Why 60% of Online Lender Borrowers Paid More Than They Expected
The Federal Reserve asked 6,525 small employers what their financing actually cost. Online lender borrowers were nearly twice as likely as bank borrowers to say it cost more than they expected.
Every year the 12 Federal Reserve Banks ask thousands of small employers a question most lenders never ask: after you took the money, did it cost what you thought it would?
In the 2025 survey, published in March 2026, the answer split sharply by where the money came from.
Sixty percent of firms that borrowed from online lenders said their actual borrowing costs were higher than expected. Among firms that borrowed from small banks, 37% said the same. At large banks, 32%. Only 4% of online lender borrowers found costs lower than expected.
That gap is the single most useful number in the report, and it is worth understanding precisely — because the obvious explanation is the wrong one.
The obvious explanation is incomplete
The easy read is that online lenders are simply more expensive. That is often true, but it does not explain the survey result.
Being expensive and being surprising are different problems. A borrower who knows they are paying a high rate and accepts it has made a decision. A borrower who is surprised did not have the information to make one. The survey is measuring the second thing: the distance between what people expected and what they got.
If online credit were merely costly, borrowers would report high costs but accurate expectations. Instead they report a mismatch — which points at how the price is communicated, not just how high it is.
Three products, three languages, one confused buyer
Federal Reserve researchers have studied this directly. In a March 2026 presentation summarizing that work, the Cleveland Fed illustrated the problem with a single scenario: a business borrows $50,000 and repays it over six months.
| Product | How the price is described | Total amount owed | Estimated APR equivalent |
|---|---|---|---|
| Product A | Factor rate 1.15 | $59,000 | ~70% |
| Product B | Fee rate 4% | $56,500 | ~45% |
| Product C | Simple interest 9% | $54,500 | ~46% |
Three offers. Three different vocabularies. The cheapest and most expensive differ by $4,500 on the same $50,000 — and nothing in the way the prices are stated makes that obvious.
Notice that Product A's "1.15" is the smallest-looking number on the page and the most expensive deal in the table. Notice that Product C's 9% looks roughly six times worse than Product B's 4%, and is actually about the same cost. A number that sounds small is not the same as a cost that is small.
The Fed's earlier focus group work found exactly what you would predict: owners initially said comparing these products was "easy," then answered comparison questions incorrectly — most often on cost. Almost all of them said they wanted costs disclosed in one standard way so that comparison was possible at all.
That is the mechanism behind the 60%. It is not that borrowers are careless. It is that three products quoted in three incompatible units cannot be compared without conversion, and the conversion is not offered.
Where firms are actually applying
The surprise number matters more each year, because more firms are going where the surprises are.
The share of applicants seeking financing from online fintech lenders rose from 17% in the 2020 survey to 29% in the 2025 survey — an increase in each of the last five survey years. Online lenders are now a close third behind large banks and roughly level with small banks as a place small firms apply.
Firms are not doing this irrationally. Survey respondents cited speed of decision and a better perceived chance of approval as reasons for going online. Those are real advantages when payroll is in nine days.
The tradeoff shows up on the other side of the transaction. Applicants at small banks were the most likely to be fully approved, at 57%. And when the Fed asked which borrowers ran into problems with their lender, online lender applicants reported the most — with high interest rates and unfavorable repayment terms as the two most common complaints. Bank and credit union applicants reported higher satisfaction than online lender and finance company applicants.
The broader funding picture
A few other figures from the same survey are worth holding alongside that:
| Measure | 2025 survey result |
|---|---|
| Firms that applied for financing | 60% |
| Applicants who received the full amount sought | 42% |
| Applicants who received some or most | 36% |
| Applicants who received none | 22% |
| Firms using financing on a regular basis | 86% |
| Firms with no outstanding debt | 31% |
Over one in five applicants walked away with nothing, and only a little over four in ten got everything they asked for. The most common reasons for seeking financing were ordinary ones: meeting operating expenses (56%) and pursuing an expansion or new opportunity (46%).
Of the firms carrying debt, 59% secured it with a personal guarantee. That detail deserves more attention than it usually gets — it means the majority of small business debt in this sample is not really separate from the owner.
What to do with this before you sign
The survey does not tell you which lender to use. It does tell you which question to ask, and when.
Convert every offer to the same two numbers. Total dollars repaid, and an APR equivalent. Not the factor rate. Not the fee rate. Not the monthly payment in isolation. If two offers cannot be expressed in the same units, they have not been compared.
Ask for the total repayment amount in writing before you accept. A lender who cannot produce a single number for what you will pay back has told you something useful.
Price the term, not just the rate. The $50,000 example above is a six-month product. Short terms make even modest-sounding rates land hard on monthly cash flow, which is where most of the damage actually happens.
Treat speed as a cost. Faster decisions are worth real money when timing is genuinely the constraint. They are worth much less when the deadline is self-imposed. Decide which situation you are in before the urgency does it for you.
Check whether a personal guarantee is attached. Given how common it is, assume yes until the document says otherwise.
What this data does not say
Two honest limits are worth stating, because they are stated in the report itself.
The SBCS is a convenience sample, not a random one. The Federal Reserve is explicit that results should be read with awareness of the biases that come with that. It describes the experience of the 6,525 firms that responded — it is not a precise national estimate.
And "higher than expected" is a self-reported perception, not an audit of loan documents. It measures the gap between expectation and outcome, which is exactly the point, but it is not a direct measurement of who charged what.
Neither limit undermines the finding. A 60%-versus-32% gap across thousands of responses is a large, consistent signal about how clearly different parts of the market explain their own prices.
The takeaway
The most expensive thing in a financing offer is frequently not the rate. It is the unit the rate is quoted in.
Trovo Capital is not a lender and does not make credit decisions. We help business owners understand what they are being offered before they accept it — which, on this evidence, is where a substantial share of the cost is actually decided.
Source: "2026 Report on Employer Firms: Findings from the 2025 Small Business Credit Survey." 2026. Small Business Credit Survey. Federal Reserve Banks. https://doi.org/10.55350/sbcs-20260303. The 2025 SBCS was fielded September 3 to November 14, 2025 and yielded 6,525 responses from firms with 1–499 employees across all 50 states and the District of Columbia. The $50,000 product comparison is drawn from a March 2026 Federal Reserve Bank of Cleveland presentation on small business borrower experiences.
Original analysis, written by operators who work with founders every week.
Approved by Trovo Capital Team
vol. 1 · no. 20
