Putting Analytics Practitioners at the Heart of Value Creation at Lloyds Banking Group

Ranil Boteju, Group Chief Data and Analytics Officer at Lloyds Banking Group discusses navigating the future of banking, pioneering data-driven innovation, and the transformative promise of generative AI

In the dynamic world of banking and finance, harnessing data to serve customers in innovative ways is the difference between mediocrity and excellence.

With over 26 million customers, Lloyds Banking Group is present in almost every community in the UK and includes some of the nation’s best-known financial brands, including Lloyds Bank, Halifax, Bank of Scotland, and Scottish Widows.

In this competitive and ever-evolving sector, the ability to analyse, interpret and act on vast volumes of information is no longer a luxury but a necessity for banks to stay relevant.

From implementing machine learning to improve customer experiences to innovating with generative AI, conversational banking, and advanced analytics, Lloyds Banking Group is at the forefront of using technology and data to transform how it serves its customers.

Heading up the organisation’s data, analytics, and machine learning operation is Ranil Boteju, Group Chief Data and Analytics Officer at Lloyds Banking Group. Boteju has been transforming businesses and creating value with data and advanced analytics for more than 25 years.

Boteju recently sat down with Orbition’s Founder and CEO Kyle Winterbottom to discuss his journey in data and analytics, the evolution of value creation at Lloyds Banking Group, and how innovative technology like generative AI is reshaping the future of banking.

The Power of Purpose

In an era of technological advancements and evolving customer expectations, the financial sector has needed to transform rapidly.

The transformation at Lloyds Banking Group is profoundly embedded in a sense of purpose.

“At Lloyds Banking Group, it’s all about helping Britain prosper, and that really resonates with me,” Boteju says. “With the skills that I have and the teams that I look after, whether it’s machine learning, AI, or conversational banking, all of these capabilities can supercharge that purpose.”

Boteju continues: “The Group boasts one of the UK’s largest data assets. We’ve got 26 million customers giving us a profound reach into the country. So, there’s a really good opportunity to drive great outcomes.”

A significant factor in the bank’s evolution is the influence of its senior leadership. “[Lloyds Banking Group] is going through a massive transformation under our CEO, Charlie Nunn. He’s created a very transformational, growth-focused leadership team.

This passionate leadership makes the confluence of purpose, technology, and leadership infectious, and clarifies Boteju’s objective. “My aim? To transform this business using data and machine learning to truly help Britain prosper.”

Evolving Views of Value Creation

In today’s fast-paced digital economy, understanding and measuring the value of initiatives, particularly in analytics, has become paramount for enterprises. Yet, as Boteju reflects, the lens through which this value is evaluated has undergone a significant transformation over the past decade.

“The meaning has changed over the last 10 years,” Boteju begins. “Back then I would have told you that value had to be tied to a revenue or profit number. We would measure every facet of analytics and assign it a monetary value. It was an incredibly intensive effort. And it was successful. We used [the numbers we generated] to justify a lot of investment and it helped us generate a lot of believability.”

However, businesses consider the larger picture and their role in supporting society overall. “Many began to realise that while analytics played a part, other factors, such as better content, enhanced human interactions, and improvements to the overall customer journey, also drove outcomes.”

With the world of business evolving, the methodologies to assess contributions from various business units had to evolve too. “I no longer want to have a single number for the value of data and analytics,” Boteju explains. “Now, we assess value based on much more specific metrics depending on the activity—be it outcomes from marketing campaigns, back-office automation or customer satisfaction with conversational banking.”

This transition from blanket metrics to a more nuanced approach indicates a broader trend in enterprises. No longer are CFOs and CEOs looking for one-size-fits-all numbers. Instead, they are urging teams to take a more granular look, understanding the different facets of value each unit brings to the table.

This shift in perspective showcases how enterprises are maturing in their understanding of value creation. They recognise that the contributions of every department are multifaceted and that true value is a composite of many nuanced metrics, not just one.

“It’s more specific to the use case,” Boteju concludes. “You can use that to understand the effectiveness of the specific analytical workload that you’ve got. You can also use it to build a business case. Many mature businesses get that now, rather than this notion of what ‘the big number’ is.”

The Future of Analytics Skills and Value Creation

As enterprise analytics and AI continue to change, understanding how to cultivate the necessary skills to create value is paramount. What’s more, opinions about how best to deploy your technical experts have shifted over the years.

