Sixty Percent of Your Borrowers Were Surprised by the Cost. At Banks It Is Thirty-Two.

Two channels serve overlapping borrowers and produce a 28 point difference in whether the cost landed as expected. Only one of those channels chose that outcome deliberately.

Sixty percent of small business owners who borrowed from an online lender reported that their actual borrowing costs came in higher than they had expected. Among those who borrowed from large banks, the figure is 32 percent. Applicants at banks and credit unions also report more satisfaction with the experience than applicants at online lenders, and the two complaints raised most often about online lenders are high rates and unfavourable repayment terms.

The obvious reading is that non-bank credit costs more, so of course more borrowers are surprised. That reading does not survive contact with the number.

A price gap would not produce this shape

If the surprise were simply a function of price, the borrowers most surprised would be the ones paying most, and the effect would show up as a smooth gradient across cost levels. What the data shows instead is a step between two channels.

More usefully: a borrower who knows they are taking expensive money is not surprised by expensive money. Merchants approaching a non-bank funder generally understand they are paying more than a bank would charge. They have often already been declined by one. The surprise is not that the number is high. It is that the number they ended up with is not the number they had in their head when they agreed.

That is a description of when information arrived, not of how much was charged.

Why the bank figure is lower, and why that matters

Large banks are not better at candour. They operate inside a disclosure regime that forces a standard set of figures out at a defined point, in a defined format, before the borrower commits. The process produces the outcome whether or not anyone at the bank intends it.

There is a second reason, and it is structural rather than cultural. A term loan resolves to a small number of figures a borrower already has a mental model for: an amount, a rate, a monthly payment, a term. Non-bank structures often do not. A factor rate, a daily or weekly holdback, an origination fee deducted at funding and a repayment period that moves with receipts do not resolve into a single number the borrower can carry around, so unless somebody does that arithmetic for them and hands it over, they will do it themselves after the money has landed. Most of them will do it wrong, and they will do it late.

That is the most useful thing in this comparison. Thirty-two percent is not a natural constant. It is what happens when total cost, fee structure and repayment cadence reach the borrower early and in a form they can compare. A non-bank funder operating under lighter requirements can adopt the same practice by choosing to, and the evidence that the outcome is reachable is that a comparable population already lives there.

The cost of the gap lands in servicing, not just in renewals

A borrower who is surprised at the first payment behaves differently for the rest of the relationship, and none of it is good for the funder.

They query the balance. They dispute a fee that was disclosed but never registered. They pay on the slower end of the acceptable range because they are managing a cash position they had planned differently. When performance slips, the workout conversation starts from a position of grievance rather than partnership, which lengthens it and lowers what gets recovered.

Every one of those is a servicing cost, and every one is absorbed by people rather than by systems. A funder carrying a materially higher rate of surprised borrowers is carrying a materially higher volume of avoidable contact, and it does not appear as a line item anywhere. It appears as collections headcount.

Run that forward across the channel and it compounds, because the channel with the gap is the one growing. The share of small business applicants seeking financing from an online or fintech lender has climbed to 29 percent from 17 percent across five survey rounds. More of the market is arriving in the channel that produces the worse expectation outcome.

How CXO Solves This

The fix is a timing and format change, and it is unglamorous.

Client Onboarding Automation generates the full cost picture, total repayment amount, fee structure and payment cadence, at the moment an offer is produced rather than when documents are prepared, delivers it, and records when it went out. That timestamp matters as much as the disclosure: it converts a claim about what the borrower was told into a fact the firm can check.

Collections and AR Automation then sequences follow-up against a borrower who has already seen the numbers, which changes what the first difficult conversation is about. Disputing a fee you were shown at offer is a different conversation from disputing one you first noticed on a statement.

Neither of these changes what a funder charges. They change how many borrowers reach the first payment holding the same number the funder is holding, and 32 percent is the evidence of how far that can move.

In most operations, far more work can be automated than leadership realizes. One discovery call is enough to size what automating it would return to your bottom line. Book it at https://cxocorporation.com/contact.