Democratizing Data Starts at the Top

Read Interview

Speaker 1:
Welcome to Champions of Data and AI, brought to you by Databricks. Thousands of data leaders rely on Databricks to simplify data and AI, so data teams can innovate faster and solve the world’s toughest problems. Visit databricks.com to learn how data leaders are unlocking the true potential of all their data.

Speaker 1:
In each episode, we salute champions of data and AI, the change agents who are shaking up the status quo. These mavericks are rethinking how data and AI can enhance the human experience. We’ll dive into their challenges and celebrate their successes, all while getting to know these leaders a little more personally.

Chris D’Agostino:
Welcome to the Champions of Data and AI. I’m your host, Chris D’Agostino, global principle technologist at Databricks. Data democratization starts with disrupted leadership. When data leaders simplify the data architecture and empower data teams with the right tools and creative freedom to do their best work, they are able to create game-changing insights for their organizations. To dive into this topic, we are joined today by Stuart Hughes, Chief Information and Digital Officer at Rolls-Royce. He is a seasoned CIO with experience in many of the UK hot start ups and dot-coms. Stuart’s goal is to create a culture and environment for every member of his team to thrive. Let’s get started.

Chris D’Agostino:
Welcome, and I guess to kick things off, I’ll ask about the what looks to be a Nerf gun behind you.

Stuart Hughes:
Yeah, so it’s great to meet you, Chris and great to be part of this. I spent the first 20 years of my life as a CTO in a lot of the UK’s really fast-growing dot-coms and kind of mid-way through my career, I pivoted to work for quite a traditional manufacturer and as one of my many leaving presents, one of them’s a poem on the other side, very odd. They gave me a Nerf gun that’s got disrupter written on it, and I think it was that time where in the dot-com world, being a disrupter was a really good thing, whereas in a traditional manufacturer, I had to kind of spin it to be not disruptive internally by disrupter in manufacturing, so it was a really kind gift that someone gave me and something I’ll treasure forever.

Chris D’Agostino:
Do you actually use it?

Stuart Hughes:
No, never. My son’s actually got one of his own because he kept stealing it, so-

Chris D’Agostino:
Nice, nice. So Stuart, you’re from Manchester, England. What was it like growing up there?

Stuart Hughes:
So Manchester is the cool part of the UK, right? So I grew up in the era where Manchester United won every single trophy, every single game. I had a season ticket at the ground. Manchester was famous for music, so Manchester Oasis, so when I grew up, Manchester ruled the world, right? It wasn’t just a city, it was kind of an attitude and I guess in some ways I kind of personified that with how I am in my life, a little bit cocky, a little bit confident and always ready to take on that next challenge, never backing down, so yeah. Coming from Manchester’s kind of a really cool thing, something I really cherish.

Chris D’Agostino:
Awesome, awesome. So in terms of the time when you realized you wanted to go into computer information systems, study that, make this your career track, what was the motivation there?

Stuart Hughes:
So it’s really funny. When I was seven, I was the computer monitor at school and I don’t just mean the computer monitor, but if the computers broke at school, it was, “Go and get Stuart, he’ll fix it.” Nobody understood I just unplugged it and plugged it back in, so it kind of has always just been the thing that I did. And all I do now at Rolls Royce is turn off and on again, much bigger systems. And I think it’s gone from there, so yeah, I have that kind of natural interest. It wasn’t in computer games.

Stuart Hughes:
You know I had a ZX spectrum, I used to code at that, it was more accidental coding. I’d do things, tampering and it’d kind of go crazy. And it just felt like a natural extension in my life, but I feel like I was like a kid who was good at football. It was really easy to just carry on doing what I was good at. I think even now, I really enjoy what I do. I really enjoy being a technologist. The senior leader, senior executive, it’s been a part of the role that I’ve learned, but if you’re a T-shaped person, my T bit is that technical, used to be a developer, never wanted to let it go.

Chris D’Agostino:
Yeah, yeah. It’s funny, when I was in high school, we were doing all of our student government elections using paper ballots and I was in the computer science club and computers were really just coming out in terms of personal computers being widely available, so I wrote the election software and so true story, I literally wrote the election software and with every vote, it would persist the results, the current state of the results to a 5 1/4 inch floppy drive and I happened to be running for president and I won and I promise that the voting… I really did. I promise, so this is 1987 South Lakes High School, or ’86 I guess it would’ve been at the time.

