A treasurer at a large European bank once described her job, only half in jest, as “making sure the bank gets paid for what it does.” She was not talking about loan defaults or trading losses. She was talking about something far more mundane, and in aggregate, far more expensive: the gap between what her bank agreed to charge and what it actually collected.
This gap has a name: revenue leakage, and in banking it is something of a taboo subject. Executives do not speak of it in earnings calls. Consultants euphemize it as “billing optimization.” Auditors find traces of it tucked into reconciliation reports that few people read. Yet the numbers, when anyone bothers to assemble them, are startling. One estimate puts the annual losses from reconciliation inefficiencies alone at nearly $100 million across the financial sector.1 Even that likely understates the problem as individual banks have reported losses running into millions within a single market. The true figure, including mispriced deals, uncaptured fees, and quietly normalized exceptions is certainly higher.
The silence is not ignorance. It is embarrassing. Banks pride themselves on precision—their systems process millions of transactions a day without error. Admitting that the pricing on a syndicated loan was never correctly loaded into the billing engine, or that a corporate client has been receiving a fee waiver for three years because nobody updated the system after the original relationship manager left, does not fit the image. And so, the leakage persists, quarter after quarter, hidden in plain sight.
Where the Money Goes
The anatomy of revenue leakage in banking is less dramatic than the term implies. There are no rogue traders or fraudulent schemes. The culprit is more prosaic: complexity, accumulated over decades, that has outrun the systems designed to manage it.
Consider how a modern commercial bank prices its services. A corporate client does not simply receive a fixed fee schedule. It receives a relationship fees structure with a negotiated bundle of rates, tiers, waivers, volume discounts, and preferential terms that may span cash management, trade finance, FX, and lending. The deal is agreed by a relationship manager, documented in a term sheet, and then—this is where the trouble begins—passed to multiple downstream systems that were never designed to talk to each other.
The product catalog lives in one system. The pricing engine in another. The billing module is in a third, often a legacy platform dating from an era when relationship pricing was simpler, and exception-handling was someone’s full-time job. Each transition from agreement to product definition, from product definition to pricing rule, from pricing rule to invoice is a potential point of loss.
The exceptions are particularly insidious. Every bank has them: the client who was granted a fee waiver during a credit crunch and never had it removed; the pricing tier that was manually overridden to win a deal and then became the de facto standard for that client segment; the product launched with a promotional rate that expired on paper but not in the system. Individually, these are rounding errors. Across a book of several hundred corporate relationships, they become material.
A Structural Problem, Not an Operational One
The standard industry response to revenue leakage is an audit. Find the gaps, plug them, move on. This approach misunderstands the problem. Audits are retrospective by design. They identify where money was lost; they do nothing about the conditions that will cause it to be lost again next quarter.
The underlying issue is architectural. Banking’s commercial model has evolved much faster than its execution infrastructure. Twenty years ago, relationship pricing was a concession – a discount here, a waiver there. Today it is the model. Corporate banking has moved toward highly bespoke, multi-product arrangements where the value of the relationship is embedded in the pricing structure itself. A client that holds its payroll, trade finance, and FX business with the same bank expects to be rewarded for the totality of that relationship, not just charged for each product in isolation.
That is commercially sensible. But it requires an execution layer capable of translating complex, multi-dimensional deal terms into consistent, real-time billing across every product and channel. Most banks do not have one. They have a patchwork of systems, each accurate within its own domain, that collectively introduce error at every handoff. The relationship manager knows what was agreed. The client knows what was promised. The billing system knows only what it was told, and it was told imperfectly.
Simon-Kucher, a strategy consultancy specializing in pricing, has documented this pattern across dozens of banking engagements. Pricing misalignment, they found, is one of the most persistent sources of hidden profit loss in financial services,2 not because banks do not care, but because the problem is genuinely hard to see. Leakage does not announce itself. It accumulates in the background, a thousand small deviations from what was intended, each one below the threshold of managerial attention.
The Cost of Invisibility
There is a deeper irony here. Banks are among the most data-rich organizations on earth. They can tell you, in real time, the mark-to-market value of a complex derivatives portfolio. They can model the credit risk of a sovereign borrower to several decimal places. Yet many cannot tell you, with confidence, whether a specific corporate client was billed correctly last month.
This is not a technology deficit in the conventional sense. It is a data architecture problem. The information needed to verify billing accuracy exists in CRM systems, deal management platforms, pricing engines, and general ledgers. But it is siloed, poorly integrated, and rarely reconciled against the original commercial intent. A bank might have six different systems that each hold a piece of the answer. None of them holds all of it.
The consequences extend beyond P&L. Consider the client experience. A corporate treasurer who discovers she has been overbilled faces an adversarial audit process to prove it. The bank, lacking a clear end-to-end view of her pricing terms, cannot quickly confirm or deny the discrepancy. Trust erodes. Relationships that took years to build can fray over invoices that, from the bank’s perspective, look correct because they reflect what the system was told, even if not what was agreed.
Underbilling carries its own risks. A bank that consistently fails to capture agreed fees is not just losing revenue; it is subsidizing relationships at a cost it cannot measure. Relationship managers, incentivized by revenue, may report numbers that do not reflect economic reality. Profitability models built on faulty billing data lead to faulty decisions about where to allocate capital and which clients to pursue.
From Detection to Prevention
The shift that matters is not from analogue to digital—most banks made that transition long ago, but from reactive to proactive. Revenue integrity cannot be treated as a back-office reconciliation task, addressed monthly when exceptions pile up. It needs to be embedded in execution, at the point where commercial intent becomes operational instruction.
This requires, above all, a single authoritative source of truth for product and pricing. Not a master data warehouse to which systems periodically synchronize—synchronization is another word for delay, but a live, integrated layer through which pricing logic flows in real time. When a deal is agreed, the terms should be available immediately to every system that needs them. When an exception is approved, it should be visible, auditable, and time bound. When a product changes, the change should propagate consistently, not drift through a succession of manual updates.
The technology to do this exists. What has been lacking is the organizational will to treat revenue integrity as a strategic priority rather than an operational inconvenience. That calculation is changing. In a low-growth environment, with interest margins compressed and fee income under regulatory scrutiny, protecting earned revenue is no longer a housekeeping function. It is a competitive imperative.
A bank that can demonstrate to its corporate clients a transparent, verifiable billing process—where every fee can be traced to an agreed term and every waiver is explicit and documented—possesses something increasingly rare: operational credibility. In an era when trust in financial institutions is hard-won and easily lost, that is not a trivial advantage.
Precision, Not Rigidity
None of this implies that relationship-based pricing should be abandoned. Flexibility remains a legitimate competitive tool. But flexibility and discipline are not opposites; they are complements. A bank that can offer bespoke terms to its best clients, and execute those terms with absolute fidelity, is more commercially attractive than one that offers bespoke terms it cannot reliably implement.
The European treasurer quoted at the start of this article eventually found her answer: nearly 4% of her bank’s corporate fee revenue was not collected as agreed. Not stolen, not written off, simply lost in the machinery between agreement and invoice. The fix took eighteen months and touched seven systems. The revenue recovered, compounded over three years, was multiples of the remediation cost.
Her conclusion was characteristically direct: “We spent years trying to grow our way out of margin pressure. We should have started by making sure we were actually collecting what we earned.”
In banking, as in most industries, growth is seductive. It promises escape from the hard work of operational discipline. But in a world where margins are thin and competition is fierce, the unglamorous task of ensuring that earned revenue reaches the balance sheet may be the highest return investment a bank can make.
Revenue leakage is not a new problem. What is new is that banks no longer have the luxury of ignoring it.
