How FDD changes for banks, lenders and fintechs: net revenue vs gross, take rate, loan-book provisioning, float income, regulatory capital and KYC/AML risk.
Most of what you learn about financial due diligence assumes a business that sells something, collects cash and books a margin. Financial services breaks that assumption. A lender's revenue is spread across a loan book that may not repay; a payments business earns a sliver of every transaction it touches; a fintech may report enormous "gross" volume that has almost nothing to do with what it actually keeps. Point a standard FDD playbook at one of these and you will size the wrong EBITDA, miss the real risks and get taken apart in the interview. This is a sector where the accounting conventions, the revenue definition and even the meaning of "capital" are different - and where the analyst who understands why wins the mandate.
In an industrial or software business, the balance sheet supports the P&L. In financial services, the balance sheet often is the business. A bank or a lender makes money by holding assets (loans) funded by liabilities (deposits, wholesale funding) and earning the spread. A payments or wallet business holds customer money it does not own. An insurer collects premiums today against claims it will pay years later.
That has three consequences for diligence:
Takeaway: In financial services, your first job is not to normalise EBITDA - it is to establish what the revenue is. Get the numerator wrong and every downstream analysis inherits the error.
This builds on the fundamentals covered in Quality of Earnings 101 and what QoE actually is, but the definitions shift under your feet in this sector.
The most common - and most expensive - mistake in fintech FDD is confusing gross transaction value (GTV) or total payment volume (TPV) with revenue. A payments processor might route €10bn of volume a year and keep 0.3% of it. The €10bn is not revenue; the roughly €30m of net take is.
The distinction that matters is between the gross figure the business handles and the net revenue it retains after paying away the costs directly tied to earning it - interchange fees paid to card schemes, partner bank costs, referral rebates, and payment-network assessments. Net revenue (sometimes called net take or gross profit in these businesses) is the number a buyer values.
Consider a simplified payments target:
| Line | Amount (€m) | Note |
|---|---|---|
| Total payment volume (TPV) | 8,000 | Not revenue - flow handled |
| Gross revenue (fees charged) | 96 | 1.20% blended fee on TPV |
| Less: interchange & scheme fees | (58) | Paid to card networks |
| Less: partner bank / sponsor costs | (11) | Cost of accessing the rails |
| Net revenue (net take) | 27 | The real top line |
| Take rate on TPV | 0.34% | Net revenue ÷ TPV |
A buyer who values the business off the €96m "revenue" line - or worse, off TPV - will overpay dramatically. The take rate (net revenue as a percentage of volume) is the metric that tells you whether the economics are healthy and, crucially, whether they are stable. A take rate that is drifting down usually means the merchant mix is shifting toward large, low-margin enterprise accounts, or that competition is compressing pricing. This is exactly the kind of revenue quality question the sector rewards.
For any business that holds balances - a challenger bank, a lender, an e-money institution, a broker holding client cash - a meaningful slice of income can come from net interest margin (NIM) or float income: interest earned on customer or client balances the business holds but does not own.
Two diligence points follow:
For a lender, the equivalent analysis is the net interest margin: interest and fee income on the loan book, less funding cost and less expected credit losses. Which brings us to the balance sheet risk that dwarfs everything else.
If the target lends money, the loan book is the business, and its quality determines whether reported profit is real. The core risk is that management is under-provisioning - booking too little expected credit loss (ECL) - which flatters current earnings by borrowing from future write-offs.
Your loan-book diligence should cover:
Here is why provisioning is an earnings adjustment, not a footnote:
| Scenario | New lending (€m) | Provision booked | Reported profit (€m) |
|---|---|---|---|
| Adequate provisioning | 500 | 6.0% (€30m) | 40 |
| Under-provisioned | 500 | 3.0% (€15m) | 55 |
The under-provisioned book reports €15m more profit - not because it is more profitable, but because it has deferred recognising losses that will still arrive. A buyer paying a multiple of €55m is paying for earnings that don't exist. Normalising the provision to a defensible loss rate is one of the most important EBITDA adjustments you will make in a lending deal.
