Data-driven banking: Inside Lloyds' 350-year transformation

Tarush Aggarwal (00:00)

Welcome to another episode of People of Data, where we get to highlight the wonderful people of data. These are our data leaders from various industries, geographies, and disciplines. Discover how companies are using data, their achievements, and the challenges they face. Today on the show, have Ranil Bhuteju. Ranil.

‍

has over 25 years of global experience in using data and analytics, machine learning to transform businesses. He's held senior leadership positions at major companies like the Commonwealth Bank of Australia, Vodafone, Standard Chartered, HSBC. He's currently the Group Chief Data and Analytics Officer at Lloyds Banking Group, where he focuses on data and AI to support the bank's mission of helping Britain prosper. He also serves as the Non -Executive Director at the Information Commissioner's Office, promoting data

‍

privacy and openness. Welcome to the show, Ranil. Thank you for having me. Of course. Are you ready to get into it? Yeah, let's go. Awesome. what does the Lloyds Banking Group do? You know, how do you guys make money and how does data play a role in helping Lloyds Bank do what they want to do?

‍

Ranil Boteju (00:50)

Thank you for having me.

‍

Yeah, let's go.

‍

Sure, so Lloyds Banking Group has been around for about 350 years. So we're one of the oldest banks, I would say in Europe actually. So we've been around for a really long time. And it is a conglomeration of quite a few different brands. So Lloyds Banking Group covers Lloyds Bank. It includes Scottish Widows, which is one of the UK's largest insurance investments and pension funds.

‍

Ranil Boteju (01:29)

It includes the Bank of Scotland, which serves our customers in Scotland, Halifax Bank, which is a building society that is very strong home loans, MB &A, which is a very strong cards retailer. There's a leasing business. There's a rental business. There's all sorts of wealth business. So it's a pretty major part of the UK economy. But fundamentally, we're driven by our purpose, which is helping Britain prosper. So all of these individual brands, they come together with a purpose to really

‍

Ranil Boteju (01:58)

really help customers across the UK and the nation to really prosper. That's really the goal. we're very purpose -driven, driven by this purpose. And so I always think about it, whenever I hire new people or whenever I talk to my team, what I say to them is, our goal and our mission is to help Britain prosper with data and AI. We've got access to one of the world's largest data sets, more than...

‍

Ranil Boteju (02:21)

More than 30 million customers, we've got about 20 million digital customers, which is more than Netflix has in the United Kingdom. So think about the richness of that data. We have a tremendous opportunity to really use that to help customers grow their wealth and to save and to secure their pensions and their retirement. So that's really the role of data at Lloyds Banking Group.

‍

Tarush Aggarwal (02:21)

it's interesting because as a banking group, one of, you you use data on a day to day basis, right? You have, you know, basic analytics, which sort of multiple of these subsadiries are going to have just as part of the banking product and the launch product or the financial products. But you also have this entire sort of data and analytics for your internal teams and, ML and all of these different functions. How do you think about, the differences between data as a product versus the data and analytics function and you know very curious to know how does that play a role with your business objectives.

Ranil Boteju (03:20)

Sure, sure. So look, the way that we think about data products, we could be building data products directly for our customers. So they could be things that enable customers to understand their savings, understand their spend, understand, kind of our future recommendations. So there's a set of data products directly to customers. There's a set of data products that we build for our internal colleagues, right? So these are things like various

‍

Ranil Boteju (03:47)

knowledge management capabilities or data products that help engineers write code or translate code or create test cases or data products that help people who work in the contact center go through call transcripts to do post -call reviews, training, or even things like knowledge management. So there's a whole set of data products for Colleague. And then we've also started to think about

‍

Ranil Boteju (04:09)

what could data products be that we build for other people that we would like to commercialize? So this is a new area. We haven't done much in that space, but that's really the third bucket. They're all equally important. So I would say that the data products that we build for colleagues are as important as the one that we build for external clients. However, I'd better say that there's a lot less risk. So when you think about some of the new technologies like generative AI, for example,

‍

Ranil Boteju (04:36)

We're doing a lot of use cases. Many of the use cases, many of the data products that we're building are focused on internal colleagues because there's a lot less risk of things going wrong.

‍

Tarush Aggarwal (04:45)

I love that you have a purpose -driven culture around making Britain prosperous. what does this map back to into, individual bits and pieces as you think about culture, the group, or is it largely driven by, each subgroup?

