The Opportunity
Embarking on a data transformation
Artificial intelligence could deliver massive value globally. According to our research, across the Middle East alone, AI could create as much as $150 billion in value, or the equivalent of 9 percent of the combined GDP in the Middle East’s Gulf Cooperation Council (GCC) countries.
Emirates NBD (ENBD), a leading bank in the United Arab Emirates, saw this opportunity and has taken a bold step: embarking on a transformation journey to become an AI-driven organization.
“ENBD had just completed a multiyear IT transformation that significantly simplified the IT landscape, digitized many core processes, and created a bankwide data lake. Advanced analytics and AI were the future growth engines to drive value from IT transformation investment,” says ENBD’s Group Chief Digital and Information Officer Miguel Rio Tinto.
With more than 20 million customers and over 30,000 employees across 13 countries, it has been a complex multidimensional undertaking. What’s more, in the financial sector, large-scale AI transformation often requires significant upfront investment and a long lead time before delivering measurable impact.
A lot of banks see AI as a huge investment that delivers very little initially. They understand it will eventually deliver, but find it takes too long. If you choose well, especially leveraging gen AI capabilities, you can bootstrap the investment through impact accrued initially while you scale up.
The Approach
Bootstrapping to scale
In 2021, ENBD embarked on the first phase of this transformation by partnering with McKinsey. “It was crucial for us to establish an advanced analytics strategy and roadmap that aligned with the business early on so that we could quickly scale use cases and impact,” says ENBD Head of Strategy, Analytics, and Venture Capital Neeraj Makin. “Sponsorship from business heads played a key role in shaping the use cases and broad adoption among the frontline.”
The focus was on high-impact, lighthouse use cases that demonstrated substantial value across business units and had executive support. To minimize change management across the front line, analytics outcomes were embedded into existing business applications.
Like all banks, we had huge volumes of uncatalogued, multisourced data of varying quality and unclear provenance. Data discoverability, understandability, and quality were significant challenges.
To jumpstart the journey, a select group of individuals with thorough understanding of the intricate data landscape were brought into the core team. This small team curated the data assets necessary for the initial lighthouse use cases. Scaling this data effort then required a markedly different approach in which a data-mesh-inspired strategy was seeded by the small central team. This team drove a fully federated approach to data governance and management across the business.
To safely scale analytics, the bank worked with QuantumBlack, AI by McKinsey, on capabilities such as an automated CI/CD framework and containerized builds that eliminate reproducibility issues. A feature store was created to deliver data and machine learning (ML) pipeline observability and reduce build time. A model validation framework was developed to standardize, validate, and auto-generate model documentation.
“As business demand for analytics grew, there was a need to transition from large-scope, long-cycle use cases to much leaner, short-cycle, iterative delivery releases,” says ENBD Group Head of Wholesale Banking Ahmed Al Qassim.
ENBD took a test-and-learn approach to this transition and leveraged data and analytics assets developed for scaling. In addition, business impact was measured and tracked using standardized ML-based synthetic control groups, closely supported by business and finance teams.
Building a robust talent pool was crucial throughout this process. Initially, ENBD focused on building data science capabilities while reskilling existing staff for roles like delivery leads and data engineers.
“When we talk about AI, there is huge focus on building technical capabilities. But focusing on change management is equally important. Techno-functional roles are key to bridging this gap,” says McKinsey partner Saadi Azeem.
As the transformation progressed, the increased need for techno-functional roles became critical to drive change management and identify future use cases. The bank initially focused on external hiring to fill these roles. In parallel, ENBD designed an upskilling program with McKinsey Academy and supported internal events like generative AI (gen AI) hackathons to build a broader AI community and develop capabilities within the organization.
The Impact
Scaling across the organization
ENBD has demonstrated scalable growth in the industry through this strategic AI transformation. Embedding scalable analytics across the business has improved operational efficiency, and institutionalized data and analytics decision-making across the organization.
ENBD can now use predictive AI to create personalized customer experiences. For example, in retail banking, a hyper-personalization capability can be used to advise mass segment clients, including first-time investors, on tailored investment solutions.
“Advanced analytics models can help identify high-potential prospects, drive better engagement with existing clients by matching them to the right products and services, and assist frontline teams with tailored insights and talking points,” says ENBD Group Head of Retail Banking and Wealth Management Marwan Hadi.
In its first two years of this effort, the bank has produced more than 100 models and built a core advanced analytics team larger than 70 people. Overall, ENBD aims to generate a five to seven times return on its AI investment through the business value generated from analytics and data-driven initiatives.
“From board decisions to every customer interaction, we intend to make AI a crucial element in each business moment and event,” says Ray.
Related Case Studies
Banking on innovation: How ING uses generative AI to put people first
How Lufthansa is using data to reduce costs and improve spend and carbon transparency
Building a next-generation carbon platform to accelerate the path to net zero
Let’s Connect
Saadi Azeem
Jawad Khan
Ashwin Naib