Chris D’Agostino:
So anyways, it was fun and so hearing that you were the go-to person as well, we’re sort of brothers in arms there. Let’s see. So you’ve worked for a lot of different companies in your career, so you’ve had a lot of different experiences. What do you think is the most memorable achievement so far with the work that you’ve done?

Stuart Hughes:
Yeah, the obvious ones are like when we IPO’d Money supermarket for a billion pounds or those corporate transactions, but I guess to think about it in my heart of hearts, it’s a different answer. I worked at a company called The Hut Group (THG). I joined really early on. We were probably turning over about £100 million and to give you a contrast, about six years on, they just IPO’d for about six billion, so pretty successful now, but I joined really early on and the way we had a website that sold CDs, DVDs, it was very specific, but we kind of knew as a leadership team that people weren’t going to be buying CDs and DVDs on a website for much longer. So we had to pivot.

Stuart Hughes:
And I’m not joking when I say it was probably in the last month of me working for the company and in two weeks we acquired a business and migrated them, and we actually migrated their website onto our tech stack before we’d actually signed the contract. And all we did, as soon as we signed the contract, it was ours, we just flipped the DNS and our website spun up. So you know, those are the kind of things that really get me excited where we did something in technology, we built a capability and it just dramatically changed the ambitions of that company.

Chris D’Agostino:
And so your LinkedIn profile has an interesting quote from you. It says, “I’m an ex-developer frustrated by management, not creating the conditions for innovation and success, so I decided to try for myself.” Can you tell you a little bit more about that? What was the experience and the environment that led to the frustration and then sort of to follow on, what do you think you’ve been able to do in your career to invert that and make the developer’s lives more enjoyable?

Stuart Hughes:
Yeah, and I think that in a lot of the companies that I’ve led and maybe the feedback I get from some of my team is, “You’re too developer focused. So you’re still thinking about the experience from the developer book.” You know, I worked for some businesses where we were a dot-net development team, we bespoke our web application, it was a good thing and then we joined the part of a really large group and their answer was, “No, no, you need to migrate that stack from where you were, which is bespoke built for what you do, to our generic ecommerce engine built on Java, and go and convince your development team that that’s a great idea.” And it’s like, “There is no way you can convince them of that.”

Stuart Hughes:
And I think that it’s those little things that yes, it might make great sense at a corporate level, but on the ground, it’s absolutely not achievable. It’s too high risk and actually, why would you put the business through that risk? And those are the frustrations that I had. And I always felt that if I could be part of that decision and maybe it steer it in, “What is it you’re actually trying to achieve?” You know?

Chris D’Agostino:
Yeah.

Stuart Hughes:
Because if it’s reduce costs, our platform’s like a tenth of the cost of theirs anyway. I go back to The Hut Group (THG) where we ran a website that was shipping about 6 million orders a year and it ran on one rack of hardware. And people were coming in saying, “Oh, we need to put Oracle in, we need…” Why do we need to put any of those things in? And it was just a complete misunderstanding of bespoke isn’t always bad, right? Bespoke is we’re really good at the things that we’re really good at. We’re very efficient and we’re very lean. What is it you want us to be good at? And I think that that’s the change now, is I can ask that question right at the very beginning before the big technology change starts to happen.

Chris D’Agostino:
Well, it’s that the simplicity of architecture sometimes is really valuable, right, and not over-complicating the design. Now, sort of transitioning into today’s modern data ecosystems where there’s more and more data that companies are collecting, they’re trying to figure out what data has value in it. Can you talk a bit about data democratization and some of the things that you’re doing there at Rolls-Royce to collect more data, make it more accessible to your business professionals, your data science community, in order to get some insights out of it?

