James Cotton, Co-Founder @ iO Sphere

Welcome back to another episode of The Graduate Guide. Today we're joined by James, James Hopkins, the podcast mate.

Thank you for having me.

So, uh, you are the founder of Hyphen, or newly named Hyphen. Newly named Hyphen indeed, which is very exciting. But can you, uh, just give us a bit of insight into what it is you do and why you started it.

So right now what we started in is we run intensive bootcamp-style training programs that help people get access to the experience and skills that they need in order to land their first job in data as like a data analyst, data scientist, data engineer, something like that. We started it because, so my background is working in industry in terms of building and scaling out of analytics pricing and data science teams. And what's really interesting in this space is that there's lots of young people who want, or lots of people in general who want access to these jobs, but most education in this space is just fundamentally not fit for purpose. You have this theoretical teacher-classroom environment where you only learn very antiseptic problems that have a right answer, and they teach you in a lot of ways just the wrong technical skills. And so we were like, well, I have a better idea. Why don't we just train people to do the job itself? We don't have this theoretical teacher-classroom environment. What we have is a synthetic company. So they come to work for us. Fast fashion e-commerce, very similar to ASOS or Boohoo or something like that. And so that means you get not only the skills that you actually require, but we help them build up real experience and then we show them how to translate that experience into a job application process so that they actually land a job off the back of it.

Sounds simple, or what should be happening anyway, but why do you think it's not happening in the education system?

I think that the incentives in the education system are usually in the wrong place. What everybody has kind of decided is, well, there's a few problems. One is that academics don't, the bridge between academia and the real world, it's way too far apart. And so you have people who are focused on the theoretical side of things and then what happens in the real world is just fundamentally different. And so you need industry to actually be the ones that define where it is that you're supposed to be coming from. What are the skills that people need? What do we actually need people to know? The other thing is that the incentives in education are usually wrong, if that makes sense. What they're trying to do is find students and deliver content at them and hope that they pass an exam. They get paid upfront, they don't take any financial risk themselves, and they just, you know, our job is to get students on courses, get those students through the courses and make sure that they pass. They don't really care too much about what happens afterwards. And so that's why with us, we've specifically aligned our financial model with them actually getting a job off the back of it. So you don't pay with one of our programs until you land a job. That means that we're very much more focused on actually giving people the right skills and experience so that they land a job. We've got very aligned incentives, and I don't think most education has that at all. They're all just about delivering content at the lowest marginal cost, which is different.

And let's talk about your experience in the education system. You go back to when you were living in Canada. What did you think your career might pan out to be when you were back at university?

I really had no idea. Like, people say they have no idea, but I really had no idea. I didn't really even think too much about getting a job because, like, I went to— I was always good at school. I'm very good at standardized testing, which is not really like a marketable skill, but it does get you somewhere in life.

Yeah.

And I was good at science and math. And so I went and I studied medical sciences. We called it life sciences then, or pre-med or something like that. And that's just because that's what I liked. And I assumed that if I went to a good university and did a a good degree like that, that was— I would be fine. Um, and then when I was in university, I was like, well, obviously I need to become a doctor, because if you're studying pre-med or medical sciences, that's basically the career path that you follow. Um, and it was only when I was in my fourth year of university that, uh, uh, I was sat down by a doctor and said, do you know what this job is really like? Do you know what you have to actually do? I was like, no, I actually have no idea. And they're like, okay, well Let me tell you, that ended that career right then and there. Um, and so then I basically had no idea what I was going to do when I graduated, and I graduated in the financial crisis, 2008.

So that, that feeling of being sort of confused and lost, how did you actually get out of it then? Did it just— was it just get a job, or was it work out what I'm good at and then try and find a job that suits me?

Well, obviously I had to find myself first, so I went backpacking around South America. Of course, need to be found. And then I went back to Calgary, which is where I'm from in Canada, and I was back at my parents' house and I was like, okay, I need to find a job. And I was like, actually, I have literally no idea how to find a job. Like, literally none. I had my CV from when I was a ski instructor and I don't even think we really had LinkedIn much. I don't even know where I found jobs back then. Job boards. Yeah, like weird job boards and stuff. Like in the newspaper, maybe. I don't even know where they came from. But no, I was— because I graduated, there was no jobs. Like, that was the general sentiment. Calgary, where I'm from, is very boom and bust. It's very driven by oil prices. And when there was the financial crisis, there was like no jobs around. Um, and I just kind of sat there like, what am I gonna do now? And so I just started applying to places, lots of places. I got a job in a bar for a bit, like in a kitchen. Yeah. Um, and eventually I got a job in customer service at the railroad.

