406 Data as a Leadership Superpower: How to Fuel Strategy, Innovation, and AI with Justin Evans, Author of The Little Book of Data

***DISCLAIMER: Please note that the following AI-generated transcript may not be 100% accurate and could contain misspellings or errors.***

Mahan Tavakoli: . [00:00:00] Justin Evans, welcome to Partnering Leadership. I am thrilled to have you in this conversation with me. 

Justin Evans: I'm thrilled as well. Thank you for having me. 

Mahan Tavakoli: I can't wait to talk about the little book of data, understanding the powerful analytics that fuel AI make or break careers and could just end up saving the world.

I. Now that's a big promise, Justin. But before we get to that, we'd love to know a little bit more about you. Whereabouts did you grow up, and how has your upbringing contributed to who you've become? 

Justin Evans: I grew up in a beautiful town in the Blue Ridge Mountains called Lexington, Virginia where my father was a professor of English literature at Washington Lee University.

And I think the fact that our household was so steeped in literature and it was in the south, really affected me in a [00:01:00] couple ways. I. First the English literature part gave me a respect for language and a desire to always find the acute, punchy way to describe something that someone suddenly gets with a metaphor that I might not have otherwise had.

And it, it served me well later when I started working in data and technology because. Often the person I'd be talking to was a business person who was expert in their business, but not an expert in data technology. And so is often it ended up being my job to explain stuff to those folks and I became the go-to person so that storytelling style that's very southern and the growing up with the narratives and the literature.

Put me in that position weirdly, actually, because when you think of someone in data and technology, you think of someone who studied computer science or statistics, not in my case. , I am the English major representing, 

Mahan Tavakoli: . Now you start out your book.[00:02:00] 

With a simple question that I think is relevant for us to level set with what is data. 

Justin Evans: There's a dictionary definition of data, which is, it's any set of facts that can be used for insights or analysis, period. Maybe a little bit boring. You can go the angle that it's. The fuel. It is the fuel per the subtitle, the fuel for artificial intelligence.

It is in a way the senses for artificial intelligence. That is to say, when an AI based self-driving car is looking out at the world, it is perceiving the world through voxels and lidar and translating all of that into data that it can overlay with a mapping application and do its self-driving thing.

And data is the way that AI perceives the world. The, I like to, in, in [00:03:00] the book, I went pretty deep, arguably too deep and really thought about data as something really fundamental to humanity, which began no sooner and no later than the moment we've learned to count. And there's this really great passage in Bertrand Russell's introduction to the philosophy of mathematics, where he talks about, Dr dramatically, like as if it's a moment.

He talks about as if the moment where the where a human being perceived three people, let's call them Brown Jones and Robinson. And he perceived this trio of people. Not just as Brown Jones and Robinson, but as a trio and a trio being something that you can abstract from any three things. A trio is something that all the number three is the thing that all trios have in common.

And I got really, I got blown away by [00:04:00] that because it seemed to me that there, there has to have been a moment in humanity where yes, we observed that. We're seeing these three things over and over again. So maybe it has a property outside of just. Brown Jones and Robinson, and that's something I can observe.

And then that observation is something I can write down or otherwise preserve for communication. And in that moment was data borne. And data is a way of us describing our reality in a way that, yes, it's reductive. We're taking three people and reducing them from complex people who love and hate and have relationships into the number three, but.

It's this directed way of perceiving the universe and it's become the new electricity. It's this power that's all around us that enables lots and lots of applications that is critically important to life in the 21st century. 

Mahan Tavakoli: That's a great way of thinking about it. And I've seen cartoon images of.

For [00:05:00] example, people walking and oozing data in that our existence, we are giving off lots of data. And you refer to the fact that to a certain extent data is like electricity or some people talk about data being the new oil, you talk about how data can become a way of crystallizing complexity.

How can data help in doing that? 

Justin Evans: Yeah the framework for the book is that I come up with sort of four data superpowers, one of which is crystallizing complex information, and then another eight or nine kind of core, core ideas and. Crystallizing complex information is something that's just inherent to data.

And the story I use in the book to explain how this works goes back to the guy who's founded Standard and Poors, and his name is [00:06:00] Henry Varnum Poor, and he had a. Very, not quite a humble upbringing. He was a lawyer, but he was the editor of a B2B Trade magazine, and he was, it was a B2B trade magazine about railroads and it was just like, I don't know, variety or billboard or whatever.

