pretty old guys. That's a pretty long story, but I'll try to be somewhat brief When I was coming out of college. Um, I I like a lot of people. I didn't know what I wanted to do when I grow up. I still don't know what I want to do when I grow up. Um, you know, 25 plus years later. But in any case, um, when I was coming out, So the thing that you did, where I went to school, there were two kinds of companies that recruited on campus. Uh, they were investment banks and there were management consulting firms, um, and which is, by the way, a terrible way to pick a career just cause, like, that's your recruits on campus. But in any case, I didn't know any better. Um, and I was pretty sure I didn't want to be an investment banker. And so it became a management consultant. One of the things I want one of the stupid things I did that actually turned out to be great for May Waas. Um, I had this picture of, by the way, in my head of a management consultant going around to businesses and putting on a hard hat and telling people have to do their jobs. And the absurdity of that never really struck my 21 year old fancy. But in any case, one of the things I did that was maybe seems stupid of the time I regretted at the time, but turned out to be great was I became the guy who actually learned Microsoft access. Um, And so whenever we were doing some work that involved data work, I became the guy calling that, um, not that these days we consider Microsoft access to be deep data work, but this was kind of before the dawn of the big data revolution. Um, and so I got to work on a lot of, um, problems with, you know, that required quantitative rigor. Um, fast forward some years, uh, after because after a couple of years of being a management consultant, I realized that going to school is a lot more fun and and you know what's business school and came out and and the first up I took, uh, can business school waas um this was him in the late nineties, and I traveled in India and I had really? You could see the India was poised to take off, um, and said they want to go there, and And I have been talking to a U S bank. Um, and they made me an offer. I said, you know, I'm not gonna take this offer because I wanna I want to go to India. I'm releasing in India, and they were very clever. They said, Well, you know, actually, we've been looking to start a division in India. Um, and go there, and you can you can start an operation. I told you this to be a very long story. In any case, I this is relevant for this reason because, um, when I went there was living in Mumbai. It and on late nineties, um, and the challenges of fascinating data challenge, um, which I never really thought When I went there, I didn't think it was in the data challenge, but they were, at the time about Ah, a little less than a billion people in India. Um, there were, um there were no credit bureaus. You think about how you figure out who you give a credit card to you in the US? It's based on credit bureau, Which tells you where that person's paid their bills. And there was nothing like that. India, Um, there were at the time only Ah, a ah, a couple million people in the whole country or had a tax i d. Number. Um, there were a lot of people whose last name was grouped our patellar saying, um and so is enormous data challenge to figure out how you to do this? The guys who were on the ground there were going and sending people into your house. Uh, who would write down, like, you know, how many servants from the kitchen? And, you know, there was, uh, there's America. And in front of the house or whatever, right? Like OK, great. Um, and they were the underwriter would just make a qualitative decision arising was never gonna be the way that we were gonna do business. Um and so I had to stitch together different data sources and do all this work. Um, and what I realized is that something like that can actually work will be transformative thinking work just as well. A lot better, perhaps than qualitative insight. So when I eventually left on that bank came home. I was dating a girl in Boston. So, um there this was back, You know, kind of the early days of the Internet when you plug in your computer into, um, um, to our listeners probably don't realize this, but there was a day before the Internet, and you have to get your computer into the wall. And it would make funny noises, uh, in a phone call outside. Yeah, and phone call cost three hours. Minutes. So dating a girl in Boston was until I came home, and I realized that there's a whole range of other claims of businesses out there, Um, both in the U. S. And around the world, which behaved with a similar level of lack of quantitative insight. Um and so I came to make the meet, the guys would started burning glass and no, they were actually named, uh, they were the name patent holders on a system that even today, um, processes things to third of the world's credit card transactions for fraud activity. Interesting problem. Why, that's relevant to what I came to do with my