And that experience at the railroad, working there, what— that maybe wasn't what you had in mind what your career would be, but how did that actually manifest itself into a career in data?

Yeah, that was a bit of a switch, wasn't it? No, like, I was actually so happy to get that job in customer service because, so I graduated in May, I got back to Canada maybe in September. I think I started in April, May of the next year. So it took like quite a long time. And that job was, I loved it actually. It was, I mean, I didn't, I'm not a morning person. It started at 6 AM, but it was, It was a step into like a real company in Canada. It's like a freight railroad. It's like 20,000 kilometers of like freight trains. There's no passengers or anything like that. And there was a whole group of us that got hired at the same time. Okay, maybe 4 or 5 or something like that. But it was, you know, you meet people, you learn, you start to understand how businesses work. I always had a natural inclination towards a bit of data. And so I was looking at sort of trends and where the rail cars were going and then eventually there was a reorg and I got out of there, basically is what happened.

Which led you into?

Service delivery, which was sort of like a different type of customer service. It was actually like, I never went into data until after I moved to the UK, actually. No, I actually had a job. If you've ever seen the movie, which is like where George Clooney is flying around kind of firing people. I had a job where me and this one other guy, we traveled around the railroad network and we were basically just like trying to standardize customer operations and ripping up all these like sweetheart deals that people had. And I did that for like a year and a half of just like traveling around, giving some customers bad news. I mean, there's fun parts to it too, but that was really quite formative.

Like, probably. And I'm not sure if sort of data was as desired a skill as it definitely is now back then. But obviously you worked as a director at the likes of Expedia, these really renowned brands. When did you sort of see that shift that data was going to become an incredibly important career to be in?

Do you want to know the truth?

Yes.

It just kind of happened. I didn't really notice because what happened is I left the railroad in order to go to— I did an MBA. I normally don't tell people that, but you can see it on my profile. I did do one. I'm a corporate sellout. I have not paid off the student debt. Don't worry. 10 years later. And after that, I moved to the UK because I wanted to keep living in Europe. I could get a visa and I had some friends here. It wasn't a lot of thinking about it. I'd only been to London once for 3 or 4 days before that, before I moved here. And then I needed another job. It was another moment of, uh-oh, I need a job. I'm a different type of graduate now. And I had worked in pricing a little bit at the railroad. And so I was able to land a job here off the back of that experience. And then when I got into insurance pricing, which is very different than railroad pricing, that's when I realized there's a whole different skill set involved. I kind of went on Coursera and learned a bunch of it before, and then I really wanted to work at Hotels.com because I love travel. And so I then spent the next 8 months trying to go and work at Hotels.com. I did like 27 interviews for 9 different jobs before I eventually got one in data, and that's when I found out what data was. It turns out I loved it.

I think this is the reality of a lot of people's careers, right, is that, um, it— you can't script a career path. I think, you know, you can try and get really emotionally attached to an idea of how your career can go to look like, but I think you've obviously gone on to have great roles at great companies and found a very successful startup, but it's never it's never a simple journey, and a lot of it is just, you know, you think you could be good at something, you think something feels like the right decision, and then you just go ahead and do it, right?

Yeah, I think a lot, a lot is about opportunities that exist that present themselves. I think the biggest breaks in my career have always been where there was a job that nobody else wanted to do but was a step up from where I was doing it. The big— like, at the railroad, I moved into tariff and risk pricing. That was a, a big one. Um, insurance pricing actually was a big one, even though it wasn't the industry for me at all. And I went into data, like proper data in tech at Hotels.com at a time where it was still really early and it was a very different industry than it is now. Back then, if you're like, I know, I like data, they basically were like, okay, well, we'll consider you. Now you need skills and experience. It's a very different world than it was 10 years ago. But I'm always jealous of people who've had that, like very— not jealous, that's the wrong word completely, but there are people who've had very, you defined career paths. I studied this thing, I then went into a grad job for it, and I progressed through it. Knowing early on what you want to do would have been pretty amazing, you think?