Stuart Hughes:
Yeah, absolutely. A bit like you, Chris, so I think you must have done lots of BI projects as well, so back in the day I did lots of BI projects and I genuinely used to have long hair and I used to use hair straighteners and anyone who’s ever been SSBI or [inaudible 00:10:36] really large data sets, it’s really cost us a lot of our hair, those kind of projects. And I think that’s what really nice about today is for me, in the past enterprise BI integration tooling, the marketers created this great vision of what it was going to do and how it was going to allow everybody to do analytics themselves, but the reality wasn’t the same. The reality was the integration tools took a long time to get data loaded in. The reality was that however big your database was, it was never big enough, you know?

Chris D’Agostino:
Yeah.

Stuart Hughes:
And the reality of the analytics tools is they were too complicated for someone who wanted something simple, but too simple for someone who wanted something complicated. And I really think that, that’s what’s changed now with modern technology and I think as Azure, PowerBI, Databricks, a lot of these kinds of technology changes have really allowed us to think about how we can democratize data better. So one of the things that I think people found counterintuitive when I arrived at Rolls-Royce was there was a big push to, “Let’s create a giant data lake, let’s put everything in the data lake, categorize all the data in the data lake and then in three year’s time when we finish building the data lake, we can get some value from it.”

Stuart Hughes:
And I guess from my background in the dot-com minimum viable products, those kind of things, we kind of spun it the way and they said, “Well, can’t you use the tool that comes with the system? So if it’s SAP, can you not just use Lumira?” “Well, yeah, I could use Lumira.” “Okay, well, why don’t you use Lumira then?” “You don’t need to extract all the data, put it over here, lose all the security model, try and re-implement…” You know, all of those things that come through. So for me, I always push use the native tool where you can.

Stuart Hughes:
But I think that the real thing that is different is where the native tool can’t do it because you want to merge in other data sets, then what I always do is say, “Rather than kicking off a big initiative to create a data pipeline, make everything productionized, why don’t you just do something minimum viable product? Just prove value. Just create something PowerBI, extract some data, use Databricks, whatever it is, but create an ad hoc one off way of doing it and prove that, that works first and prove that you get the insights that you want and that there is value.”

Stuart Hughes:
And I think it’s probably 50/50 right? I think sometimes people do that work and they realize there wasn’t a correlation, right? Or the answer isn’t what they wanted it to be, so they’ll chuck it away, right? And then I guess once you’ve proven the value, if there’s something there, then productionize it. And I think that was a real change. I think when I joined Rolls-Royce our start point was, “I want to create this thing in production that’ll last forever.” And of course then, the most expensive ways of doing it. And now we’ve switched to, “Do it in the native tool, if you can’t do it in the native tool, Databricks, PowerBI, non-productionize, minimum viable product, get the value and then if there’s really value ongoing, then let’s look at a data supply chain, let’s look at making something that’s more repeatable because, by the way, the run cost of that is much, much higher.”

Stuart Hughes:
And I think that’s really changed the way that people think about it and it’s almost made PowerBI not a dirty word word, because I think in some companies it’s deemed as shadow IT or all of those kinds of things, right?

Chris D’Agostino:
Yeah.

Stuart Hughes:
It’s prove the value before we productionize it.

Chris D’Agostino:
And given that you’re in a regulated environment, you’re collecting data from aircraft engines and those aircraft engines fly on commercial jets and people’s lives are at risk, and so they need to perform well, can you talk a little bit about as you think about innovation and collecting data and gaining new insights, what do you need to for the environment that you’re deploying these things into to ensure that safety is first, as I’m sure it’s a top priority there?

Stuart Hughes:
And that’s it. One of the phrases that summarizes Rolls-Royce is, “Trusted to deliver excellence,” and I think that we absolutely take that to heart in everything we do. Some of the applications that my team write, they are safety critical applications. So they go through an external governance from the aviation authorities to say, “Yes, you can adopt that model to predict the life of an engine, to predict the life of a part.” So the governance that we go through at Rolls-Royce is probably much higher than I could’ve dreamt of when I was in one of those small agile start ups, and I think that that’s really important.

Stuart Hughes:
The second thing we’ve done is to create a data passport process, so we understand the lineage of all of the data, where did it come from, what can it be used for and then where are we going to put it? So any data that comes into Rolls-Royce from an engine, from an customer, or some data that we generate from those data sets, goes through a very, very consistent process where it’s reviewed to make sure that yes, we can take that data and we can put it there because it’s not military, or it is military or it’s export control compliant. And then once we’ve done that, we have to understand what have our customers given us permission to do with that data and again, make sure that we honor those commitments that we make.

