duflo who just won the Nobel Prize was a classmate of mine. 
Kirill Eremenko: Wow. 
John Johnson: That’s how I know her and it’s like, wow, I know all these really famous academic people. So one of the best parts about it is that I met a lot of really smart people who really improved my skills as an economist. The other hand it was really challenging and everyone who’s there was the smartest person in their class in undergraduate, and then you get there, you’re not the smartest person anymore, you could get a 19 on an exam out of 100 and that was passing. And so when I look back on my time there, I think to myself I’m really glad I worked hard and made it through. It was very difficult at the time and humbling. But it set me up so well for my future career. And I’m just so thankful for the opportunity, both for the people who I met there, but also just being trained really well as an economist. It was really one of the most challenging things I’ve ever done. 
Kirill Eremenko: Yeah. And to your point, that’s an amazing place to be. I would 10 times out of 10, I would choose to be the dumbest person in the room, than the smartest person in the room. If you’re the smartest person in the room, there’s nobody to learn from. There’s nowhere to grow. I would much rather be the dumbest than, okay, I’m the dumbest, but at least I can learn so much right now. 
John Johnson: Oh, I definitely had people to learn from. 
Kirill Eremenko: Yeah. That’s awesome. That’s awesome. So you moved to Washington in 2001 but there’s about eight years until you founded your first company, which we’ll talk about in a bit, what did you do for those eight years? 
John Johnson: Yeah, so I worked for a large economic consulting firm and I had a really successful career there. I learned from a lot of interesting people. There were a number of these different types of firms. This particular firm was a bigger firm. And basically what I found was I did have a lot of success. I was told repeatedly that I was a superstar that I was on the right track, but if I could be a little more patient and just waste my turn, I would in the future be able to do what I wanted to do. And I think if you’re a successful, driven person being told to be patient is probably the worst thing you could ever tell them to do. And so, along the way, I started to realize, not only did I have ideas about what I thought, how a company should run, what I wanted to do, but I was willing to take a bet on myself and say, “Okay, well then if you think you can do this better, then go do it.” 
John Johnson: Now at the time people looked at me like I had six heads, “What are you doing? You think you can go do this yourself. You’re a young punk kid.” Maybe I was, but it turned out to be the best thing I ever chose to do. 
Kirill Eremenko: Yeah, absolutely. Well I remember when I was starting my first business. Oh yeah. When I was leaving my employment the person, the most scared was my mom. She was thinking, “What are you doing? All that security, you’re throwing it away.” 
John Johnson: My business partners, as it turned out the people that came and did this with me both of them within the first year, they both, their wives were pregnant with their first kids. It was crazy and it was during the 2009 recession. So one of the things that has been striking is that as we’ve been going through the coronavirus crisis, which is providing different challenges, I am surprised how much it reminds me of the kinds of things we had to do back in 2009. But I do think the good news is in these types of economic settings, it’s incredible opportunity. It’s a real chance for companies to innovate, to be aggressive, to think about how they can serve in different ways. But look, I never realized I was going to be an entrepreneur. 
John Johnson: Now, people who know me are like, of course you’re going to be an entrepreneur, but I didn’t appreciate that. And I didn’t know how much having a vision and having a way of thinking things should be done. That’s the core of entrepreneurial people I find is that you think you have a better mousetrap. And it really bothers you when things aren’t done the way you want them to be done. 
John Johnson: And so I had not appreciated how much career success and validation comes from, at least for someone with that makeup, having a value system, having a set of behaviors that you subscribe to in your company. And so, yeah, we’re a rigorous company. We do analytics and litigation work, but we also have a core set of values that we follow, which is both about how we do the work, about how we treat our employees, how we manage the company. And it’s interesting because I just had not appreciated how much of my personal validation would come from those things as much as also from doing really high-quality work. 
Kirill Eremenko: Fantastic. Tell us a bit about your values. What are some of the core values? 
John Johnson: Well, I think the biggest, the name of the company, Edgeworth Economics and Edgeworth Analytics, Edgeworth was a 19th century economist who basically discussed the gains from trade. And there’s something called the Edgeworth box, which is where you could find the point where you can trade such that everybody’s made better off and nobody’s made worse off. That’s actually, and so in a introductory economics classes across the country, you’re taught the Edgeworth box and you have these little curbs and you find the optimal point. That’s the corporate philosophy, we were going to create a system where everybody could be made better off. 
John Johnson: Part of what we do at our firm is we train young economists to contribute in different ways to advance their career and that’s at all levels, even our undergraduates who come and are going off to graduate school. We’re trying to do is provide them with opportunities so they can be better economists, proceed whatever their career [inaudible 00:12:55]. And so from that respect that is kind of the core value of the firm. Now tied with that, of course is also highest quality work, being responsive to our clients, being smart about customer service, those kinds of things. It’s about though an experience for our clients and our employees, which really puts them in the best position to succeed and really being a team. 
Kirill Eremenko: Fantastic. I love it. I love when those are very thought through and putting the clients first and taking care of the team, of course, amazing set of values. And I’m sure it’s a culture once you build it. I think you mentioned before the podcast, you’re about 80 people in your company right now. Once you build a culture like that around those values, anybody joining naturally has to either fit in or other people will notice that they just don’t fit in with those values. 
John Johnson: And some of the most painful things you deal with when you have a company, is that when people don’t fit the culture, even if they seem to fit on other dimensions, that’s actually one of the things you have to be committed to the culture first and foremost. Even someone can be the best economist in the world, but if they don’t fit in with the culture, then that’s not good enough. You got to, once you commit to a culture as a firm, that’s your DNA, that’s what makes it distinct. So that’s part of what we look at when we recruit. And that’s part of what we’re looking for in people that we hire and develop and mentor and bring along. For me as a business owner now 10, 11 years in I’m looking at what’s the next generation?