‍

Ranil Boteju (04:58)

No, it's very consistent culture. whether you're at Scottish Widows, whether it's on the Bank of Scotland, Halifax, it definitely does feel like Lloyd's. the way that I describe it is it's built up over, many hundreds of years. There is definitely much more of a community spirit. So our purpose helping Britain prosper. This is not a top down corporate affairs PR driven thing. It's something that is

‍

Ranil Boteju (05:23)

know, pervasive across the organization and what you'll find is individual branches do all sorts of activities to really help, you know, for example, local charities, to help local communities. We do a lot of work in that space. One of the things that I get involved in, my team gets involved in is quite a lot of work to really encourage, for example, students across the UK to really consider data careers, right? So we have...

‍

Ranil Boteju (05:46)

lots of people in my team, lots of the data scientists who would speak very regularly at all sorts of schools and universities, colleges, apprentice groups, and really try and demystify what data is and try and get people from disadvantaged backgrounds to really consider data as a gross. So that's an example of something that we do in my space. it's a very, I describe it as,

‍

Ranil Boteju (06:13)

I've worked at banks all over the world. Lloyd's has much more of a, it's a different feel and I feel, you know, it's really driven by our purpose. most of the organizations I've worked at, the purpose is usually something like, let's be number one in market sharing, in wealth management, or let's be number one in Asia. These are very, I've described them as, you know, quite finite goals. Whereas I think with Lloyd's, helping British prosper, it's an infinite goal, right? It's something that is so

‍

Ranil Boteju (06:40)

always striving to achieve that. It's much more about leaving a legacy and that has really impacted the overall culture in a very noticeable way.

‍

Tarush Aggarwal (06:50)

I love companies which have an underlying goal which unites, especially large companies which have multiple groups such as Lloyds. What is the culture like in your groups tend to be like?

‍

Ranil Boteju (07:14)

Sure. Absolutely right. So Lloyd's, as I said, it's been around for 350 years. We're in the middle of quite a significant transformation. I think it's been described as the world's largest transformation in financial services. So over the last few years, we've spent probably more than three or four billion pounds investing in data and technology in the data space. So I've been at Lloyd's for about almost three years now. So I've been pretty much...

‍

Ranil Boteju (07:38)

trying to transform how we think about data, not just within the data teams that are across the bank. So my biggest area of focus is to really cultivate within the team, a growth mindset. So I want everyone to really think about what can they do to really transform and improve how we're using data and AI, and really to move away from people who are, you know, focused on risk and control and trying to keep the ship steady to

‍

Ranil Boteju (08:04)

thinking about how they can use these technologies to really transform customer experiences, to transform colleague experiences, to really differentiate ourselves using this. So growth mindset is pretty much something that I'm trying to drive. The second aspect is really around curiosity. So I feel with data AI type roles, you've got to be really curious, you've got to be really thinking all the time about how can you use these technologies to solve new problems that no one's thought about.

Ranil Boteju (08:32)

you know, that no one's asked for, right? So it's a very new area. People won't come to you and say, I have this problem. Can you solve it with AI? the people that own those problems don't have the skills to do that. So what you need is a data team where people can actually look at a problem and then spot, here is an opportunity to solve that using the tools that have. So that curiosity is really important. And then the aspect of the culture I'm to really drive, not just through my teams, but across the bank is continuous learning.

‍

Ranil Boteju (09:00)

So if I think about everything that I'm now dealing with, generative AI, large language models, foundation models, ethics related to a whole bunch of topics that we never had hallucinations, none of this was on my plate three years ago. It's all thought that we've had to learn. And if I think about five years ago, I wasn't really working on cloud, right? That again, so what it's meant though is over the course of my career, I had to continuously learn and get my skills updated.

‍

Ranil Boteju (09:26)

And so I feel really passionately that for anyone working in data and AI, it's less about what you know, it's more about how you can learn new things and deploy those. And so really trying to build a culture and really foster continuous learning and really amazing to see we now hire, 100, probably about 120 data graduates in the UK. And all of them, they graduate, they come and work for us.

‍

Ranil Boteju (09:51)

but immediately they start to learn new skills. They do all sorts of online courses. the younger generation of data science and data engineers are really embracing the concept of continuous learning.