Stuart Hughes:
So our passporting process is a really, really important part of us knowing what data is where and where can it go. By by having such rigor in that passport, once it’s in the right place, actually we have a little bit more flexibility to work and we have more agility in what we do. So at Rolls-Royce, our data supply chain’s been reliable, understood and effective. That gives us the ability to be agile. We don’t have agility in the sense that maybe start ups have in that way.

Stuart Hughes:
I think the final piece, really, is I’m a person who fights against the concept of shadow IT. So do I want someone to go and create a crazy access database that does was SAP does? Well, no. Okay. But do I want to give a user that’s probably a little bit more of an expert than average and give them access to Databricks and give them access to some of our data stores? Absolutely, right? So that person has now found a new ceiling, they can learn more, and they might become one of our data scientists of the future or do some advanced analytics. So for me, it’s really encouraging the business users to be able to do more for themselves. I’ve a background coming from marketplaces and one of the things you learn in marketplaces, you can’t write the features fast enough. You have to really on third parties, the community to deploy features and move the platform along so you can keep up.

Stuart Hughes:
So being able to enable the people in Rolls-Royce to do things for themselves is a really, really important thing and I remember one of the things you and I talked about, Chris, where you were talking about having to train some of your data scientists who used to be maybe data engineers or having to bring those people through and I promise you we’ve got the opposite problem in Rolls-Royce. Everybody’s got a nuclear physics degree, astrophysics degree. Our engineers want to write code, want to write really advanced algorithms, advanced models. They’re really accustomed to fluid dynamics and aerodynamics and thermal, so actually what I’ve tried to do within Rolls-Royce is give people the tools so they can go and do it in the safer way. So make it easier for them to do it the way I would like them to do it, then to kind of write some crazy macro in Excel that I’m sure we’ve probably had out there in the past.

Stuart Hughes:
So yeah, for me, it’s really about empowering people. I think empowering people is making it easier for them to do it the right way, making it easier for them to choose the right tech tool, rather than stopping them, berating them, because they’ll just find another way anyway, right? These people are super, super smart.

Chris D’Agostino:
So Stuart, for other executives that are in similar roles, one of the challenges that I saw leading the data engineering function at a top 10 bank is there’s this tension between the engineering department and the line of business where the engineering department is trying to build a suite of tools or an ecosystem that supports all of the use cases, has all of the data management functions, the lineage, that passport of data that you talked about. And the line of business, at the same time, needs to run a P&L, and so the shadow IT, I think, emerges because there’s this bit of tension between the time to delivery, speed to market from the engineering team, and what insights the business might need.

Chris D’Agostino:
But the risks, I think, from my perspective, I’d love to hear your feedback, is oftentimes the environments that are being done in the shadow IT are not full production, they don’t have maybe perhaps access to the right data, the security controls are not in place and then probably most importantly, is they won’t scale when they need to go to production, or the code quality may not be exactly where you need it to be. So can you talk a little bit about what advice you would give to somebody that’s in your type of role, where they’re seeing this phenomenon take place in the organization?

Stuart Hughes:
Yeah, and I think my background, again, has helped me. I think in large businesses, “IT said no,” would be acceptable. Whereas in mine, when you work in start ups that hyper-growth, there is no, “No,” right? There’s always a way, we just need to work it out. And I think that my attitude being that, at the start I think it scared my team a little bit, because they’re like, “Oh, saying no was our way. So now we’re going to have to come up with a smarter answer.” And I think we managed it, which is we created a team that is only small, it’s less than 10 people, but that is dedicated to people, to help them self-serve, so do it in the right way, and they’re only dedicated to providing the tools for them to do it in a supported way.

Stuart Hughes:
So we provided Databricks for some of our data analytics, we’ve provided a power platform so they can build a power app. We’ve provided PowerBI on the power platform. PowerBI can actually be pointed at Databricks, PowerBI seems to be able to connected to everything in SAP, so what we’ve really tried to do is say, “We will make it easier if you use these tools.” I think the number one thing that that team has produced other than this lovely, safe environment where people can put their app, put their model, put their analytic, is they’ve also been really clever and they’ve created a design style guide that when you open the power platform of some visualization software, you basically load a load of themes, so they basically instantly makes your report look beautiful.