John Johnson: I’ve had a very successful career. I’m not saying my career is over, but for me there’s a big part of my next 10-year plan is really my legacy. Can I leave this firm in the hands of another group of economists that can take the torch and continue? And so when you’re trying to do that, you’re looking for people that are again, entrepreneurial that understand the culture. That are really good at what they do, but also who are willing to take over a vision for how they want to transform the firm. Obviously if you’re passing the legacy, you’ve got a certain core set of values, but they have to be able to shape it, develop it, make it their own. And so that’s what I find is my challenge. That’s my personal career goal for the next 10 years is how do we do that? In a smart way, but those are different challenges. I’m just really proud of this fact that most companies don’t get to the point they can even start to think about that. We’re lucky that we can. 
Kirill Eremenko: Yeah, no, that’s amazing. That’s amazing. 
Kirill Eremenko: Hope you’re enjoying this amazing episode. I’ve got a cool announcement for you and we’ll get straight back to it. Virtual Data Science Conference. Curious? You’ve probably heard of DataScienceGO, the conference that we’ve been running for the past three years in Southern California. Maybe you’ve attended, if so, it was super cool to have you there. But maybe you weren’t able to attend for the reason of being in a completely different country, or the flights were too long, or the timing wasn’t perfect. There could be plenty of reasons why you weren’t able to attend. But now, we’re bringing DataScienceGO to you. 
Kirill Eremenko: This June, we’re hosting DataScienceGO virtually and you can attend and get an amazing experience there. Guess what, the best part is that it’s absolutely free. Just head on over to datasciencego.com and get your tickets today. This will be our very first time running a virtual event. Nevertheless, we’re still going to combine the three key pillars of fun, amazing talks and networking into this event. You’ll hear from speakers like John Krohn, Sam Hinton, Hadelin de Ponteves, Stephen Welch, and many others. 
Kirill Eremenko: Plus, you’ll be able to network with your peers. This event is going to be epic on all fronts and we’d love to see you there. Head on over to datasciencego.com/virtual and get your ticket today. The number of seats is limited. We’d love to have everybody there. For our very first event, we’re limiting the number of seats to make it more manageable. Make sure to get your tickets today, if you want to be part of this. On that note, I look forward to seeing you there. Now, let’s get back to this amazing episode. 
Kirill Eremenko: Tell us a bit about what you do. You’ve painted a picture of the company. It sounds like an amazing place to work and it has a great culture and I’m sure you take great care of the clients, but what do you do? Let’s start with Edgeworth Economics, the company that’s over 10 years old. 
John Johnson: Yeah. So the economics company, basically a lot of our work is in litigation and there we have a number of people that are expert witnesses. And so what we generally do is we are involved in litigation cases where there are large data sets. My example is I had a case many years ago where I had the data on every chocolate candy bar sold in the United States. 
Kirill Eremenko: Wow. 
John Johnson: And I had to look at the data and figure out what happened to pricing over a 10-year period. And how to go towards some chocolate factories, learn about the industry. And so our economic consulting litigation work, we basically parachuted into different industries. We get the data, we learn all everything about how it works. And then we develop our expert opinions in different litigated settings on how either a workforce works in our labor work or how a product market works in our anti-trust law. That’s what the litigation company does. 
John Johnson: Then the analytics company, which is more recent, that takes that same skill set with these highly credentialed PhDs and statisticians and MBAs. And now we’re also help companies with their practical issues like, okay, outside of litigation, how can we help you? And that business there’s a number of different angles. One part of it is HR related where we do a lot of education on HR analytics, in consulting on HR analytics. Then just general business analytics where companies have a data question and you’re like, “I have a data question and I need help. I don’t exactly know what I need. I just know I need some help. Can you get me there? Can you just help walk us through how to structure a data problem, how to think about a data problem?” 
John Johnson: And then the other part our analytic company has been doing, which is interesting is with the coronavirus taking over the world, we’ve actually launched a coronavirus impact study, where we basically are posting almost on a daily basis, different analytics on the coronavirus. And this is more of a public service, but it actually been interesting because there’s a lot of companies that are interested in this, where we have dashboards for different states, there’s the reopening in the United States, showing what the cases are doing on a daily basis, matching the demographics. We have a lot of reports on the restaurant industry, working on a report of the alcohol industry and how drinking has skyrocketed during the coronavirus, but a lot of different types of things like that. So we have a pretty broad-based business, but the common theme is rigorous data analysis explained simply, and that transcends everything we do. 
Kirill Eremenko: Okay. Wow. That’s really cool. And it’s a great foundation to start like you did the Edgeworth Economics where you have to be rigorous. You’re going to court, you’re presenting to, I guess juries. You’re acting as a witness. You can’t go wrong there, right? Can you maybe give us an example of like the, if you’re able to share from the 10 years, one or two of the most memorable cases that you had to deal with? 
John Johnson: Yeah. I mean, I can. Obviously our testimony is usually confidential, but once I’ve testified or decisions have come out, I can talk about that little bit. I’ll give you one really interesting case involved grocery products where I actually testified where I worked on the case for almost eight or nine years of my life. 
Kirill Eremenko: No way. 
John Johnson: Yeah, right. And I had data on every grocery product sold or shipped in the Midwest for a 10, 12-year period. And it basically it was an anti-trust case involvement was alleged allocation of markets where one competitor agrees to not compete with another. And that’s again, that’s- 
Kirill Eremenko: What does the anti-trust mean? 
John Johnson: So antitrust is the part of the US law that has to do with competition economics and how firms… And so there’s certain laws in the United States, particularly with called the Sherman Act, which you cannot agree with a competitor on pricing. And you cannot agree with a competitor on how whether you can serve a market. So you can’t say, “Hey, you take these clients, I’ll take those clients and I won’t compete with you and you don’t compete with me.” Okay. Not allowed. That’s illegal, that’s actually a criminal violation, actually a criminal violation in most parts of the world. You can’t discuss with your competitors, you got to be very careful. So a lot of the types of antitrust cases we deal with involve either communication between competitors, alleged communications and what the consequences were. 
John Johnson: So in this particular case, what I was asked to do was a lot of things, but ultimately I had to process these terabytes of data on grocery shipments to see if there was a change in pricing or if the prices were different over time. And what I actually testified in court about was what would the damages have looked like if something had happened? So the jury gets to decide if something happened or not. And I had to calculate, okay, well, assuming it did happen, were there damages? And so in that case, I remember I had to… I’m trying to explain to the jury that what the other expert had done, had not taken account for lots of other factors that would have affected prices. 
