For the past decade, African banks have competed fiercely on customer acquisition. Branches, agents, USSD codes, and mobile apps were all tools to sign up as many customers as possible. That race has largely been won. The new, harder competition is not about who has the most customers, but who understands them best. On this front, most banks are falling behind, and the root cause is that they treat customer data as exhaust rather than as infrastructure.
Data Fragmentation Across Systems
In a typical bank, customer data is scattered. The app team holds one view of the customer, the cards team another, the agent network a third, and the call centre a fourth. These systems do not communicate in real time. Consequently, a bank with millions of customers still communicates with all of them in roughly the same way, pushing identical messages about the same products to people with vastly different needs. Customers tune out, engagement remains low, and marketing spend quietly evaporates.
The Role of a Customer Data Platform
A customer data platform (CDP) offers an unglamorous but effective fix. It is important to distinguish a CDP from a customer relationship manager (CRM) or a data warehouse. A warehouse stores data for later analysis. A CRM tracks interactions logged by staff. A CDP does something different: it resolves the many fragments of a single person—app login, card swipe, agent deposit, support call—into one persistent profile, and makes that profile available for real-time action. The commercial value lies in acting on that data immediately.
Real-Time Action Drives Value
With a unified data foundation, the economics shift. A bank can recognise that a customer who just received an inflow is a candidate for a savings product rather than a loan, and reach them through their preferred channel at the moment the message is relevant. In my work building engagement infrastructure, the pattern is consistent: unify the data first, and previously stubborn numbers begin to move. Engagement rises because the message fits, and churn falls because customers no longer feel like strangers to their own bank. These are not marketing abstractions; they are line items a CFO can see.
Challenges and Governance
None of this is automatic. The same machinery that enables personalisation can tempt banks to over-message until customers disengage entirely. Data governance is not a compliance afterthought but a prerequisite. Consent must be treated as something customers grant and can withdraw, not buried in terms nobody reads. A platform that ingests everything and protects nothing is a liability waiting to surface.
African markets add unique complications. Many customers are thin-file, with little formal credit history. They carry multiple SIM cards and switch devices, making identity resolution genuinely difficult. A large share of activity still happens offline through agents, so any honest single view of the customer must reconcile physical and digital interactions. Solving identity resolution under these conditions requires hard engineering, not a configuration setting. Institutions that treat it lightly will build expensive systems that produce nonsense.
Caution on AI Adoption
The current rush toward artificial intelligence in banking warrants caution. AI is only as good as the data beneath it. A model trained on fragmented, duplicated, poorly governed records will not produce insight; it will produce confident noise at scale. The banks that benefit most from AI are not those that buy it first, but those that did the unglamorous work of unifying and cleaning their data beforehand. Sequence matters more than speed.
Policy and Talent Gaps
Policy also plays a role. Nigeria’s data protection regime and the work of the National Information Technology Development Agency have set expectations around customer information handling. Clearer rules, far from being a brake, give institutions the confidence to invest. The other constraint is talent. We do not have enough engineers who understand both the regulatory weight of financial data and the practical craft of making it usable. Closing that gap will require deliberate investment from banks and technology firms, not poaching alone.
Looking Ahead
The institutions that prosper in the next decade will not necessarily be those with the largest customer numbers or the flashiest apps. They will be the ones that decided, early and seriously, to treat customer data as core infrastructure: to unify it, govern it, and act on it with restraint and precision. The asset is already on the balance sheet. The only question is whether African banks will finally choose to use it.