Stuart Hughes:
So it’s a full design system, it helps people make things look beautiful, but we only support that on the systems that we want you to use. So it really helps people find their way. It’s like the breadcrumbs that help them find their way to our supportive way of doing things and I think that’s really, really important. But what I would never do to somebody is say, “Just because you don’t want to do it our way, you’re on your own.” Because that’s still wrong. Those people have still got to deliver for that business unit and still help our customers achieve, so you’ve still got to be careful.

Stuart Hughes:
But yeah, one of the challenges that we still have today, and it’s very difficult, is we have a lot of engineers who want to use R, we have a lot who want to use some of the other analytics, MATLAB, that’s what I was trying to think of. So people who want to use MATLAB, and actually, we want them to use Python and PySpark, right? So it is a convergence that has to happen, but I guess I go back to my original story of trying to convince the dot-net developer to switch from bespoke.net to Java on a kind of off-the-shelf platform, is I think you’ve got to make it easier for them to make that transition rather than black and white force them to. Because they almost dig their heels in more.

Stuart Hughes:
But the more we provide the design system, the more we provide tools that are highly scalable, the more we say, “Well, you can do that on your laptop, but it takes like 19 hours. Where if you put it in Azure, it takes three seconds.” The more of a reason that people are going to wander towards your tool and make the changes, and they work out, but actually it’s not that different after all, and it’s something that they can really buy into. So yeah, for me, it’s create a team that empowers people and give that team a strict remit that they must not do it for someone. So they mustn’t say, “Oh, we’ll do a power app, or I’ll build you a model or I’ll build you an analytic.” I think that’s a really important part of how you get that self-serve culture together.

Chris D’Agostino:
So Stuart, Rolls-Royce has a really interesting model in terms of the way in which you make money in the aviation space for aircraft engines. It’s by the hour, as I understand it, in terms of usage and so the company is really incentivized to keep these engines running and keep them in flight. So really, the modeling that you and your teams do for determining mean time between failures and making sure that you’ve got the supply chain for the parts to repair engines as they need to, the engines are in the right location geographically, you’ve got the repair teams in place, or they’re trained up through the aircraft company, can you give us some sense of how that modeling works and how it really ties into the P&L of the company?

Stuart Hughes:
Yeah, so our servitization, anyone who does servitization ends up with Rolls-Royce as a case study because I think 20 years ago that we were the first to kind of pioneer in this way. It’s really important to start with the yes, we do and extend the time on wing, so the engine to fly longer, but we’ve gone nowhere near the safety barriers, right? So this is about within a very, very high regulated safety industry. So just to put that caveat on it. But yeah, everything that my team does really, is forecasting to make sure that the total care product that we sell to our customers is going to behave and work for the customer and also obviously work for us and deliver the revenue that we need.

Stuart Hughes:
So we do some unbelievable things, so we can work with an airline to say, “Where do you think this airplane will fly from and to?” And then we will forecast for the next 30 years that route, how much it will cost them in fuel, what maintenance that we’ll need, which months they will need that maintenance in and really it’s a bit like buying a car and know everything that that car’s going to have to do as the customer from a repair, maintenance, overhaul kind of angle.

Stuart Hughes:
From our internal side, we can also tell you which parts we would have to buy, where that inventory would have to fly to and be when we need to do the repair, how much it would cost and do a financial forecast, because obviously we’re a FTSE 100 company, so we have a lot of audit around our finances and that’s a really key part of it as well. So yeah, it’s an incredibly complicated forecasting process that we do. If I think about the mechanism, we have a component called the blue data thread, so that brings engine data, so when the plane lands, we take a lot of data. On our newest engine, it’s about half a gig of data per engine, per plane, per flight. So you can see a significant amount.