John Johnson: And how do you explain that to a jury? That’s a fairly simple concept. I actually talked about having a taco truck at a restaurant and imagine you have a taco truck. And how do you convey that if my chef got sick or suddenly the cost of the meat for my tacos went up? This other expert was not counting any of those things as potentially affecting prices. Do you really have to be creative? I did a good billion runs of data analysis with huge data sets, but then a lot of the art was how do I take all this analysis and show in a very pinpointed way to a jury? Okay. Here’s what happened. So the jury ultimately ruled in our favor and that was kind of cool. We won the case, but also I remember one of the most compelling moments is up on the stand talking and there’s four men of the jury with this little old lady and I’m talking and I’m looking at the jury and she’s taking notes. I can see her smiling at me. And [inaudible 00:23:13] knew you were teaching. And so that’s one of my favorite stories. 
John Johnson: I mean, very few cases actually end up in a jury trial as it turns out. People that go, it’s all about trial. Actually, most of my work is in depositions or lawyers are questioning me and briefing and hearings before judges. But that’s the kind of stuff where I learned a lot about an industry, I processed a ton of data, but the real hard work after processing the data and coming up with my opinions is how did I translate it to a common group? An audience that was not interested in data. I called for jury duty. And that was a lot of the challenges that I faced in that case, but that was really rewarding. 
Kirill Eremenko: Okay. Wow. So the jury in this case ruled that there was collusion. 
John Johnson: That there was not collusion. 
Kirill Eremenko: There was not collusion. Okay. Interesting. Wow. What about any cases where the result was very like, for instance, somebody, I don’t know, what’s the most severe consequence of a case that you were involved in? 
John Johnson: The economic part of a case is only one part. So it’s important when you do this kind of work. I’m usually asked to opine on certain circumscribed issues. I don’t, as an economist, for example, in cases where there might be criminal violations, I’m not usually asked to opine on whether someone did something. That’s not the role for the economist. How am I supposed to know two people agreed to something? So, but the reality is there are gigantic fines. There are gigantic criminal penalties. There are gigantic civil penalties that come with it. And so we could be dealing with cases where the damages could be billions and billions of dollars. It’s called Bet-the-company litigation. And so I don’t want to underplay, it’s very stressful because companies that are hiring you or plaintiffs that are hiring you, there’s a lot at stake in these cases. 
John Johnson: And so what we do at our firm, is I have a reputation for being very scientific, very driven by the data. People who hired me want an objective opinion based on the data, but one that is I am very meticulous about the science and how you apply the science of statistics and economics. And so that’s just what I do. But the stakes are always high and you don’t win every case. Even times, and this is one of the frustrating things is even times where I believe the economics we did was the right economics. That’s not always the definitive ruling in a piece of litigation. Judges can decide for a whole host of reasons or there’s different standards. And so being something that’s good enough to be published may not be the standard by which a court is saying, they’re going to apply to whether or not an economic analysis is good enough for the opinion [inaudible 00:26:00]. 
John Johnson: And that’s just a reality of what you have to accept when you’re an expert that I can do the best economics in the world, but there’s other things that affect it as well. As long as I’m confident that the opinion I’ve offered is consistent with my standards, is responsible, is the truth. That’s what you have to be comfortable with. 
Kirill Eremenko: Got you. And so about this case, I just wanted to understand it fully. So you’re using patterns of movement and sales of grocery products in the Midwest. Your goal, and what you were able to do was to demonstrate whether that there was no collusion between two companies just by, through the sales patterns that you’re seeing overall? 
John Johnson: It was a little more, it was more along the lines of I was able to demonstrate that the prices weren’t changed in a way that was consistent with collusion. So basically we had, there were several facilities that were at issue and basically a lot of it dealt with how far our grocery products shipped and who could you turn to or competition in the face of a price increase? That’s the kind of stuff that we deal with. So yeah, there were maps in that case, there were how far the product shipped, there were, what were the demand factors that changed different types of products? How often were they bundled? How big were the customers? One of the things that comes up quite a bit in our cases is very large customers have buying power. Walmart is always the classic example. Walmart usually gets the best price, no matter what, [crosstalk 00:27:26] a lot of buying power. 
John Johnson: So that’s a big, competitive factor that has to be accounted for when you think about these things. So that was the basic idea is really getting into the data at a granular level to understand where product flowed, who bought the product, what were they looking to for competitive? What were the other competitive conditions? Those kinds of things. 
Kirill Eremenko: Got you. Through many years of working on cases like that, and what’s the number of cases of the top of your head that you’ve done? 
John Johnson: Probably about 100. 
Kirill Eremenko: Through working through 100 cases so rigorously, some of them taking many years, you’ve probably developed an approach to analytics. What is your approach? 
John Johnson: Well, and I think, yeah, I have, I think that it’s what I rely on and this is true in both my litigation work and my analytics work, although they’re slightly different focuses, but in both at the end of the day, it’s about what is the question we’re actually trying to answer? Always coming up with it, very concrete. Here’s the question, what is the data that’s available? And then how can you rigorously test what you’re looking at? And rigorously testing, it can be things like different statistical techniques. One of the issues that comes up quite a bit on the litigation side, and it comes up on the consulting side, but a little less is the importance of averages. And how averaging data can create all sorts of problems if you’re actually trying to understand very nuanced phenomenon. Sometimes averages are perfectly fine for an answer, but oftentimes averages can obscure tons of variation and experiences and economic phenomenon. 
John Johnson: And so what I believe in doing is you actually test, you can test whether an average applies, you can go in and you can do statistical tests. And I wish I could say I’m the one who came up with the statistical tests. I’m not, it’s a whole legacy of statisticians. Although sometimes talk to lawyers, you think I was the pioneer of some of these tests, but the point is whatever the problem is, it’s thinking hard about question, data. What’s an answer that’s actually useful? And then what are the limitations of the answer? And if you follow that process, whether you’re answering a question for your company in a analytic sense or a question in litigation sense, it’s going to lead you to the right answer or more important it’s going to highlight what are the strengths and weaknesses of the analysis? 