Do you think that?

Well, I don't know. I would have like missed out on a lot of pretty good experiences.

But yeah, because I think, you know, obviously you'll be— you'll have a lot of people that are coming to you to try and help you land that job in data. But these kind of— these, these people that are coming to you, are they, um Are they 100% sure that they want that career in data, but they're just struggling to get it? Or do they just think it might be the right next step for them?

Oh, that's a great question. We've had 3 people so far, so like barely any drop out because they decided that a career in data wasn't for them. But the thing about data is that data is super broad and the skills exist in almost any job actually that you can end up doing, right? Because it's a lot about problem solving, trying to understand think critically about problems. There are some technical skills associated with what we do in data, but you can use those skills in a lot of different roles. And so, most people come to us though knowing that this is the kind of role they want to get into because data, like, these are interesting, they're creative, you can work on all sorts of different problems, you can work in different industries and different geographies, you can be really flexible in terms of how you work. And so, they're good jobs. And so, most people come to us knowing that they want them, if that makes sense.

And on the employer side, because I think a lot of companies nowadays kind of realize they build a product and they're like, oh, hang on a second, we actually need to, to, to scale, to become like a big company. Nowadays you do need a really great understanding of data. It's a lot of what people sell now. So I think the actual, uh, process of employers coming to you Where do you think they're maybe going wrong in their understanding of the actual significance of understanding data in business now?

Do you mean the employers or the—

The employers, yeah.

I think that there's a— Oh, there's so many problems here. We're actually, one of our big expansion areas for next year is we're going much more into the corporate training side of things. But what most people fundamentally don't understand what data is, because it is an actual thing, if that makes sense. So they just think about it maybe in terms of like, let's make better decisions or we need to invest in this. You've got data, you've got the ability to understand what you've done in the past, operate differently, make changes, but you also have the ability to predict things into the future, which would be more your machine learning forecasting type of things. You've obviously got AI, but the problem is that people don't know any of what this means. And so they know that they should, they think they should be doing something, but they literally have no idea what they should be doing. And so there's a lot of opportunity to get people to do things in the right way, better drive value from data, actually use it in a way to like help grow their business as opposed to just kind of panicking. And what we see actually a lot is like at the exec level, people who came up in their careers before, you know, data in the current format was a thing, they don't even know who to like ask questions to or how to start to think about it, if that makes sense. And so they're almost a bit scared of it.

Is there an example of a company that you've you've worked with or worked for, or even you know that previously didn't have data and then they've since, like, investing in the people and the team to actually build that out for them, have seen like a mesmeric rise in their business?

Um, I'm having one of those moments. I mean, there's got to be many companies like this. Um, it's weird, it's hard for me, like, the companies that I've worked for are, they were always sort of digital native, if that makes sense, except for, I mean, the railroad was a bit different, but it was also, I left a long time ago. And so, if I look at Expedia or WorldDreamer or something like, they were always tech companies that were built sort of data first, if that makes sense. So, those are not springing to mind. I'll think of an example.

Sure. But I think, I guess, generally the overarching point here is that, and I'm finding it now even with the recruitment industry, I think it's— you've seen so many problems where it's just, it's very commission-driven. No one really cares too much about what happens to someone in that journey. They're not really looking into, on an analytical way, why someone might be good in a role and And I think a lot of the mismatch is actually caused by just the lack of understanding of the data of people, of humans. So how do you see data not just in the recruitment industry but being applied for general good in the world as well?

Well, I think there's tons of ways that data, like, even if you can talk like the LLMs and chatbots and stuff like that, there's obviously people are worried about how that will affect employment, but there's huge opportunities just around, you know, if you think about the medical applications, you know, being able to diagnose people a lot faster. If you think about it from an education standpoint, when you can actually start to give people at scale personalized tutoring, which is not, you know, it's not economic to do that in any other way. So there's humongous areas you can use these things for good, but you can do it like there's also much more lower-level use cases, where people are spending advertising dollars really inefficiently, which means that they're unable to invest in other areas that are— if you think about how you optimize a sales funnel, if you can be much more effective at taking a good product to market, a product that is adding value to people, if you can be much more effective about that, more people can get access to that product and you can scale it, you can deliver lower prices, all sorts of good things. Like that.