Stuart Hughes:
If you look at some of the older engines, the data fits on a floppy disk, so there’s a real difference in what we get. We do also get some data live, so we have a one way stream from the plane into Rolls-Royce at certain points in the flight, and on the more modern airplanes, this’ll become more frequent and more data loads and sends will happen, but it’s very, very expensive, these things are 40,000 feet in the air. So yeah, we get the data in, we import it in through Azure, some data pipelines, we do some stream analytics, we do some batch analytics, and we run some diagnostic networks on the data that’s coming in and then we assess some of the alerts where things are out of what we would see as a parameter. With humans or with other models, or humans with models and to make sure that we take the right action based on the safety spec that we have for that engine.

Stuart Hughes:
That can be, when it lands, we work with the aircraft maintenance team to make sure that something is done. It could be an inspection, but it’s all the time, what we’re trying to do is make sure that when the airplane wants to take off, that it doesn’t not take off because of a Rolls-Royce engine. That’s a really important thing. While, of course, safety being the most important factor in that.

Chris D’Agostino:
So Stuart, we’ve talked about data democratization, the architecture, the way in which data moves, the governance components of that and how you need to keep a good check on the data. You talked about the data passport. You talked about the data science teams and the type of talent that you’ve been able to hire and bring into the organization that are PhD level curious people. Can you give us an example of a project that you’re working on now that kind of combines all these things and some of the lessons learned and how it’s moved innovation forward?

Stuart Hughes:
At Rolls-Royce, our data science team is called R Squared, and we create what are called challenges, so anyone in the business, from someone in the lowest roles to the most senior people can raise an innovation challenge. We created this concept called digital value premium, so it basically means what is it worth if we do this? It could be cash saving, it could be cash avoidance, it could even be revenue generated, and people will push ideas into this and then the data scientists pick them up in hopefully a sensible order. So one day you might be looking at what we talked about in gathering data on ticket sales. Another day you might be taking some temperature data from one of our ovens that we do casting in and we worked out that if the blade was in this position, it would have more of an opportunity to have an problem in quality than if we put it in this position.

Stuart Hughes:
Just really taking these challenges that people couldn’t do self-serve, right? They couldn’t do it if you… With self-service tools, with democratized data. These are really specialist skills that people have and that you need to hook into. But almost like making that the backlog of those people democratized, right? So the best idea is what they work on. The idea that’s going to generate the most value is what they work on. And I think that’s really exciting. The idea is that, that enable us to keep the engine on the wing longer, so it’s less disruption to the airline. It’s less disruption to customers. These are great things, but actually, helping our general business make better decisions I think is just as important.

Stuart Hughes:
When COVID hit, we found it really difficult to know what the engine flying our revenues was going to be, because we didn’t know if people were going to fly, whether they weren’t going to fly, which airplanes were going to fly, which airplanes weren’t. And you know, what’s great about Rolls-Royce is there’s always like 52 smart people who can tell you how it’s really obvious what we should do, right? So the obvious answer was to go and get some data from the web, how many people are looking on travel sites, how many people are looking at flights, what’s the booking conversion rate on a flight? And then tie that data set up with what tickets have been sold. So we start to see what’s the difference between the number of users looking for a flight, what tickets have been sold and then try and marry the tickets that have been sold to the scheduled flights.

Stuart Hughes:
Because one of the things that happened during COVID is that the airlines would keep three flights in the schedule and then on the last minute, they’d merge them all onto one plane and then just say to everybody, “You’re getting on that flight.” So for us, it was really, really difficult to know how many engine flying hours we were going to have and of course, that’s a really important part of our business. We were able to gather all those data sets in, chop them into a big blob into Azure and really start to do some analytics on that and get going really quickly.

Stuart Hughes:
And what was amazing for me was our chief customer officer, our sales director coming to me singing the praises of me team, we’ve never had such visibility. “I didn’t even know that this could be done,” you know, and there’s a Friday meeting and it’s one of the best attended where it’s just, “What’s the next thing we’re going to do? What’s the next thing we’re going to do?” And you’ve got three, four, five of the most senior directors in Rolls-Royce attending this call. The CFO even attends the call, right? Something that comes from what data scientists are doing and saying, “We’ve looked at this, we’ve compared our engines to Boeing’s engines, we’ve compared our flights to Boeing flights. We’ve looked at planes that have Rolls-Royce engines versus planes that have non-Rolls Royce engines.”