John Johnson: I think people get a false sense of security around data that just because they’re using data, it means they’re going to get the right answer or the rigorous answer. Data in and of itself is not enough. It’s well thought out data analysis that can get you to the right answer. But what is the right answer? Oftentimes the right answer is, wow, this is interesting. It’s pointing in this direction, but you know what? I better consider XYZ as well. All right, let’s go back and get more data. 
Kirill Eremenko: Got you. Got you. Why analytics? You founded Edgeworth Analytics, the second company in 2019 when already data science is a hype. Everybody knows data science. It’s a very attractive term. How do you distinguish between data analytics and data science? 
John Johnson: Yeah. So what I think of it as is data science involves a whole host of both the technical machine learning, algorithms, programming capabilities, data capabilities, new data software. I mean, I think there’s a pretty broad sense of tools. I think analytics though is more true to what it is that I do in my litigation work in my consulting work, which is really focused on what is the question you need to answer to function as a business? What is the problem you’re trying to solve? And how do we get there? It’s not about selling you on let’s build a dashboard. Not that I haven’t [inaudible 00:31:04] dashboards, but that’s not data analytics. A dashboard is data. What the data means is the analytics piece and what I’ve found in my experience, dealing with companies very large sophisticated companies is the number of times that first of all, they actually don’t understand their data. 
John Johnson: They don’t understand in a large organization that the sources of data they need to draw upon are from different places. And they actually don’t understand that, how do you get to an answer? So analytics in my mind is our ability to serve our clients in a different way where we can say, what are the practical questions you need to answer? I’ve kind of felt that I mean, I can give you a great example. I did a pitch for a case a few years ago, actually. I’ve been thinking about analytics for a while, and I had done a pitch for a case for a fairly sophisticated company. And they’d asked about some things related to website analytics and stuff like that. How do you track who was on their website? 
John Johnson: And we put together a really sophisticated proposal with a lot of unique insight, how you would build statistical models to actually identify what the web traffic would look like, who the users were. Things that we thought were really insightful. And the response we got back was, “That was mind blowing what a great proposal. We never thought of that. But we’re just looking for a dashboard provider.” So that was a really formative experience because I realized that lots of people are talking the language of analytics, but it’s such a broad field. What we are selling is actually something very specific, which is you have a data problem or a problem. You think you have the data to process it, but you’re not sure you need help. How can we help you? 
John Johnson: It’s kind of like having a smart team right behind you to say, “Look, let’s work together. We’re not interested in selling you on, “Hey, why don’t we have a…” I mean, if you want to keep coming to us with problems, of course we want to help you, but I’m not trying to sell you a software package or a dashboard. What I’m trying to help you is say, “Look, I’ve got a pretty important targeted problem. I need some smart people to get me an answer that I can understand, and that can get me to where I need to be.” That’s what we do. And I think that’s what differentiates us. 
Kirill Eremenko: Got you. And just so then, why these three branches? So I understand the coronavirus impact study and I noticed on your website, there was one interesting one recently about where employers have to in some cases like you were doing analysis into, or someone from your team was doing analysis into whether employers… In what cases employers need to make the decision between diversity and experience? And that as I understood it, there was this legacy that maybe from a long time ago when diversity wasn’t a priority of an employer and they weren’t as careful about it, they hired a lot of people not in a non-inclusive way. And now those people have more tenure in the company. And now when they’re doing job cuts that that tenure is going to be taken to account. So they’re going to be stepping on the old minefield of not being diverse. So that was a very interesting analysis. 
John Johnson: Look, the thing that when you look at what we’re offering as our services, the HR focus is partly because it is a direct link to a lot of the work we’ve done on the litigation side. We have a very active labor and employment practice where we deal with companies a lot of times. So we felt like a really great place to focus in part was on HR issues because there’s a ton of them. And it’s a place where HR managers are not trained as much in data. I actually taught at an HR school that was part of where I was a professor. And so we find that that’s a great place for our educational function because that’s an area where data is important that people really need some help. We do general business analytics. We have one client that we’re helping them with retail store data, trying to assess how much inventory they should have, for example. 
John Johnson: And so we can use general data analytics in that sense with different businesses. And then the coronavirus is partly as I said, a function of the fact that this is a really exceptional time and we’re like, look partly just as a public service. And partly because people are really interested, we think it’s important that if we can provide some insights and get some good data work out there, we will. But I think the more important thing is when you are a firm that is used to doing what we do, which is parachuting into industries to help with specific problems. What we’re really marketing is not so much just HR or just business analytics or supply chain analytics. It’s that there’s a process we bring to the table that is very rigorous, that can work in a lot of settings. And that’s really our marketing hook quite frankly, is that what we’re here to do is help you identify your question, identify your data, get to the heart of the matter for you. 
John Johnson: We have something, we call it our data blueprint plan, but it’s basically we help lay out a blueprint for you of how to solve your problems. So that’s what we do. But those are just some areas we’ve called out. I mean, there’s also a little bit on polling because ironically I did write a book a few years ago on data in our everyday lives. And it came out in 2016. And what happened was when we went out to do media work on the book, the only thing anybody wanted to talk about was polling. And I’m like, “I don’t want to talk about polling all the time.” But as a result, we have a lot of it’s another showcase where we teach. So it’s this interesting intersection of where are the areas where we can provide value added and teaching about data, to translate into how we can help people learn about their data. 
Kirill Eremenko: Understood. So let’s talk a bit more about this parachuting into industries. I loved your example from the beginning where you had data on the chocolate bars from a whole year, and you had to, which you said you had to go and tour chocolate factories to get domain knowledge. And I hugely respect that about data scientists who practice that when you not only just focus on the data and get insights from that, you actually go in and get that domain knowledge. What does domain knowledge mean to you and what maybe what are your best practices for acquiring it fast? 