It was, a trade magazine. Not exciting. And he was so passionate about railroads because he believed that railroads were the key to American growth and the key to connectivity in our world. He grew up in this little tiny nothing town of Andover, Maine, where that had no, no market. And he thought, railroads is the mechanism that we would put something like connects a place like Andover, Maine to the world.

So he got really passionate about railroads, but he saw this problem that the railroads were expanding out into Colorado and California and Maine, and Georgia and Texas and the bankers. Who were funding them were in Philadelphia and New York and Boston, and [00:07:00] they had no experience with these railroads.

And so what he did is he took the, this at the time, very complex and still complex technical world of railroad engineering and business, and turned it into a manual. He would mail a, a questionnaire to all the railroad heads and write it down, publish it as a manual and give it to the bankers.

Then the bankers felt that they understood each railroad company to the extent that they could fund them and fund the expansion of, the American Dream which Henry Var Andour really believed in. And of course, Henry Var Andour successors crystallized that even further down to triple A bonds and triple B bonds and junk bonds.

And to me, it's a wonderful example of how all the complexity that, of the engineering, the business the land itself can be crystallized into a single number like AAA for a banker to understand. And data has that [00:08:00] power in many other facets of our life. And it's part of why it's a superpower because it allows us to have an influence on these impossibly complex thing.

Through a simple means. 

Mahan Tavakoli: That's a beautiful example in describing how, , data can help us understand the world and simplify it in a way, it becomes much more accessible to us now. There is use of data, therefore, in simplification and in understanding, but . How can you make sure that you are using the right data and you are channeling it and simplifying it correctly? 

Justin Evans: Yeah it's a critical question to how we do our work as data people. And it's something that requires vigilance and talent [00:09:00] on the port of, on the part of the the person wielding the power.

I think it's, part of that is that we have to first acknowledge that data is not reality. Data is a description of reality. The a lot of, some of the fun ideas about data come from earlier parts of our history with this census. And I profile this fellow Herman Holis, who in effect invented the computer in order to process the census.

And he called. The census, or maybe it was just a term of art at the time. The census takers were enumerators. A term I really like for some reason. Because it's, its the job of a person to go into a town, literally to walk into town where, there, there could be a Faulkner novel happening in this town.

There were all these relationships and history and warfare and, whatever, feuds, and they turn it into a thousand people, 70 or white. 30 are black, some are rich, some are poor. Some have [00:10:00] kids, some have, some don't. And it just reduce it to a handful of attributes. And we have to be humble about that act of reductiveness.

When we do our work. And the, there's a, there's another section in the book where I talk about. What makes a good and a bad data person, and I do have some examples of bad data people in the book. But the people who are good are those who have usually the term is used with lawyers or financiers, but I think that data people have a fiduciary duty to their client and the client.

There's always a moment in, in, in the relationship of a data person and a client, and that client could be a pain client or it could be an internal client or just someone you're supporting. And there's a moment where you sit down and you as a data person, really have to sit there and understand what the person's trying to accomplish.

And in that process, you're trying to see the world through their eyes. And that's when you take on your fiduciary responsibility and you [00:11:00] go. Swimming out into the numbers and you know that you're gonna do well by your client when you're respecting that kind of core vision of what they're trying to accomplish.

And it's, there's no magic formula for it. I think it's a matter of talent and skill and care, which is why it's important for people to think about ethics and mission when they are thinking about using data, because you can use it poorly and you can use it wrongly, 

Mahan Tavakoli: absolutely. Now, every client that I deal with, they are in essence a wash with data the question becomes, okay, data is oil, data is electricity. We've got all this data. Where do we go with this? What are the types of questions we ask? What is the type of formulations?

Justin Evans: I think about data, and I admit that this is a personal approach as I think about data through the lens of innovation versus answering questions. [00:12:00] So the way I've used data in my career has been to help companies, but not only companies nonprofits as well, use data to fulfill their mission and have data be a part of how they do business.

And one. Example I use in the book that I really like, even though it was not my project, it was watching Amazon. And for that matter, Walmart from afar learn how to use their e-commerce data, their online sales data to create an advertising business. And in that sense, two things were happening. One, they are a company who needs growth.

Everybody needs growth. They realized that the data they have, that where their customers are transacting with them was giving a signal about what other things those people might like to buy. And the fact that those purchases were happening in their internet environment means that not only could they show them ads on those internet pages, but also [00:13:00] they could, once the person had seen an ad and then bought something.