According to Boteju, the approach to analytics in enterprises is undergoing a profound shift. This involves a move away from broad, centralised analytics towards more outcome-specific measures that resonate more closely with individual business objectives.

“The whole notion of federating analysts and data—that is where we need to go now. It’s become a much more mainstream capability,” Boteju says.

But what does this decentralised model entail? “In five years, I think the analytics people will be almost invisible within teams. [Analytical] skills will be much more widespread,” Boteju predicts. Such a shift underscores the expectation that analytical skills will not remain confined to specialised departments but will be woven into the fabric of various business units.

Boteju emphasises the need for the continuous growth of skills in enterprise teams. “Our objective is to continually develop competencies, introduce new specialisms, and then integrate them into business teams,” he says. “It’s not a ‘once-and-done’ thing.”

He cites generative AI as a focal point now, but its skill distribution is set to broaden over time. The evolution of Natural Language Processing (NLP) at Lloyds Banking Group serves as an apt example. What was once a specialised skill in the central team has now been adopted by federated specialists building conversational bots within their teams, illustrating the changing dynamic of skill deployment in the enterprise.

But with these changes, how does one ascertain value? Boteju reflects on past challenges: “Years ago, I was leading big analytics teams centrally and found that I just wasn’t close enough to the [operational execution] to really understand the value.”

He continues: “As teams have become closer to business operations, it’s become a lot easier [to recognise the value]. The big shift is actually having these capabilities embedded where revenue and value are actually being delivered.”

The trajectory of analytics in enterprises, as envisioned by Boteju, paints a picture of integration, evolution, and closer alignment with business goals. As data skills become ubiquitous and analytics specialists increasingly work in tandem with business units, the path to a clearer understanding of value creation is bright.

Banking’s Next Frontier: The Potential and Pitfalls of Generative AI

As technology continually reshapes the banking and financial services sector, generative AI is increasingly viewed as a powerful force for change. Since ChatGPT stormed onto the market and into the public consciousness in November of 2022, its adoption for business use has been notably brisk.

“For the first few months, I was taking a wait-and-see type approach. However, what I observed very quickly was organisations all over the world finding quite useful and value-accreting use cases,” says Boteju.

In comparison, other much-discussed tech trends like quantum computing, blockchain, and Web 3 struggled to provide meaningful benefits to enterprises in the short term.

The swift integration of generative AI into vendor offerings and its potential to solve pressing issues in banking further demonstrated its growing influence.

“Across the bank, I saw various teams starting to use [generative AI] to solve actual problems,” he says.

These experiments were driven not by a desire to be ‘on trend’. Rather, generative AI was providing a means to solve problems that had defied other techniques.

However, despite the excitement surrounding generative AI’s potential and a genuine interest in generative AI’s problem-solving capabilities, Boteju also recognises its present limitations.

“[Generative AI] is very new. So, there’s a lot of hype around what you can and can’t do, particularly when it comes to scaling these capabilities. And that’s where I think a lot of the big players may have oversold where we are.”

Boteju categorises the primary applications for generative AI in the financial sector into three distinct groups: direct customer interactions, a human-in-the-loop approach to assist professionals, and addressing specific technical challenges like understanding metadata at scale or writing code quickly.

“We’re focusing on the second and third categories,” he says.

The desire to focus on non-direct customer interactions at first stems from generative AI’s current inability to consistently generate accurate and context-aware content, which can pose a risk in customer-facing situations where precision and trust are paramount.

Additionally, scaling such a technology presents its own set of complexities. While generative AI can process vast amounts of data and generate information, ensuring its outputs remain consistent, accurate, and free from unintentional biases remains a challenge, especially as the technology scales across varied use cases.

In closing, generative AI certainly harbours the potential to redefine facets of banking and financial services. However, as industry visionaries like Boteju suggest, a measured and cautious approach is vital, especially when customer trust and stringent regulatory standards are at play.

Generative AI’s entrance into the banking scene has sparked curiosity, experimentation, and caution. While it presents a promising avenue for problem-solving, its infancy necessitates a measured approach, especially in sectors where trust and reliability are fundamental.

As the landscape of data analytics in banking shifts towards a more federated, outcome-specific approach, understanding and harnessing new technologies like generative AI is critical.

The journey of value creation at Lloyds Banking Group thus underscores a delicate balance between innovation and responsibility, pointing towards a future where data practitioners are instrumental in weaving technology and trust seamlessly together.

This article was featured in the first Edition of our Driven by Data Magazine. You can download the magazine and read more articles like this by clicking here.