life. Um, they were they were frankly board that made they said well, g were always been in financial services and financial services companies. Banks? Are they dealing with money? They're very clinic ated where people absolutely not quantitative, Where they never not giving analytical insight. And so they landed in the world of HR and Civil HR is about people people decisions are almost never made based on data. What happens if we can make, um, HR decisions? Um, based on data based on actual patterns of how the market works and help people move their careers. And that was the inside started burning glass. Um, I, uh uh took over the business in a year or two later. Um, and, uh, that's been a, um you know, really transformative insight for me, which is that there's a huge opportunity out there. There's remain huge opportunity out universe, universe, huge opportunities out there. Excuse me. Tongue tied of tried a, um, to really bring data driven transformation in parts of the world that people never thought possible. The first challenge that bring last two gone. We were looking at the front door of a company who deserves to get a job. Um and ah, a lot of students are graduating university right now are are facing that dilemma and realized that all too often its software that's reading their resumes and so we were looking at instead of keywords trashy, understand real patterns of placement And what what actually works? What bears out in the market and, um, fascinating. Um, but we ultimately realized that there was an opportunity really, um, refocused we're doing. We came to be quite a format that that work continues to be successful. But realize that it wasn't the question of how do you match people in jobs, Or how do you sort through a stack of resumes that really mattered. If people don't have the data to know what kind of jobs are available and where those jobs will take them, they'll never be successful in likewise for companies. If companies are always trying to, um, by their talent to acquire their talent, um, right in the here and now, the delivery successful is the right person is either there or not. The world is full of all these cosmic coincidences. And for those for students were thinking about planning their careers. One things that always, um Ahmad by is some of the coincidences that bring. People were fabulous that their jobs and how did they get to that very specific job? And a lot of you trying to figure that out for yourselves right now. And a lot of that right now too often is random. It's random because people don't know where jobs are. They don't know what the landscape of opportunity looks like. They so we decided, Look, to do what we've been doing it to look it instructive at quantitative data around how people get jobs we had We realized that there is no data in the formal sense around employment transactions. There's things like resin raise and job descriptions. Um, and so we realized that we really need to do is to be able to trade structured meta data so that we could start to look for patterns. Now, once we had done that, we said we've got really good at being will take a job description of breaking apart and translated to our own language that we can, uh, we can see similarities between jobs and aggregate up information. What happens if you apply that to the whole universe of jobs? Right. You know things like, uh uh, you know the government statistics that we hear over last month and you turned me added three million jobs. Or these days, Unfortunately, it's not adding jobs. Um, those air based on surveys and the idea of a survey is you can't constantly see the whole world. Um, like you couldn't ask everyone in America survey. You know who they're gonna vote for for president. And so you take a small surveying, you try to extrapolate. Well, the problem is that that means you've got to keep your categories really broad. The extent of the Bureau of Labor Statistics tracks a job called a computer programmer. Well, for those of you may be studying computer science know that there is no job called the computer program is absurd. Are you a ruby developer? Are you? Ah, working C plus plus their you know, what are you doing? Um and and so we realized that, um, to really get to real insight about where opportunity is and what it takes to unlock that opportunity, you needed to, um, think differently, which is not about doing a survey. But you realize that most hiring transactions today happen on online. Which of us is actually still crunching out a resume and putting it on bond paper and putting into an envelope and putting a stamp on the envelope and sending it in right so we know can analyze, like 85% of all right. And we understand what are the specific skills that unlock those jobs, how to real people move up in their careers. And it's been transformative for us. It's been transformative for me in my career, um, to have a chance to deliver real insight. Um and, um, we think it's really transforming the world's of education and work.