John Johnson: Yeah, so as an economist who is a statistician I’m not only driven by the data. I mean, the field that my PhD is in is econometrics, which is the merging of economics to statistics, right? So the idea there is that I’m a professional economist. I think about how companies work, how does a firm work? I have all this theoretical knowledge from my training, but then I have the practical knowledge. So what I need to do to be smart about data problem is I have to understand the business the best I can to answer the questions. So how do I try to do that? Well obviously it’s dictated by the question I’m looking at, but I’m often focused on pricing data. That’s a big part of what I tend to work with. And so what are the practical realities of the business? 
John Johnson: I always tell clients, I don’t have a monopoly on good ideas. I’m not here to come in and tell you, “I know your business better than you do.” I want to hear what you think about your business. And it doesn’t mean that the data will always support every person’s views. There’s a degree of confirmation bias you have to be very careful about. But to disregard what the business people actually think about their business is actually to throwaway, potentially a lot of useful information. Of course, I let the data ultimately speak, but I can craft better testing of the data, better structure to the data. If I understand at least where I’m starting from or where the business people are starting from. And that’s a critical translation. I mean, there have been times where I’ve worked for companies where they’ve given me their data, I’ve done an analysis. 
John Johnson: I’ve come back with a result that they thought was just completely, “Oh my gosh, that can’t be true. What did you do wrong?” And I had to help bring them along and explain the process and say, “No, let’s step back and show you exactly what we did. And we did this, we did this, we did this, we did this, and this is the answer. And this is why.” And you bring them on the journey with you and they get there. So for me, that foundational work and it’s face to face, it’s going in person, it’s showing interest, it’s listening. Those skills are critical. It’s not just the sitting in my little computer room, running my numbers in the back room. There’s a whole other part to it that you have to do. 
John Johnson: So I find getting to the right people, being focused on the question, being respectful of people’s times. I mean, people running a business don’t want to spend infinite time with you, but I think if you can get that foundation with some really targeted, focused questions, that’s helpful. So back again, not to sound like a broken record, but if you’ve framed the question correctly, that’ll really allow you to determine, okay, what is the domain knowledge I need? What do I think is really essential? Who is the right person to talk to so that I can get there quickly? 
Kirill Eremenko: Got you. For a business person listening to this, like a leader, a director, an entrepreneur, a business owner, at what point should they realize and how that they have a problem where they need to engage somebody like yourself or a consultant, external consultant, what’s a telltale sign that you can recommend for people to look out for that, “Something’s going wrong. I don’t know how to fix it. I need help.” 
John Johnson: That’s a tricky question. Because even as a leader of a company myself, sometimes it’s way after the fact you’re like, “Ooh.” I think what I would say is always if there is a practical outcome that you think is really important to your business and you’re having uncertainty or you’re trying to think of yourself, “Gosh, I don’t know that I have full information on this really important decision. I think that’s a good time to reach out.” I think another time you can reach out is just simply, yeah, after something’s gone wrong, you can reach out and we can try to reconstruct it. It’s always a little better if we’re more forward thinking. We often talk about with executives having a seat at the table. People throw numbers around all the time. But what we try to do is give you numbers that are actually going to answer your question and be defendable, not just a number for the sake of cherry picking a number that sounds good. 
John Johnson: So another place where we found when we’ve done our courses for people who have gone on site and taught in groups about HR and other analytics, sometimes it’s as simple as facilitating communication between business leaders. I’ve been surprised how many times you’ve been in a meeting and it’s like you talk about some concept about retention, let’s say, and some other person who’s in a different division’s like, “Oh, I have data on that too. This data would be useful.” Just getting those people talking about the different data sets they have. So one of the things we’ve tried to do is break down the barriers to people reaching out. What I mean by that is you’re always afraid to call a consultant because you think, “Oh my gosh, they’re going to charge me a fortune and they’re going to just turn this into a never-ending project.” 
John Johnson: And what we try to do is we try to be very focused. Part of our outreach, I mentioned our blueprint program. The idea is just simply if you have a quick targeted thing where you just want some advice, you can reach out to us for a small engagement just to talk through something. Now that doesn’t mean you’re going to get all your answers, but we’ve tried to facilitate communication with potential clients so they can understand better. Not everybody can embark on a giant data project, but maybe they just need to talk to someone intelligently for an hour about or a problem they’re having. 
Kirill Eremenko: Okay. Got you. That’s a great approach to get just some talking going. The difference between something, I don’t know like if this term is used before, I just thought of it now. So there’s a concept where people, consultants come in and especially with a big company, there is executives, like you said in that situation when you told them the answer and they’re like, “This is not possible.” What’s the difference between or how to identify a lip service analytics versus a hardcore numbers don’t lie analytics? If a consultant’s coming into my business and all they’re telling me is what I want to hear, I’m happy, but how do I know that they’re not bullshitting me just keep me happy? 
John Johnson: Look, I think there’s a few things I would look to, all right. So one thing I always look to is what is the background of the people you’re talking to? And I don’t mean that to be obnoxious. It’s just like there are people that actually do work in… Who are actually practicing this kind of work on a daily basis. Then there’s people that kind of are pretending and can put together a pretty PowerPoint, but a pretty PowerPoint alone is not enough. The other thing I try to say to people is if you’re getting data analytics work, if you can’t understand exactly what you’re being told. If I come into you and I start talking about, “Well I ran this child test and the child test rejects the pooling of the model and so then I did a kind of control for this or that or this.” And if it just sounds jargony and somebody is not actually explaining to you what they’re doing and what’s driving the results, that’s a pretty good sign that you don’t have the right person. 
John Johnson: I make sure my clients, when I talk to them, understand exactly what I did. Even the technical stuff, that doesn’t mean I’m doing a full-fledged tutorial on regression analysis and the [inaudible 00:44:01] or whatever. I’m not talking about that, but they have to understand the intuition and what the levers are underneath the analysis. If someone’s not explaining that to you in a way that is useful to you, they’re probably not the right person or doing good work. The best work, this is an old phrase from MIT, explain it like you would to your grandmother. And so I put such an emphasis on the explanation part because I think if someone can articulate something clearly and concisely, that is a very good sign they understand the problem well and they know what they’re doing. If they can’t, that is a very good sign that they probably haven’t thought it through rigorously enough. 