‍

Tarush Aggarwal (10:02)

Talking about your data team, what does the data footprint look like? How big is Lloyds as a holding group and how and what is the data? How big are the data teams across the entire group? Lloyds probably has about like 85 ,000, 90 ,000 colleagues, mainly in the UK, but we have about probably

‍

Ranil Boteju (10:16)

So Lloyd's probably has about like 85 ,000, 90 ,000 colleagues, mainly in the UK, but we have about probably three or 4 ,000 colleagues in India now as well. So we've built a technology center in Hyderabad. So with that, we have probably about 4 ,000 people in data roles, and we've grouped them into a set of data job families. So I know with quite a lot of precision how many data scientists we have across the bank.

‍

Ranil Boteju (10:45)

how many data engineers, how many people in data management, data strategy, and also an MBA. So we've categorized people into all of those different jobs. So it's about 5 ,000 people, which is, you know, it does make us one of the biggest data teams in the UK. And what we also know is, for each of those job families, we've also assessed people's skills. So we know exactly what competencies they have. And we're now focused on ensuring everyone has

Ranil Boteju (11:11)

very individualized learning plans that we're continuously helping colleagues upskill as well. So it's a pretty good footprint. We're upskilling people as well. And we're also hiring, we've scaled our graduate programs. It'll be about 120 data graduates a year now from this year.

‍

Tarush Aggarwal (11:23)

Wow, if there's one piece of advice to somebody listening in who's super excited about the culture you've created and the work Lloyds is doing, what is one piece of advice you would give them in building a career at Lloyds?

Ranil Boteju (11:43)

Look, what I'd say is I think the whole concept of being curious and continuous learning and growth mindset. So what I would say is always try and assess yourself against those three characteristics. More people are starting to really understand that those are the three drivers that really predict success. So if someone at university, try and think about the things that you can do to demonstrate that you have a growth mindset. So whether it's, you know,

‍

Ranil Boteju (12:10)

organizing hackathons or organizing all sorts of activities for your fellow data scientists and data engineers. Try and demonstrate how you've learned new skills that you didn't have to. So anything that demonstrates a real passion for continuous learning. And then also curiosity. So think about how can you demonstrate you've been really curious about something you've then found out about it and tried to solve a problem. So I think those are three things that I would really focus on. The other thing I'd say

‍

Ranil Boteju (12:37)

We're in a world now where we're awash with all sorts of online tools, right? So if I think about generative AI, there's probably about 20 or 30 amazing tools out there. Play around with those, learn new things, test them out. Those are things that are always going to help you but also demonstrate that you've got a real passion for learning and curious and growth mindset.

‍

Tarush Aggarwal (12:44)

What does the data stack look like across the group? Do you have any consistent stack? I'm assuming there's a lot of different data stacks. How do look at that? 

‍

Ranil Boteju (13:11)

so it's 350 years old. We have lots of different brands, right? So historically, we've been very fragmented. So we've had just in the BI space, multiple tools. What we're now doing though is we are now converging on a common set of tools. it's public, we made these announcements, we're converging on the Google Cloud Platform for a lot of our data analytics tools.

‍

Ranil Boteju (13:33)

What we want to do is have a consistent set of tools so that we can build a pattern and then really use that and scale that across the organization. a much better way to work if you have consistency across your organization. It makes it easy for things like career pathing,

‍

Ranil Boteju (13:52)

build once, deploy many, scaling, support, evergreening, all of these things become a lot easier if you've got a simple application is a very big part of our agenda. Having said that though, I'm very conscious that in the world that we are, we also don't want to over double down on too few supplies. So things like portability are very important. And then in the world of AI,

‍

Ranil Boteju (14:18)

very strongly believe that as one of the UK's largest banks, we have a real duty to really foster the open source ecosystem as well. So a lot of our infrastructure for AI modeling does enable us to bring in external models, open source models as well. And that's something I'm very keen to really propagate and encourage.

‍

Tarush Aggarwal (14:38)

What is one thing which you have been really proud of in your tenure at Lloyd's which is is caused a real business impact?

‍

Ranil Boteju (14:47)

One of the things I'm really proud of is the focus that we had in the organization on data culture. So I've been running data strategies for about 15, 20 years now, right? So I've always focused on fairly standard things like data platforms, AI, data management. At Lloyd's, when I thought about our strategy, I also introduced data culture as a core pillar, And the focus of that one is in addition to all of the technology around data and all of the...