Amazon and Walmart could go back to the advertiser and say, oh look, somebody saw your ad, and then they bought something, which is the dream of every advertiser and I. Not only that, but Amazon and Walmart very cleverly built up an advertising business in a way that benefited all their stakeholders.

Because if the seller of something online is able to put an ad in the e-commerce environment and then have somebody buy more, then in theory the person who bought more is happier because they found something they wanted. And the person who is selling is happier because they sold more of that thing.

And Amazon is happy 'cause they're getting paid on both sides and. It was such a successful model that for Amazon at least it became 8% of their whole business. And so I wave my writer's magic wand and say, okay, this is the rule of 8% that you can generate 8% from data that's lying around in your environment.

[00:14:00] A nonprofit that I advised for a time was a group called Denver Urban Gardens, who creates community gardens in unused and unattractive land. And they wanted to do a project under, have underway a project to measure the impact of putting in an urban garden somewhere. And the point I'm telling by telling this, giant story about Amazon and this.

Small community story is that whatever your mission is, there is data lying around that you can use to either measure the impact of putting an urban garden into one corner of your town, or measure the impact of millions of advertising and e-commerce transactions over the course of a year. But the data should be embedded into the.

Organization's mission, and that's how I come at data. It's [00:15:00] less, I'm overwhelmed with data. I need to decide yes or no. And in this one instance, I needed to make a decision. It's more like, how can I systematically put this into my organization so I'm doing something better and smarter all the time using data.

Mahan Tavakoli: Therefore, what is the advice you give to clients and organizations on how to do that? Where should they get started? Where should they go in order to become more data-driven? 

Justin Evans: , Two things.

. One is that the general manager, the leader of the business is almost always gonna have an intuition about how they can use data. It's just a spark. They know their business. They don't know data, but they know their business. So I walk into the room and I say, how do you think you can use data?

And I'm going to believe what they say, or mostly believe it. They might not understand the how, but they will usually understand the what. That's a vision and that means that they're committed. The second [00:16:00] thing to consider is what kind of data do you have? Every organization exists for a reason. You have conquered you if you exist and have continued to exist.

You have conquered some space where some stakeholder. Cares about you and they like the way you're doing something. You have your audience, you have, you own this space, so your audience and your advertisers or your stakeholders. And once you've conquered that space, that means you have expertise over your stakeholders.

You have expertise over this market, or it may even be a geographic location, but you know something about that and that knowledge means data and that data can be harvested by you. Either by licensing somebody else's data or getting free government data about that thing, or maybe you're gonna commit, which is a commitment to generate data about that thing.

And once you have those two things and those two ideas, you really [00:17:00] lack nothing but resources, time, and will joking to get at the opportunity there because it's, that's really the kernel of it. And those two steps, I would say, are really the most difficult. The last step then is to step into that world where you're accepting that you're gonna be dealing with data now, not what you were dealing with yesterday.

And that requires quite a lot of humility and confidence, weirdly, at the same time. So the humility part is I'm not talking about what I already know. I'm talking about data. So I have to be humble and I have to accept the guidance of experts. The confidence part is never, ever let the experts tell you what to do, because you have to be able to translate.

This is where people get lost. You have to be able to translate the concept back into business, back into [00:18:00] English, or you're being led astray. You really do see people get enamored with the balderdash and the jargon and the acronyms and the great client, the great general manager is one who can firmly be in both worlds at once.

The world of their expertise that they're completely confident in, and they demand that their experts translate everything back into and an acceptance that there are new rules when they're dealing with a new medium. In this case, data and ai. 

Mahan Tavakoli: That is beautifully put. So the way I visualize it, , is that there is a sense of direction that initially is there.

, You come up with a map based on the data. But recognizing that map is not exactly the territory. It goes back to this conversation about AI's ability to [00:19:00] augment rather than fully replace judgment in that it pulls together the right data to give you some right framing that you need to believe in.

I. However it requires your judgment and experience to then decide based on that. 

Justin Evans: Yeah one of my favorite people that I interviewed in the book an entrepreneur named Adam Green. , And Adam Green's father had died of dementia, but Adam was sure that his father had died of dementia due to loneliness. He was sure that after his mother died, his father had become increasingly lonely. And loneliness acts as a kind of.

