South Africa’s banking sector remains the powerhouse of the continent, accounting for almost 30% of Africa’s banking income, with revenues forecast to hit $30bn by 2030.
This is largely built on a bedrock of disciplined capital management and sophisticated regulation that has helped enable the “Big Four” banks ― Standard Bank, FirstRand, Absa and Nedbank ― to anchor 82% of domestic assets through decades of economic volatility.
However, this traditional fortress is being challenged. While incumbents traditionally rely on their scale to drive revenues, specialised players such as Capitec and TymeBank are increasingly using digital agility to expand services, cut the cost to serve and gain ground.
The defining trend over the next decade will be the ability of South Africa’s established banks to pivot to agile, AI-enabled operating models fast enough to maintain market share in a rapidly expanding African banking ecosystem.
Analysis from McKinsey’s latest research on African banking shows that banking across the continent has grown at about 17% annually in constant currency terms over the past five years. Profitability has been equally striking. African banks delivered returns on equity of 19% in 2024 and 17% in 2025, well above the global average of 10%.
Much of this momentum is occurring in markets where banking systems are still being built. Digital payments are transforming transactions and extending the reach of the formal financial system. In these frontier banking markets growth is primarily driven by new customers entering the banking sector.
However, in mature markets such as South Africa, which already has high levels of financial penetration and one of the most developed financial systems in Africa, growth is more likely to come through productivity, innovation and new revenue models.
The shift is already visible. Competition in banking is evolving in ways that would have been difficult to imagine a decade ago, with financial services increasingly embedded into everyday economic activity.
Payments are now integrated into commerce platforms, credit is offered within supply chains and digital marketplaces, and new platforms are reshaping how financial products reach customers. In many cases banks provide the regulated balance sheet while platforms control the customer relationship.
This opens the door for fintechs, telcos and digital players to scale embedded financial services, with Africa’s fintech revenues expected to grow fivefold to about $47bn by 2028. For banks this means embedding into ecosystems rather than owning the customer interface end to end.
Discovery Bank’s model linking wellness to banking rewards and Vodacom’s Airtime Advance show how financial services are becoming part of broader digital ecosystems. In this world the bank is no longer a destination; it is an invisible, trusted layer within a platform.
To compete in this environment, banks will need to move AI from experimentation to the core of the enterprise. While AI is already used across fraud, credit, risk and customer engagement, much of it remains fragmented. The real opportunity lies in scaling it across the organisation to reduce costs, accelerate decision-making, and unlock new revenue streams.
This will require a reset of integrated technology and data infrastructure, as well as governance and talent, to translate AI into sustained productivity gains and stronger financial outcomes.
This is ultimately a race on speed. Banks that move decisively will use AI to streamline operations and shorten lending cycles for consumers and small and medium-sized enterprises. But speed without trust will not hold. Automation must be secure, compliant and explainable, with Protection of Personal Information Act (Popia) grade consent embedded across key models.
The prize is a lower cost to serve and the ability to reinvest in higher growth, fee-based businesses. Those that hesitate risk falling behind more agile competitors already building AI into the fabric of how they compete.
This is ultimately a race on speed. Banks that move decisively will use AI to streamline operations and shorten lending cycles for consumers and small and medium-sized enterprises.
For large incumbents this will require a fundamental mindset shift. In a market where neobanks release features weekly, the traditional six-month update cycle is no longer viable. Leading banks are replacing legacy systems with cloud-native technology and application programming interfaces (APIs), while simplifying their offerings to focus on what they can deliver exceptionally well.
South Africa’s banks now face a clear inflection point. They enter this phase from a position of strength, with deep pools of capital, trusted brands and millions of customer relationships, supported by a well-established regulatory environment. But advantages only endure when they evolve.
While improved GDP growth and declining money market rates may offer some relief to banks still recovering from the challenges of Covid-19, these benefits are likely to be offset by the scale of investment required in technology, data and AI to remain competitive. This will demand greater efficiency, resilience and agility, as well as a willingness to rethink how banking is delivered and where value is created.
South Africa built Africa’s most powerful banking system through scale, discipline and strong institutions. The next decade will test whether it can translate that strength into leadership in ecosystems and AI and redefine what a bank looks like in a digital economy.
• Kuyoro is partner and head of the Africa financial institutions practice at McKinsey & Co Africa, and Sambandan is a partner and head of the McKinsey SA financial industries practice.
Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.
Please read our Comment Policy before commenting.