Yeah, I think if we're using an example that people might actually know and see it publicly working is that in football, you know, English football, that is.

I don't know anything about sports. I think half the people that come, half the boys that come on our course, males, they always like, I want to get into data because I love soccer stats or football stats. And I'm like, I don't know anything about sports, I don't do sports.

But I think whether you like sports or not, if you look at teams in the Premier League like Brighton or Brentford, and I think everyone knows the example of Moneyball, the film, right? I think that was like, that's, that's why I think where everybody from a young age was like realizing, oh, okay, like data is important, right? It's, uh, and it can make or break performance as well. Like, how have you seen, um, you know, performance actually being tracked by data, like, and people improving themselves? Through that?

Well, there's— I actually think this is one of the big gaps that still exists out there, right? Which is like, it's very difficult for people to get objective views of their performance around skills or experience or whatever it is, even within the company that is required, right? Because if you like— and so I don't think anybody's doing this particularly well, right? It's difficult for people to be objective. It's difficult for people to actually understand what are the requirements for anything. Testing as like a standardized testing rule is not a particularly great way of assessing where you sit on a spectrum, if that makes sense. So, I actually don't think, I mean, I don't know anybody that's doing it well from a performance perspective in companies. Do you think there is someone?

No, I guess I'm asking you.

I actually think it's a huge opportunity because like, for example, if you take a technical skill that exists in data, there's a programming language called SQL. It's the one that you use SQL, it's actually relatively simple in terms of programming languages, but it's the one that sits between all use cases and databases, right? So like if you want data, you need to do something in SQL most of the time. The level of SQL that's actually— so one, the level that's required is usually not where it is, but also people don't really understand when they're coming into data that SQL is the language that they need. And so there's no good way of actually people benchmarking their own performance against where they need to be. And then improving it so that they understand how they can get to the next level. And that's a very simple, straightforward use case in data. If you got into something like storytelling or collaboration or how you work well with others, how you lead a team, there's no really good ways of assessing that objectively across industries, functions, teams.

Yeah. And I mean, I'm interested as well, how students could apply data to their actual job search, right?

Well, so this is something I actually like. I also— every student that comes through one of our courses, I always teach a session called like, everything good in life is a funnel. Life is a funnel because there's so many opportunities to actually like just be much more numeric and structured and analytical and think critically about your job search and actually get to the results a lot faster. And you don't need like complicated systems or anything like that. You just need— You need to track your own performance because if you do, you can get to— that's one of the ways that we actually make sure that all of our students get to a successful outcome, right? Which is like, are you aware of enough jobs? Are you applying to enough jobs? Are you being selected? No. And if you understand where you're failing in your funnel all the way through to getting a job, starting that job and passing your probation, right? If you can understand where you sit relative to your peers, relative to good benchmarks, what it is where it is you should be, you can change your own behavior to adjust and actually land a job. So for example, are you applying to it? Like, is the problem that you're like, you're not aware of enough jobs that you could apply to? If so, you need to get better at researching, at finding jobs, at understanding titles you could apply to. Is the problem that you're not applying to enough jobs? In which case, how can you speed up your application speed? How can you do those kinds of things? Is your problem that you're not being selected for jobs? In which case, is it a problem with your CV? Is it a problem with your LinkedIn? Is it a problem with your networking approach? Right? And like, we can actually create a funnel from start to finish. Everything good in life is a funnel, right? And show you where it is that you're falling down so you can, can land a job.

Yeah, it's, it's fascinating but also slightly scary, right? I think because you do hear this, um, this type or this phrase thrown around quite a lot that if you don't adopt data and AI and everything, like, you will fall behind. You will you will miss out. Like, I think that is just how careers and life is being done now. But I think if you will actually try to break down the significance of why, like, an 18-year-old, beginning of their career, beginning of their life, really should start getting into data, what would you say is like some encouragement to start just, you know, getting comfortable with it?