In most sectors, free cash flow can be swept up to the buyer. In regulated financial services, it cannot - the business must hold regulatory capital (and, for banks, liquidity buffers) against its risk-weighted assets. Growing the loan book consumes capital. That has a direct valuation consequence: some of the cash the business generates is trapped supporting the balance sheet rather than available for distribution.
Diligence needs to establish:
This reframes the equity bridge. Beyond ordinary net debt items, a regulated financial institution may need a minimum capital carve-out - cash that looks surplus but is legally required to stay in the business. Treating trapped regulatory capital as distributable surplus is a classic way to overstate the equity value.
Financial services businesses distort the usual net working capital analysis because customer balances, settlement timing and deferred revenue behave unlike trade receivables and payables.
Red flag: If a business is holding customer money on its balance sheet as though it were its own, question everything else in the accounts. The client-money boundary is a test of whether management understands its own regulatory position.
Financial services carries a category of risk that never appears as a P&L line until it appears as a catastrophe: compliance risk. Weak know-your-customer (KYC) and anti-money-laundering (AML) controls, mis-sold products, or an unlicensed activity can trigger fines, forced remediation, customer redress or the loss of a licence - any of which can erase the value of the deal.
FDD (working alongside legal and regulatory advisers) should probe:
These rarely produce a clean number, but they belong prominently in the FDD report structure as risks the buyer must price or protect against in the SPA.
Many fintechs grow users faster than revenue, so the diligence question becomes: does each customer actually make money, and when? Unit economics - contribution per customer after direct servicing and funding costs, set against acquisition cost - tell you whether growth is value-creating or value-destroying. A business acquiring customers at a cost it will take four years to recover, in a market where churn runs high, is buying volume, not profit. Distinguish genuinely recurring revenue from transactional flow, and apply the same SaaS-metrics discipline where the fintech sells software alongside its financial product.
Expect a question that tests whether you can separate flow from economics. Something like: "A fintech tells you it did €5bn of payment volume and grew 60% last year. What do you want to know?"
A strong answer:
"My first instinct is that €5bn of volume tells me almost nothing about the value of the business - I need to get from gross flow to net revenue. So I'd want the blended take rate: how much of that €5bn does the company actually keep after interchange and scheme fees and any partner-bank costs? Then I'd want to know whether the take rate is stable or compressing, because 60% volume growth achieved by onboarding large, low-margin enterprise merchants can mean net revenue grew far more slowly. After that I'd look at where the income actually comes from - if a big chunk is float income earned on held balances, I'd stress it against a lower interest-rate assumption, because that's a rate-cycle windfall, not durable revenue. I'd also check the client-money position to make sure held balances aren't being treated as the company's own cash. And on the risk side, I'd flag the regulatory perimeter and KYC/AML controls early, because a licence problem or a redress liability can outweigh anything in the P&L. In short: get to net revenue, test its durability, and price the regulatory risk separately."
That answer shows you understand the sector's economics rather than reciting a generic checklist.
Financial services is the sector where the accounts are most likely to look like a normal business and least likely to behave like one. The numbers that matter - net revenue, take rate, loss rates, capital headroom - sit behind the headline figures a founder will happily quote you, and the risks that can sink the deal often never touch the income statement at all. Master the translation from gross to net, learn to read a loan book, and respect the regulatory constraint on cash, and you will bring something to a fintech deal that a generalist analyst simply cannot. In this sector, the balance sheet is the story.
The Transaction Services Interview Programme (€119.99, one-time) includes a dedicated financial-services and fintech module covering net revenue vs gross, take-rate analysis, loan-book provisioning and regulatory capital in the equity bridge. Enrol today.
Hundreds of candidates prepared their interviews with this programme. Those who landed the role have one thing in common: they worked the cases before walking into the room.