Stuart Hughes:
It’s a really incredible thing that opened up by just having that new problem and that new way of looking at it and having kind of infinite compute and having really powerful tools that allow you to do data analytics.

Chris D’Agostino:
So Stuart, one of the things that we aim to do with this series of Champions of data and AI is to help other aspiring people in our field, what lessons learned can we give them, and so when we think about where we are today in our careers, what advice would we be giving ourselves coming out of university? So if you can touch on that?

Stuart Hughes:
Yeah, and I hope my peers and my colleagues don’t mind me saying this, I think that one of the things that it took me a few years to learn is that the senior leadership team isn’t really a team. It’s a collection of individuals who’ve got very, very clear accountability and responsibility for a specific area, but they’re very loosely coupled as a team. So I think when you’re lower down in the organization, you’re looking up and you see this great team, you know, they must do everything together, make every decision together, but it’s just not like that at an executive level. Executive level, you are a loosely collected group of people with a shared set of outcomes and objectives.

Stuart Hughes:
Why is that really important? Well, I think it’s really important because one of the mistakes you can make is trying to be in their group. Because actually you’re trying to be in a group that doesn’t really exist. These are people who are friendly, professional, work together, but the shared goal is the success of the company. So how can you do it differently? And I think that this is a little bit of after the fact this is what I learned, but I’ve always been promoted the day after I succeeded after running at the fire, so wherever the big fire was that one of the very senior directors was concerned about, is I would drop in, get in there, take accountability for it, fix it and then come out and make that person look good, right?

Stuart Hughes:
And actually, that person, nine out of 10 times, would then say, “Oh, Stuart’s a great guy. Let’s promote him.” And I’ve kind of just gone round in my life doing exactly right. “Where’s the fire? Okay, I’ve got a great extinguisher, I’ll go and put it out.” And it’s always, always, always helped me. And I know that that’s like the anti of what the personnel or the HR or the people department would tell you, which is you have to plan your career, you have to think about everything in minute detail. To me, it’s not that. It’s you need to deliver great outcomes and actually, the biggest outcome that any senior person is going to want is something that is removing the current thing that they’ve got that’s one fire.

Stuart Hughes:
So for me, that was a great strategy. Also, how did that help me? Well, you know, one of the things that happens when you go to executive recruitment is… Well, there’s a few things that I didn’t expect. The first thing is you do lots of personality tests, numeracy tests, maybe most people know that’s going to happen, maybe they don’t. The next thing is you then get interviewed by a psychologist for four hours. Now, that’s scary when you’ve never done that before. So you’re kind of like, “Oh my god.” Everybody has insecurities, right? You go like, “Well, this person is going to get in my head and find those things.”

Stuart Hughes:
And actually having those great firefighting outcomes has given you is you have a real solid sense of confidence that whatever happens, I’ll work out the answer. And actually, what they’re really looking for is someone who’s super solid, someone who, “I know this person is going to be able to do that.” Someone who talks about going in and achieving quite big outcomes. So I did something and it changed the outcome of the company, and again, that’s a really key thing that people are looking in an executive recruitment process is they’re looking for people who are going to come in, come up with, we can call it strategy or come up with a vision, or something that will change and then drive that through to execution. And they’re looking for great examples, right?

Stuart Hughes:
When I always coach people, I say, “Please don’t tell me that we ran out of pens and you went and got some pens.” What I want to know is we had a digital workplace strategy, it meant that we rolled out Teams to 2,000 people in four days and that enabled us to then start working when at the time you couldn’t work. Those are the things, the great examples that get you through that executive recruitment process. I had a really strange experience in one of my first executive recruitment processes, which was for a company called TUI, which I think are the biggest package holiday company in the world, and the person said to me, “Oh, you’re the next generation of CTO. You’re like where this disruptive thing came from, right?” And I didn’t even say that, those weren’t my words, those were his words.

Stuart Hughes:
And actually, that was kind of cool for me, because I picked up what he said and it gave me the confidence to just be me. So I talked about how I behave and how I do things and I think it’s a really common piece of advice, be authentic, be yourself, but for me it’s, be authentic, be yourself, but talk about the outcomes, talk about the change that you drove, give them a reason to put you through to the next round of that recruitment, because I think that’s a really important part of it and for me, I often get feedback that, “You’re a very technical CIO or a very technical CTO or CDO.” But they always give me the job, right? So they say that with an element of negativity, but then they always give me the job and I think that’s something. Be proud, if you’re a technologist, to be a technologist because I think that might be what people are looking for.