Kirill Eremenko: Oh my gosh, absolutely. Absolutely. I would love to ask you if there’s something you can explain on this podcast as an example of maybe like, I don’t know a simple, not regression, statistical model or something. But before that I just want to say we do the same thing. I completely stand by that because in our tutorials we have the Machine Learning A-Z course, which is one of the top selling courses in the world, top taken courses in the world on machine learning. And we specifically focus on the intuition behind the models whether it’s Support Vector Machine, Support Vector Regression, and so on, Bayesian inference and the [inaudible 00:45:22], we don’t do the mathematical part because people can learn that elsewhere. Intuition is very important. So yeah. So is there something that you can give us an example of how you would teach somebody? 
John Johnson: Why don’t I tell you actually something that’s real practical? We’re working on it right now. So as you know with the coronavirus issue right now, the United States is starting to reopen and one of the unique peculiarities of what’s going on right now is we have 50 states with different governors. We don’t really have a federal response right now. We have 50 individual state policies, which in fact they’re even varying within state. 
Kirill Eremenko: Wow. 
John Johnson: Okay. So it’s quite interesting. And so there’s this momentum building for, “Oh, the economy is reopening.” What we’re doing right now, we’re running a study where we have collected systematically County level data on the coronavirus cases across the country. What you’re seeing a lot in the news media is these anecdotal stories about, “Florida reopened and the next day there was a spike in cases.” 
John Johnson: Well that doesn’t make any sense on its face. Right? Similarly, you’re seeing stories like, “Oh, Georgia reopened and the number of coronavirus cases have gone down. Clearly there’s no problem. We’re back.” Okay. These super simplistic narratives drive us nuts. So what we’re trying to do is actually use what are called, the technical term that is called a natural experiment. We have all this variation now and we have time series data where we can follow a number of cases and we can follow the reopening. And in fact, we’re getting some cell phone data where we’re actually able to track movements to see how much the movements are increasing across states as they reopen. And we’re actually trying to link, okay, what are we seeing in terms of number of cases in the reopening to get a real estimate of that in a statistical sense? 
John Johnson: And so what you’re living off of is the variation in the opening, the changes in time and the number of cases. However, there’s a complication. The amount of testing is also varying at the same time. But what we’re noticing is that some of these states that are reporting, the cases are going down. It’s actually good they’re testing us. Think harder about these issues. So we’re putting together a study, we’re trying to disentangle this and look, this is a hard problem and I’m not pretending to have all the answers, but it actually is… So we have a question. We have different sources of data. The variation of the data looks like at a minimum it’s at the County level. And for some we have zip code level variation, we’re trying to get a good measure of mobility changes with the laws and what else could be driving the coronavirus response? 
John Johnson: So that’s the way we’re laying it out. And I can’t tell you what the answer is because we’re working on it. It’s a real time experiment. But that’s the idea. And so to me, if you could understand well, looking at the number of coronavirus cases on the same day as a reopening doesn’t mean anything because we know this occurs with a lag. The number of people that are flowing in and out will be a really important determinant for sure. We need to know what that does and we have to know how much testing there is because the testing critically is going to influence what the data says. I think that’s an important starting point. Now we could have a discussion. You might say, “Well what about mask wearing? Is there something on mask wearing you could put into the model?” Well sure we could. I don’t know if we can get it at every County level, but there are surveys on that. 
John Johnson: Or what about the prior experiences with respect to hospitalization or different demographics or different occupations? There’s lots of things we could do, but the point is if you can understand the basic framework and what we’re trying to live off of, which is the variation in the reopening over time so we can really look to see what’s going on. I think you’ve got a good insight into what our model might show. 
Kirill Eremenko: Got you. I’m starting to understand what you’re saying. I’m feeling it as if I came to you with a problem, you sit down when you walk me through, even before you’ve done the analysis, you’re walking me through, “This is what we’re going to do. We have this, we can add this data, this is the plan, this is the steps.” So I’m with you along the way. I’m your ally. 
John Johnson: Right. And you may have another idea like, well John, did you think about the fact that the experience in Europe, maybe it was different and maybe there’s some data there that could provide a good benchmark for a different thing. Yeah, of course. I didn’t think about that. We could bring that data into the picture. The point is to make you an ally, is a great word. I need to partner with you. I’m not here to be telling me what to do. Yes, I’m going to provide my expertise to get to an answer. I’m going to tell you what I think, but we don’t have the monopoly on the good idea. Right? The work product will be better and you’re more invested in what I’m doing if you exactly understand what I’m doing as opposed to me just speaking from authority and saying, “Well, we’re going to do this.” 
John Johnson: Granted, I don’t want you to tell me whether I should do a heteroscedasticity correction with zero correlate like [crosstalk 00:49:52]. Okay. But that’s not the point. But on the basic intuition and thinking through the issues, I think there’s some room there to really shape the analysis and I think that just creates better work product. There are times where people give me suggestions gently that’s probably not make sense to do for this reason, that reason. But you don’t have to be insulting to people like ideas are ideas. That’s part of the creative process. 
Kirill Eremenko: Okay. Understood. And while you were speaking, I had this interesting idea that you are starting Edgeworth Analytics. You are at a advantage in the sense that you probably have access to a lot of data sets that other people wouldn’t even have thought of. Right? You can- 
John Johnson: It’s interesting, we’ve been doing a lot… the data sets from our litigation are confidential so we don’t have access to those. But we do have I mean one of the interesting things we do is we’re constantly looking at what different data sets are out there. And I can tell you, one which we’ve been looking for is bicycle sales data because there’s a big series of stores in the US that bicycle sales have skyrocketed right now. So it is amazing how many unique data sets are actually out there for use. When the pandemic started, we actually used data from open table, which is the reservation system. We actually have a map we built, which shows how quickly the reservations declined as restaurants were closing across the country in those first two weeks. 