Ranil Boteju (15:14)

algorithms and the data skills, having a pretty heavy focus on upskilling the broader organization. So we have a massive focus on ensuring that our colleagues in non -data roles are also suitably upskilled and literate about how to use data. So that's a learning I've kind of had over the last few, maybe decade, that unless you invest in your business partner's data skills, you will never really achieve the ROI. So that's something that we put a lot of effort and focus on. You know, we spend a lot of

‍

Ranil Boteju (15:43)

with our senior leadership team, with the CEO, with his team. And we spend hours getting our PhDs and our experts to explain some of these really complex topics so that they know as much about these things as almost as we do, which is fantastic. So that's something that we've done, which I don't see many others do. And that's something I'm quite proud of.

‍

Ranil Boteju (16:09)

Yeah, one of the challenges has been around trying to move off legacy. So we have target data that we get to. The reality though is when you have a very fragmented set of siloed data systems of insight trying to converge onto a single stack is very complex. It's expensive, takes a lot of time. That is quite challenging, right? So often the knowledge is no longer exists in the organization. There's lots of code that needs to be unpicked.

Ranil Boteju (16:39)

those things have been quite challenging. And the older the organization is, the more legacy the harder it is. Sometimes, if you're in the years of startups and organizations where they're starting with a clean sheet of paper, they get to build their stack, rounds up with the latest technology. That is a lot easier than having to transform a very complex legacy business. It's a lot harder and requires very different skills to what you'd need in a Fintech or in a Google or an Apple type

‍

Tarush Aggarwal (16:55)

And the big question on everyone's mind is AI and we know companies are responding to it in different ways. How does your AI strategy affect your general your general data and analytics strategy at you, is AI going to be part of the same group? Are you looking at it with different headcount? How does it impact roadmap and the business?

‍

Ranil Boteju (17:08)

So that's something that has been challenging, but we're really facing into that.

‍

Ranil Boteju (17:33)

So I'm not sure if you're familiar with the Trojan horse, right? So from the Iliad, right? So effectively it's a way to smuggle in like a concept or in the case of the Iliad, was a whole bunch of Greek soldiers in Detroit. But AI has been the Trojan horse for data and analytics. So I've been in this space for 25 years. And to be honest, data has always been something that CEOs and boards have been interested in, almost like a bit of a hobby.

Ranil Boteju (18:00)

They wanted to talk about it. They were interested. But it was seen as a super interesting hobby. What's happened with generative AI and AI in particular, the last two or three years is now people really understand that this is a general purpose technology that will fundamentally transform society in the same way that electricity and steam has done. And so all of a sudden the whole data agenda is now something that everyone takes very seriously. So for the first time in my career, I am

‍

Ranil Boteju (18:29)

meeting up board on a very, very regular basis to talk about all sorts of data topics. So to summarize, what I'd say is AI has made everything related to data super visible, important, and people are taking note. It needs to be very integrated with your broader data strategy. So basically, for you to do AI, your data needs to be organized, it needs to be in the right environments, you need to have the right tools, you need to have the right data cataloging, data products, you need to have the right...

‍

Ranil Boteju (18:57)

access control. So everything related to data is now even more important. my view is that, you know, data and AI, these are really pervasive, technologies, they won't be done by central teams. So at Lloyd's, have a centralized, decentralized model. So there are data and AI teams all over the bank.

‍

Ranil Boteju (19:15)

And then we also have a central set of central teams. have high -end experts. You need to have both. But really, the reality is, if you fast -forward to, three four years, the way that we think about data now will be the same way that we think about typing, word processing. So if you rewind 30 years ago, there were people called typists, and there was a concept called a typing pool. And if you wanted a memo, you'd send it to their dictator, and they'd type it. Now everyone does their own.

‍

Tarush Aggarwal (19:38)

Yeah, that makes a ton of sense. Ranil thank you so much for your wisdom today.

Ranil Boteju (19:43)

analytics and data will be the same, right? So most work will be self -serve and everyone will need to do this is my very strong view.

‍

Thank you. Hey, look, really enjoyed talking to you and hope your listeners got to learn about Lloyds Banking Group. So whether you're in the UK or India, we have a ton of roles. Check out our website, we're growing a bit, but particularly in data and analytics.