A self attacking mechanism, if you will. It's like a stress response that then can then make diseases like dementia worse. And so he was sure that his father had his demise had been accelerated by [00:20:00] loneliness. So Adam set out to find a way to score seniors in senior homes and detect whether they were lonely and.

He had to pat onto the cover the couch cover cushions and pat his pockets and figure out how he would do this. And he found that there was already a loneliness score that was a standard, it was a seven question test, which was already too long. So he shortened it to three, but then somehow who'd go into the senior homes and he would do these programs and see people become less lonely.

But it wasn't reflecting in the score. So he had to re-attack it a different way because he realized that there was sample bias. If I ask you a question, whether are you lonely, you're gonna defensively say, oh, what? I'm not lonely. And that's what was happening with these tests. And so he had to reframe his toolkit, which is I think where your question started [00:21:00] based on.

A passive approach to the question. It had to be kinda observational. So the way he solved the problem, and this is where the chapter ends, is he came up with a way to simply do a phone call to the seniors, ask them about their day. They were able to use words that they were using actually correlated to loneliness.

And so if they use certain vocabulary words and certain intonations and other things that were very subtle, they would be able to passively score the person who's loneliness. And that would take away the sample bias and therefore they would actually be able to score someone on a before and after about how lonely they were.

And it was this kind of marvelous trick that they were able to pull off by. Just coming at their data toolkit from a number of different angles and got this almost kind of magical result. 

Mahan Tavakoli: , What an outstanding example it takes that iteration before you are able to actually get it right. Now. [00:22:00] AI has also made interpretation of data a lot easier. It drives on data, but it has also made it a lot more accessible to whether it's individuals or small teams and organizations.

I would love to get your thoughts on. What role do you see AI playing in data rather than data playing in ai? Data is the fuel and the driver of ai. What role does AI play in data? 

Justin Evans: I think it's a very acute and insightful question. I think AI is going to weirdly accelerate the cycle of creating data. So right now you'll read a lot of.

Headlines now about the big AI companies running outta data. I don't have the numbers in front of me, but it's, there's, a cajillion. Bytes of data that you [00:23:00] can get from these big web scrapers, and then you can license or steal all the books and then you can steal or license all the magazine articles and eventually your scraping capabilities are gonna effectively outpace the ability of humanity to create new content and you will effectively plateau.

But I also, so that's one narrative, but I also think that there's. What AI can do is create is create data. So if you have some AI application that has perception capabilities that AI can in effect, write down what it's perceiving on an ongoing basis and thereby. Generate a larger and larger pool of data that can then be processed either by its own application or somebody else's application.

And the, when you to help people who haven't thought about it a lot, the what the kind of, the core [00:24:00] way that AI is taught to perceive is through the role of very poorly paid data annotators. And, an hour day, there would be a kind of proverbial, proverbially, horrible summer job of like data entry, right?

It's that kind of thing. So you're basically in a computer program. You're sitting there and you're shown a picture of a dog and you write DOG enter, and then that's the job. And you are teaching image by image and label by label. You're labeling these images. You're teaching the computer to speak human.

And that is an effect. And that exercise became the famous ImageNet database that Fei Fe Lee, who's now at Stanford generated, that created, that was the heart of so many ai, visual AI applications. I. A ImageNet I'm going down this down some tangents, but ImageNet ended up being like a 35,000 image [00:25:00] database that could, describe porcupines and different types of shag rug.

And it was just super obscure. And of course, it's only grown. So now we can teach the AI itself to be a data annotator. So it can say, this is a rose, this is a pink rose, this is a yellow rose. This is a a Harris rose, a Queen's rose. Create more data and that can feed the cycle.

And it's something I'm doing in my own work now and it's it's a powerful capability. If you just think broadly and generally that we have so many smart devices in the world. All these smart devices are in effect little independent data generators. And with. The power of ai.

We can classify and label and organize all this data and just keep the flow, keep the supply going at a tremendous rate. And so that's like a high level thing that sounds journalistic and paranoid, but I think from a practitioner's point of view, it's actually pretty great because you can [00:26:00] direct that of course, right?

It's not just about observing the world for no reason. You have a problem statement. If you're trying to do something great for your local park, and you could set up a little device that can measure the number of people coming in and out of the park, and you're in effect using AI to generate data to serve your purpose.