um So our work is, um, you know, is very much of built around the world of big data. Um, it's, uh, divide our work into a couple of parts. Uh, so part of what we're doing is, um, is just aggregating the data, essentially scraping, um, as we know most, job posting on the web, Hundreds of millions people have their crew histories on the web in various places. And so it turns out this is not the part of our work. That's the highest technology. But it's actually, um, operationally very difficult. Um, we're used to thinking about things like, um, robots and spiders. Being essentially kind of robot is being infallible for those who are in the field. Know that, actually, these air very fragile technologies they break, um, and so going out and having an army of 50,000 robot today going out to all these different sites, um, and tracking when is the site broken? When is your robot broken, or is there a decline in jobs? Because, um, you know this thing going around with robot or because there's just less hiring activity going around takes a lot of operational expertise. So that's one aspect of we do next is Then how do you create, um, ineffective data store for massive multi dimensional analysis? If I want to be able to have data accessible that can slice it across dozens and dozens of dimensions across billions of records And, um, you know, there's, uh there's a lot of complex data storage and rendering capabilities, but another branch of, of, of the worst is around. How do you translate all of the different ways that people have describing themselves? Um, this is, um, for more technology technically minded of you, and the audience is what's often referred to his anthologies. Um, so you an example? Um, if you're a, um, have the title of associate, your job is probably different if you're working at an investment bank. Goldman Sachs, a consulting firm like McKinsey or a store like Walmart, Right? So being able Teoh, divide those up. Or conversely, knowing that the Associated Wal Mart and the Guest Services Representative and Target are doing the same work, um is also really important. So so that's a third branch of what we do, because that's what allows us to know that we start. We got the data we put in a place we can analyze it. Now we have a Rosetta stone so we can actually aggregated up and see what's going on in the market. And the last piece of this is the ability to apply that ontology. You've come up with a language, right? You've created um, you know, some people were tired were listening today. May have heard of it was, uh, used to be, have never called Esperanto. Oh, and the notion waas that let's get the whole world the key barrier. Teoh Teoh, World peace is that we can talk to each other. And so let's get the whole world to change. Language will stop speaking English and inner do and French and whatever else and will also be guess, Paronto And And of course, that didn't work very well. We're not trying, Teoh, Um get one to speak. Um, Esperandieu, But we trans everything toe Esperanto on the back end. To do that means not only defining the language, but it means actually having the translation capability. Um, so that when you see that job title which says associate at Goldman Sachs, you can say this is not a retail associate. This is a junior investment banking job

You know, I like Teoh. Think about my style of management as, um, as management by question. Um, you know, But it's also in some ways, um, management by story. And here's what I mean by that, Um, I think that, um, one of the most powerful capacities of the human brain is its narrative capacity. That's why our brain is today the very best pattern detection device that there is. We're able to look at very limited sets of evidence, more able to construct. We know exactly what's going on. And here's why. That's relevant in the world of management. Because ultimately, if you're a manager, you need to have a good sense for where he where are you going, where you where's your organization during where you taking things? One of the best ways to do that is actually to create a narrative. Now, let's be clear. A narrative doesn't mean fiction. We're kind of nonfiction narratives, just in the same way that the best the best Signs books the best history books often read like fiction in a way, because they're so engaging right, but because they're built around a narrative and and on and telling a really compelling and powerful and transformative story. If you've developed that, then what you always need to do is do a couple of things. Number one when you encounter, um um, you know something that you're learning or something that people in your organization of doing you can ask. First of all does this is this support the narrative that we've created first I was just take us where we're going. Just take us off course. Second, each thing that you encounter along the way that your team is finding helps you figure out whether or not that narrative continues to be nonfiction or whether you're veering into fiction because you never want to build an organization of driving organization based on fiction. Um, and so that gives you a framework for testing things, Um, for pushing hard on things for pushing people on what they're doing. Is this consistent with where we're going with our business purpose? Um, that means that Ah, and that's, uh that's what brings me to the idea of management by question, because the yes, we want to empower people. Um, it's really important to empower people to make sure that people have opportunities that people have opportunities to speak to contribute, to be creative. It's also important Teoh to push, uh, and asked the tough questions. We have to ask ourselves the tough questions because that's how you can make sure that, um, everyone in the organization is achieving what they can, and the organization is living up to its mission and potential.