Stuart Hughes:
And I think my final piece of advice is, my personal view is that if people are T-shaped, the long deep bit is your skill and mine is technology. I understand technology right down from the database, APIs, writing code, architectures, that’s just me and that’s what I enjoy most. So maybe that’s what gets me the role. Showing people innovative, disruptive, different things, talking with a passion about the outcomes that we’ve driven. But I think what keeps you in the role or allows you to go further on is the breadth. So it’s very easy to overplay the, “I’m a great technologist, so I’m just going to answer all technology questions.” And that’ll just frustrate people. You won’t be seen as a team player. So I think, yes, your technology or your technical skill might get you the role, but actually it’s the breadth that keeps you in the role or allows you to move up within the company, or it’s the breadth that becomes your reputation in the company.

Stuart Hughes:
In Good to Great, there’s a concept of level five leader and I would never say that I’ve achieved it, but I honestly do try and achieve some of the goals that it talks about. You know, I see where I am in the world now about coaching and helping people, not some of the things that I found as negative that people would say to me, “You’re bad at this, you’re good at that. You need to change this.” Helping people understand that in that context, the way you reacted wasn’t the right way, but actually, the way you reacted would be really useful in this other context, so don’t stop. Use what you did in that context, but in this context and then have a conversation with them about maybe how could we have done it differently? What were they trying to achieve? What was the outcome?

Stuart Hughes:
I think that coaching other people, as you get more and more senior, it’s the bit you have to enjoy. You’re more and more detached from the day to day pressing of the keys or the delivery. I always joke with people that most of the things on my CV recently, I didn’t do. I just led the teams that led the teams that led the teams, but actually, I’m really proud that I did lead the teams and I coached and I mentored and I helped them. And now I just see my job as removing the obstacles before they even get to them. I see my job as always working six months ahead, thinking about what obstacles are going to get in the way of delivery in six month’s time and making sure that I help clear them out of the way before they even knew they were going to be a problem.

Stuart Hughes:
And I think that maybe that’s the difference in the Stuart now versus the one that just got annoyed with management making stupid decisions. While I can’t go back to being that delivery technical person, but what I can do is try and clear the road for people to have an easier life. Because working in IT, working in technology, it’s not easy. It’s difficult. It’s not repeatable, it’s not like we work in a plant that generates the same thing over and over again. It’s very artistic, very creative and only [inaudible 00:42:45] people understand that and I think that’s something that I like to hold dear, that I can clear the road and help these people have an easier ride because I know how tough it is to be in a development team.

Chris D’Agostino:
Yeah, so I mean, in summary, it sounds like really sort of three things. The first was, don’t be afraid to work on the really hard projects and ones that have the potential to fail, but be one of those people that can help rescue those projects and make them successful. The second is to, while you’re very deep in a particular area, make sure that you pay attention to the broader skill set that you need to succeed inside of an organization and integrate in with the management there. And then as you become a manager and start leading teams, be that force multiplier, mentor those teams, take pride in leading them even if you’re not doing the hands on stuff anymore because you’re going to lead people to be able to succeed with those big projects that you were working on earlier in your career.

Stuart Hughes:
Absolutely. And I think if you do those three things, you’ll be seen as different and I think that’s a really positive thing. I think that the next generation of people isn’t the same as the last generation of people and I think that if I can help change things and move things forward, I think that’s a really important part of what we’re doing.

Chris D’Agostino:
Awesome. Well, Stuart, thank you. Everyone, Stuart Hughes again with Rolls Royce. Appreciate your time and I’m sure we’ll chatting again in the future.

Stuart Hughes:
Thanks very much, Chris.

Speaker 1:
Thank you for joining this episode of Champions of Data and AI, brought to you by Databricks. Thousands of data leaders rely on Databricks to simplify data and AI so data teams can innovate faster and solve the world’s toughest problems. Visit databricks.com to learn how data leaders are unlocking the true potential of all their data.