Mahan Tavakoli: , There are lots of different potential uses for every type of business and organization. When you think about that generation of data, I would love to. Touch on a couple of other elements that are in your book and in your title as well. You say make or break careers. How will understanding of data and analytics impact our careers?

Justin Evans: I had a moment a few years ago in my career when I was joining a traditional company that wasn't really digitally minded, wasn't really data minded, [00:27:00] and. Of course the whole advertising and marketing industry was and is becoming more data and AI oriented every day. And there was this kind of moment where people just started to bounce out of the business because they didn't get it.

And I say that completely without judgment. They were people who bounced out of the business who I thought should bounce out of the business because they were curious. There are people who bounced outta the business that I thought were the best that we had, but they. Just couldn't bring themselves to learn a whole new thing.

And I actually thought it was a bad thing. There were a lot of people who can be arrogant about it and say the dinosaurs, but I don't think that's right. I just felt like it was a, it was just a missed opportunity that it couldn't. Bring everyone along and let them apply their talents to a data-centric world.

And it was actually one of the reasons I wrote the book because I thought if, because I was an English major, [00:28:00] because I like to explain things to clients because I like to simplify things, maybe I can be the person who can write down these four superpowers and eight key ideas and have those people not bounce out and instead embrace it.

My goal in the book is to have all these data stories, not told by me, but about these other seekers and practitioners. If I can have all these stories wash over you, it's it's I think about it as almost like reading like a whole bunch of Sherlock Holmes stories. Like by the end of that, you're gonna be thinking like a detective.

You're gonna be like, oh, you have chalk on your sleeve and therefore you're a school teacher. And I really want people to be thinking about those data use cases and data applications by the time they finish the book and not be that person who will get bounced because you didn't embrace it and weren't curious.

Mahan Tavakoli: I appreciate that and I find that with you sharing those stories and when people read the stories. It makes application and understanding of it a lot [00:29:00] more accessible. It's a little bit again, like the analogies that they use in a lot of startups. When, for a while everything was Uber of this, Uber of vet because when you, once you understood Uber, you were able to understand whatever else they were talking about. So I think. With reading these different examples, you can then visualize the role data can play in your organization a lot more effectively than just understanding the data structure by itself. So that's why I really appreciate those stories. That other thing that you say that I think is relevant to this you say data, you describe it as ideas with structure.

Justin Evans: There is a repeated framework that's inherent in every data story where there's a hero who is the person who is asking the question. [00:30:00] They perceive some happy valley in the distance that reflects a world without obstacles in this certain area.

And they have these obstacles in front of them that they want to get rid of. And data becomes the power of the sort of the, to keep going with that kind of mythic. Construct data becomes like the superpower that they take on in order to overcome the obstacles and achieve the Happy Valley. It's very much like a fairytale.

And actually, as I wrote the book, I I got looser and more comfortable in the structure and the stories came more easily to me. But the, one of the, one of the, and often but it's always about these people solving a problem. One of my favorite. People I interviewed in the book was Sharon Green, who is the head of statistics at the Bureau of Communicable Diseases in [00:31:00] New York City, and she, I.

The Bureau of Communicable Diseases is this totally nondescript building in the middle of Queens, and you're so happy that they're there because what they do is they track like West Nile and leprosy and gonorrhea and salmonella, and they are there to identify these outbreaks and stamp them out. And so Sharon Kay Green was the head of this group, and one day in February, 2019, she got brought into a room that had this meeting called the doc of the week.

So the doc of the week is they would bring in some expert and they would tell them like, what's going on? So one day in February, 2019, the doc of the week was someone who was like, yeah, there's something happening in China and there's we, it's deadly and we don't have a test for it, and there is no cure.

And the team was of course, chilled because New York City is a city of 8 million people and there are a million foreign visitors every year. Many of whom come from China, and they just knew it was coming. [00:32:00] And what this team had to do was take a disease that had no test and no cure, and figure out a way to identify hotspots and stamp it out.

And in this case, it truly was a heroic effort on the part of this team as they, they actually just made a very kind of small, herculean effort. 24 7 labor. And I don't think that any of them saw the sun for months and months. And they were dealing with true crisis with New York City having the, one of the highest death rates in the world, higher the deli, higher than Hong Kong, higher than London.

And they realized that they could apply the work they had done for salmonella to covid, and all they had to do was so the main way you detect a hotspot. You take a, some space, a square mile in New York City and you say, how often does somebody get salmonella in that, in [00:33:00] a square mile in New York City?