It's actually, I agree, like these are the skills for that will set somebody up for the future. And we really talk a lot within our own world about like durable skills versus, you know, more transient skills. The technical skills, which are how you do things, are transient. They'll come and go. You can learn new ones. It doesn't really matter. But what we try to focus on a lot within our programs, but actually are the basis for like most careers, are much more about the durable side and what it— where you can take these skills and go into any career, right? And those are about critical thinking, analytical problem solving, being able to recognize a problem, take that problem structure into something you can solve, validate that with experience. That's what we really do in data, which is very different than like it being a bunch of stats, if that makes sense. And so what people tend to get worried about, scared about, right, is that if I go into data, I need to know statistics. I need to know math at like a, you know, third year at Durham kind of level. And that's just not true at all. In fact, like, one of the biggest misconceptions in the world about data is that you need to have really good math skills. It is totally possible to go all the way through into like high levels of data with only having like GCSE stats, maybe first year of university equivalent or something like that. We mostly deal in like trends, averages, medians, distributions. Of course, there are jobs that involve really advanced math, but they're the minority in data.

What's the real barrier then? I'm trying to work out why, if you're saying it is as simple in terms of just a mathematical understanding and people might build it up in their head as really being really complicated, where do people stop? Why is it?

I think a lot of people just filter themselves out at that point, which is like, actually, I am not— I was never good at math, and so therefore I can't do these jobs. But what's really like sad, just misfortune, I don't know, the way math is taught in schools, right, is like, here's a calculation, go and calculate it. That's not a particularly good way of getting anybody to buy into it or enjoy it. What's more interesting is like, actually, I've got a problem, which is like, I want to be able to detect credit card fraud, or I want to be able to understand who's going to win the Premier League, or I want to know who's the best player to add to this Team. It's a team, right? Yeah, yeah. Right. I want to know who's the best player to add to this team. Right. And then it's like, okay, but actually there's a problem. I can understand that there's a problem. Now let me take that, figure out how to structure it, then I'll go and solve it. You can get a lot more buy-in that way. Where I think— so problem number one is people just don't even attempt because they think they're not good at math. That's really like, we can overcome that. I think the problem that exists after that is that people think once they've decided to go for it, then they think that it's only about math or it's only about technical skills. And so they spend all their time like learning to code advanced Python, or they might learn how to build like a neural network, which is a really complicated form of deep learning or machine learning. You know, it's what Google Search was always built on. But actually what they're missing is like teamwork skills, business acumen. How do I understand how the business works? How do I understand how to set goals, measure things? They're missing analytical skills, critical thinking skills, they're missing professional skills, how to make slides and communicate, run a meeting and that kind of stuff.

Have you got a case study within what you do at Hyphen where somebody's come in with just no background that would indicate they would be employable for a data role and have come out the back of it employable?

That is what we do. We take people from any background. They have to be motivated. Like, our courses are intense and we take people who have— we've taken recruitment consultants, you know, people who didn't go to university, were working in an agency, recruitment agency, and train them up. And then they've landed a job in data and we can do that in 10 weeks. So that's like, it's not hard once you have the right motivation, you've got the focus. I mean, we've got hundreds of case studies like that because that's that is what we do. We obviously have people who come from science backgrounds and stuff like that.

But, um, so I mean, you know, forget— correct me if I'm wrong, but there are loads of students that are like just applying for jobs, um, that they think they might be good at. They're getting rejected and they just feel incredibly lost. If they went through your program, that is one sure-proof way of making them like employable in a, in a really needed career path, right? Like, that is a fair statement to say.

Yeah, that's what we do. Yeah, it's like, so it's not, I mean, it's not a guarantee, right? Like, um, in general, we get to about 75-80% get a job within, in data within 6 months of graduating. And that goes up to about 75-80% over 12 months, something like that. So there's always people who, um, some people don't go into data off the back of it. Some people, you know, choose not to, um, look for a job, but it is, it is very, very possible to go and get a job off the back of what we do. And a lot of that is when we give you— it's what we, you know, it's our thing, right? Which is like, we give you the experience. It's not just about like, here are the skills you need, but like, actually, let me put you in a realistic working scenario. Let me give you real problems. Then let's take a step back and show you how we can translate that into your CV, into your interview questions. We do interview prep with them the whole way, practice interviews. And then we make sure that they go get a job by holding them accountable to the job search afterwards.