John Johnson: It’s actually a pretty cool graphic. And so I think part of it is when you’re creative about data, you can find, “Hey well I know about this data set or oh, this would be a really good data set to try to draw upon.” And I think that’s part of it is that when you’re a data geek, you’re just constantly looking for different data and when you’re working with a client, you help them identify that data within their own company. 
Kirill Eremenko: Got you. I want to talk a bit about HR because we’ve talked about coronavirus, about some data analytics. HR is clearly a big part of what you do in both companies. Do you have an interesting example of maybe HR case or an HR consulting project that you did just to give us a sense for what it means a analytics in HR? 
John Johnson: Yeah. So, look, I think what I would talk about generally, I’d be a little careful about how I catch it. But let me start out generally. The kinds of things that we do with HR analytics is someone has some kind of underlying labor problem. Let’s say a retention issue with their workforce or they’re just finding that they’re losing good employees and they don’t understand why. The kind of thing that we could help with is right. Okay, well what are the kinds of levers that could be affecting retention in that kind of situation? What are the competitive circumstances? Who are you losing your employees to? Are there pay disparities? Are there other industry factors? Are there dissatisfaction issues? 
John Johnson: So again, framing questions in a way that’s like, okay, I am seeing a practical business reality that I’m struggling with. There’s oftentimes we work on things related to diversity goals and things like that, or companies who try to be more diverse. And how did they do that? Things like that. So if there’s any number of different types of questions that we can help with. We generally find them in the recruiting, the retention, absenteeism or just diagnosing problems that wow, we’re having an issue and we don’t know why. 
Kirill Eremenko: Got you. And you can do all of that through data analysis. 
John Johnson: Yes. And again, same process though. I mean I have to understand the company and I have to understand a bit of the culture and I have to understand how the business works, right? Because there’re just certain factors. Different companies operate different ways. A company is just as diverse as a product. So just like some cases I was about before learning about chocolate candy bars, learning about company Xs culture and where they’re drawing their workers from and what they do and what their policies are is the input into a good data analytics, HR analytics project on the other side. So it’s the same process. It’s just applied more specifically to the labor part of the world. I mean, one of the reasons why I’m so into HR analytics is because one of my other specialties is what’s called labor economics, which is really the field of economics. About how the labor markets work, how a wage is set, those kinds of things. 
John Johnson: And so I find that fascinating and that’s partly what drove my interest in the HR piece of the world. I think it’s a really critical, overlooked part of successful businesses. And I also think it’s an area where the data can be so powerful, but I think HR professionals are a little bit more afraid to use data in maybe some other areas. 
Kirill Eremenko: Why is that? Why are HR professionals among those? 
John Johnson: It’s obvious they are people, people, right? I mean, who goes into human resource management? A lot of people that do, they’re worried about helping people and they have really great people skills. I said, I taught at an HR school, a top HR school. And even there, the one statistics course they took, those HR managers took was terrifying for people. And so I think it’s moved in a direction that is more data oriented. But I also think the other reason why HR people sometimes they’re a little bit afraid of data, is because they think it’s replacing the people skills. It’s divorced from the fact that you’re dealing with people and then all those other, I don’t mean this insulting soft skills that are so critical to being an effective HR person, listening, understanding what the problem is, getting to the heart of the problems. 
John Johnson: A lot of people interaction, the data does not replace that. The data compliments that so you can effectively diagnose the problems. But I think the pre disposal, and again, partly because data scientists do a pretty bad job of explaining things, they’re not sensitive to that. So I think that’s where the disconnect comes.
Kirill Eremenko: Got you. What are your thoughts on that? Do you think that in the long run, in that kind of 5 to 10 years data will replace the soft skills to their HR function or not? 
John Johnson: I don’t think it will completely replace. I think it’s pretty impossible. Look, one of the greatest challenges of running any organization is dealing with the people. It’s just the practical. That’s an age-old axiom. People make a company, that will never go away. But the data being a more systematic part of management, the data being a basis for making decisions will continue to be important. The ability to square what the data is showing with what we know about people and how we then craft management strategies I think is where HR is headed. And I think that’s going to be a hallmark of effective HR programs and HR systems at your companies. 
Kirill Eremenko: Okay. And I like Andrew Ng who’s a founder of Coursera, I like his quote that data is the new electricity. 100 years ago, only about half the US was electrified. Now I can’t think of a single business, even a farm in the middle of nowhere that’s not using electricity. With data science and data analytics is going to happen like that, but much faster within 10 years or so, I think maybe 15. All companies are going to be using this because of competitor pressure. So my question would be as a business owner over, let’s say a fortune 500 company or a large organization or any kind of organization for that matter. To what extent do I rely on consultants or external parties who come in and do the data analytics for me? And at what point should I consider building my own analytics function internally and can it help having the two for a certain period of time? 
John Johnson: Yeah. Look, I think this is the famous make-or-buy decision [inaudible 00:57:18] all sorts of, right. I think it depends on how much do you plan to integrate the data analytics into your work? Do you have the right people in place? Do you feel qualified to actually build out an analytics organization? I mean, one of the things that just in the existence, I’ll give you a new simple example from my business. Just in the 20 years I’ve been doing economic consulting in the 10 years as a professional economist entrepreneur with my own company, one of the biggest and most difficult hires we make are PhD economists. And in the time I’ve been hiring, it used to be that my main competition for PhD economists were universities. [crosstalk 00:57:54] were coming to consulting or you could be any university. And during the course of last 10 years, suddenly these maybe you’ve heard of them Amazon, Google they’ve started to hire economists and good economists. 
John Johnson: And so I’ve found now my competition for people tends to be more of those companies. Right? And those are the cream of the crop. They’re pretty sophisticated buyers so now if you’re a company that specializes in something else, the undertaking of building a data analytics organization could be challenging. I think ultimately if you’re a big enough company, you want to integrate it, I think that is definitely part of what you’d want to do. But I do think consultants will be pretty important even in that context to validate what’s being done, to get a good sense. Even if you have the most effective data analytics organization, you still need sometimes that outside set of eyes to help. And then what about all the other companies that aren’t fortune 500 companies that just can’t dedicate the resources to build an internal analytics function themselves? 