And let's say it's, 1% of the time. And then if you have a month where 2% of the population is getting salmonella, then you know you have a hotspot. And then they send in a team and the team does an investigation. So how could they apply that to Covid where there is no trend? There is no baseline.

They don't know the 1% thing. It sounds very simple in retrospect, but what they figured out how to do was to take the testing the positive testing proportion. In other words, if the average across New York City was 4% po testing positive, but in a period of time, in a space there was 8% positive testing, then they knew they had a hotspot.

And by the way, they had to persuade the city authorities that this was the right way to approach it, which was no easy matter. And so once they had devised the technique and then once they had persuaded the New York City officials, then when they identified these hotspots through, by the way, this elaborate and brilliant system of gathering up [00:34:00] lab tests from every hospital in the city into a secure computing environment where if you know the special login, you can analyze it, they were able to identify the hotspots and then floods that zone with PPE and testing equipment and stamp out the hotspot.

And to me it was, and at the end, I got so emotional when I was writing the books, I was meeting these people and Sharon Green was just like, that was my wartime service. And she's I do it again. I. What a badass she was such a badass, and she was like this, she was five, six, with a PhD.

It was, not, she didn't look like a badass. She just was the baddest of the bad asses. And I just love meeting those people and it really felt truly heroic to me. What some of these people's work was, 

Mahan Tavakoli: That is beautiful and. It goes to the point that you make, that data could [00:35:00] just end up saving the world as well . Now, would love to get your thoughts, Justin, . What is one shift you hope every leader makes after reading your book or after listening to this conversation? With respect to. Data and leading with data in their team and organization.

Justin Evans: I believe that data is latent in every organization and its use is latent in every organization to unleash growth. And it doesn't necessarily mean growth in sales of a course or profit or, but of course it could be a nonprofit, it could be an academic mission, it could be a social mission. And I think the data is there to help you achieve your goals.

And my sense of mission personally is, can I help you [00:36:00] see that opportunity? And I. If nothing else, feel confident after or by reading the book that you won't be intimidated. And I think that's a real, I think that's a real issue. I think that people have four levels of data denial. The first is that, it doesn't really affect me.

The second is that this is really intimidating. I. And the third is, even if it is, even if I can get over my intimidation, it's too hard for me. And then the last is I think data is evil because I can see it offering me this, I can see the ad on the internet of the sneakers that I saw two weeks ago and it's following me around the internet and it's creepy.

And if you can get through those four things and then you can get confident. That's what I want. I've I personally have experienced many times in my career the [00:37:00] data bully. There's someone who knows more than you do, and they want to feel big by making you feel small, and they're gonna use all the acronyms and they're going to dazzle you.

And really what they want is for you to walk away thinking That guy knows more than me. I'm gonna leave him alone. And. It gets back to my prescription for leaders that you have to be able to translate these data ideas into English and into business and into your own language because they do. There is no data concept too complex for you to do that.

And once you've taken on that confidence and that fluency and that power. You are in a position to capture your 8% growth. You're in a position to use data to fulfill your mission more. And because data is inherently a tool of scale, it's guaranteed that it's going to drive a lot of power in what you're doing once you get through the hard part [00:38:00] of harnessing that innovation.

Mahan Tavakoli: That's a powerful point and I love that you come back to that potential 8% growth in that. The way I visualize it that often used analogy of data being the new oil is that every organization is on top of an oil field. Picture takes picture somewhere in the Middle East, wherever you want, where you might have tapped into couple of the oil reserves, but there's lots of oil.

Under the surface. Some are deeper, some are closer to the surface, but there is lots of oil to tap into with the right understanding and right approaches. That's why I really appreciate your book and your insights. Justin, where can the audience find out more about your book and also follow your work?

Justin Evans: My main social presence is in LinkedIn. You just look me up. Justin Evans Harper Collins Leadership or Samsung where I work. [00:39:00] And I've also got a substack the data story on Substack.

Mahan Tavakoli: Really appreciate the conversation, Justin and your book, the Little Book of Data because it does help leaders understand the power of the analytics, that fuel AI as you put in the title, but also make it a lot more accessible through the stories that you tell.

Thank you so much for this conversation. Justin Evans, 

Justin Evans: a real pleasure. Thank you 

so much.