So one thing that I'm trying to wrap my head around, and I'm sorry if I'm, um, like, this is quite like, uh, revolutionary in terms of like a way to actually solve this problem where there, there is— this is a sector that I imagine does have quite a lot of jobs available, but there's just not the talent needed, right? But then if someone's feeling hesitant about, okay, when I get into that job, what does the future of my career then look like? What are the branches that you can go on to do from a career that starts in data?

Oh, excellent question. I mean, I think you can go into pretty much any role because right now, see what people think of with data, right, is they're thinking like you're doing some kind of mathematical calculation to work on some kind of problem. But really what you're trying to do, it's almost like a strategy function for the business, right? You're trying to go in and say, actually, I can take a complex problem that we don't really understand. Amorphous, ambiguous, right? And say, actually, I can understand that problem. I can then structure it down into problems that we can understand, we can solve. I can then understand how I should solve them, I'll solve them. I can validate the whole thing and bring it back together from my experience and then go back to the business and say, I recommend that you do this. And if we do this, it's going to have this impact in terms of, you know, revenue growth or cost management or something like that. And that is just business, right? That is just how we're doing it, but we're doing it from a data perspective. You can, so you can take that skill and you can go into any function.

And the opportunity cost of learning how to do this, whether you think you're like a very people, human-centric person, is there is literally no opportunity cost to just seeing if you are someone that's interested in data. It will be beneficial no matter what you go on to do, right?

Yeah, exactly. You can, I mean, I actually think the best, I mean, I focus a lot on analysts, so data analysts, because that's my background. But the best ones are people people, like the ones that want to get out of their chairs, they want to go meet people, they understand how the business works. Because it's one thing to do a project, it's another thing to get that project adopted. Right. And so that's the hard part, which is not— the hard part is not building a model or running some kind of algorithm or something like that. The hard part is like understanding the business, structuring it, and then actually taking the output to deliver some kind of benefit for the business. People, people, people, people. It's really weird saying that, right? They're the ones who do that best because they have the ability to actually build relationships with the stakeholders, the other team members, everything like that.

And without trying to render what you do redundant, what is your message back to academia and the school in terms of how they can integrate what you're doing at an earlier stage?

I think they just have to do better. I think The— well, there's kind of two different views of this, right? One is that like an academic approach is like actually people can learn for the sake of learning. I think that's kind of an elitist approach, if that makes sense. People generally take education because they want to get to an outcome, right? And that outcome is usually either getting a better job or a new job or being able to perform in their job in a way that they can't right now. And that's what they need to keep in mind, which is like, people are coming to you because they want something in the end. That means you need to be much more responsive to what they want. If they want a job in data, you need to be training them for that. They need to be much more employer-led. They need to support people all the way through. That's what I think.

It's been a really fascinating conversation. I think we could go on for ages, but I'm going to ask you my final question now, which I ask every single one of my guests, is that When you first started off in your career and you didn't know at all what you wanted to do, you might have had a certain idea of what success might go on to look like in that career. But fast forward and now you've found your purpose and you're being able to give that back to other people. What is your definition of success in your career and business, and how has it changed from the beginning?

In the very beginning, I was focused. I mean, I was working in a 15,000-person company. It's very hierarchical. You're like, actually, how can I 'get the next title, how can I get the next, you know, promotion or performance review or something like that?' That's out the window now. For me, it's— and I've really— and that actually probably happened for a long time in my career. Now it's the ability to actually have impact and change something. Like, fundamentally, this entire system around education into, like, jobs is fundamentally broken. This is like a weird system that everybody has subscribed to. We're going to pay 9% extra tax for the rest of our life in order to go through this system for the chance of getting a job. And now I get so much more meaning out of like, actually, we can help people do it better. We can do it cheaper or more efficiently, and we can actually help people get what they want. And when you see somebody come from a background that has nothing to do with data, that would have never thought they could access one of these careers, they go on and get that job. Like, that's incredible for us.

James, thanks so much for coming on. Appreciate it.

Thank you for having me.