John Johnson: We have found, we have had clients that we’ve worked with where we’ve taught them how to do certain things. So there’s also that collaboration where the handoff of like, “Okay, well we did these initial analysis for you. Okay, here’s how you could do them.” I’ll tell you, more often than not, they come back to us anyway. And so that’s just a practical reality. I do see that as a trend, but I think there’s just a whole host of companies that just aren’t going to be in a position to build out organizations and effective organizations and there will be some that do. And so, yeah, the Amazons, the Googles of the world, they have exceptional data analytics, but other companies may not. And so it’ll evolve and 10 years from now, who knows? But I still think there will be a need for outside people to come in and validate to look at or to help and all those companies that just can’t do it. 
Kirill Eremenko: Fantastic. John, I just looked at the time, it’s been crazy amazing this hour, I have a whole list probably another 10 questions I would love to ask you, but we’re running out of time. So I’ll just ask you one last question before we go to the contact details and where people can get in touch with you. And that question is what would you like for the coming three years, let’s not go five or 10 years, just three next three years with this whole coronavirus situation, with this mayhem happening in the world. There’s a lot of business owners who are scared, who are fearful, who are trying to hold their resources as closely as they possibly can and they’re trying to survive in this winter that’s come upon us. What would your wish to those of them listening to this podcast be? What would your recommendation or some inspiration for them right now? 
John Johnson: Well, I’ll try. I mean, look, it is scary. I mean I have a successful business and I’m scared and my colleagues are scared. I think everybody needs to be empathetic to each other first. Back to just a simple point. This is unprecedented. Economists use the term unprecedented way too much. This is unprecedent, right? If there’s good news, I think there’s two things. First because this came on and because this is a health crisis married to an economic crisis, if we can get the health crisis under control, hopefully the economic crisis can be then controlled. So hope first of all for eventually vaccine treatments, some of those things are really important to a recovery. But as a business owner in this environment, what I would say is it is often the case that in these adverse times is when most innovation comes. 
John Johnson: If you’re thinking about it as an opportunity, what can you do in this environment to either retool your business, to think about what is a unique angle for your business or just what are some things you’ve probably wanted to think about that the good times might have obscured? I think people that take the downtimes and try to use them to aggressively set themselves up for the backend, are the ones that end up so much more successful in the long run. So, I mean that’s the strategy we’re doing at our firm is we’re like, “Hey, what can we do proactively to set ourselves up to get through today and then for the future, be ready to take advantage of the new opportunities we weren’t thinking about?” 
John Johnson: And so I think that gives me some comfort, at least as a business owner, that I’m doing what I can and also just be kind to yourself. These are really difficult times. I’d remind myself that. I said to one of my colleagues yesterday, “I’m not responsible the fact there’s a recession right now.” I think it’s, know that other people are hurting too. You’re not alone. 
Kirill Eremenko: Fantastic. Thank you. Great advice. John, where can people listening to this find you in case they’re interested to read more about the amazing studies that you’re doing or your team is doing or maybe even get in touch for some consulting projects or discussions? 
John Johnson: So if you go to our website, www.edgeworthanalytics.com and we also have the economics edgewortheconomics.com. You can send me an email there. That’s the best way to get in touch with us. That’s where all our studies are and descriptions of our work. So that would be the place to look. 
Kirill Eremenko: One word, edgewortheconomics, edgeworthanalytics.com. 
John Johnson: Yep. Two different [crosstalk 01:02:43]. 
Kirill Eremenko: Got you. Awesome. And is it okay for people to connect with you on LinkedIn? 
John Johnson: Absolutely. 
Kirill Eremenko: Fantastic. Fantastic. John, thank you so much for your time. It’s been a huge pleasure having you on this show. 
John Johnson: Thank you. It was a really great conversation. Take care. 
Kirill Eremenko: There we have it everybody. Hope you enjoyed this episode as much as I did and got lots of valuable takeaways. For me, I really enjoyed the part where John was talking about data science or data analytics in the space of litigation and what that means and how those proceedings go and how they are able to extract insights through data and then present it in front of a jury or in front of a judge. I found that very interesting. It’s [inaudible 01:03:27] analytics that we don’t often think about it but as we could see from or hear from what John was describing. There are lots of interesting data sets, lots of challenging problems that need to be addressed that lots of huge cases that are happening in the space from the last few years. So it was very interesting to get [inaudible 01:03:46]. 
Kirill Eremenko: Of course, the business side of things. I think that was very valuable for me personally as a business owner I was sitting on the edge of my seat listening to what John was saying and hopefully if you’re a business owner or if you are in the management of a company or in the management team of a company, then we’ve you got lots of useful takeaways as well. And if you did get lots of valuable takeaways and you’d like to get in touch with John or follow their company and see what else they’re doing, then check out to their websites and their profiles at edgeworthanalytics.com or edgewortheconomics.com. You can also connect with John on LinkedIn. And speaking of all of these amazing resources, you can find everything that we talked about, including all these links and any other resources that we mentioned, plus transcript for this episode plus the video version for this episode. Yes, this episode is available in video. 
Kirill Eremenko: If you’re just listening to the audio on Spotify or SoundCloud or iTunes, you can also find the video on our website all this available at www.superdatascience.com/369. That’s www.superdatascience.com/369 and make sure to check that out and connect with John, he’s a great thought leader in the space of their analytics that you would probably want to be connected with at least on LinkedIn. 
Kirill Eremenko: And one final thing, if you know somebody who is a business owner, a director, a manager at a company, an entrepreneur, a founder and they might need a refresher on economics or data analytics or they might benefit from some interesting advice that John had to share today, send them this episode. It’s very easy to share. Just send the link www.superdatascience.com/3-6-9 and they’ll be able to watch the video or listen to the audio. Find everything that they need there. 
Kirill Eremenko: There we go. So that was us for today. Hope you enjoy this episode with John Johnson and talking about data analytics specifically in the space of litigation and data analytics consulting. If you find it useful and I look forward to seeing or hearing you back here next time. And until then, happy analyzing. 
Show All

Share on

Related Podcasts