Original research
A personal loan can leave you with less cash than you expect
Key finding
The amount in big type may be the loan balance, not the cash that reaches your account. Start with the cash that will reach your account. A 10% deducted fee turns a $10,000 cash need into an $11,111.11 loan. You repay the full $11,111.11, plus interest.
This is a 17-program sample, not the whole market. It cannot predict your offer. This working report is public. Publisher review remains pending before its final research release.
- The lowest advertised APR is not a typical rate. Most public records do not show what a typical borrower paid.
- Only 6 of 17 programs published a complete state map. Your state may change access, amount limits, or the lender.
- 9 of 17 programs used a different legal lender or one that can change by state, partner, or offer.
- Your offer is the final word. Check cash received, monthly payment, total repayment, loan length, and the legal lender.
The ad leaves out facts that can change your deal
A fee can leave you short of the cash you need. The lowest APR may not be close to the rate you get. Your state can change the rules. And the name on the website may not be the lender on your contract.
Those are not edge cases in this sample. Only 6 of 17 programs published a complete state map. Nine used a different or variable legal lender. Only 1 of 22 records about funded loans supported a limited comparison with a later posted rate.
The public page shows only part of the loan
A rate page is built to show what might be possible. It may not show the cash after fees, the price most borrowers paid, or the company named on the final note. State rules can change the answer again.
To see how often those gaps appear, we checked 17 programs with different business models: online platforms, finance companies, banks, and credit unions. This is not a ranking or a picture of the whole market.
We used official lender pages, disclosures, filings, and regulator records. When those sources did not settle a question, we left it unknown.
Questions about the research
What did Rung study?
We checked fees, loan amounts, rates, state availability, and the legal lender behind each brand. We also looked for public evidence about people who received loans.
Why did Rung choose these 17 programs?
They cover different ways personal loans are made: online platforms, finance companies, banks, and credit unions. The sample is built for contrast, not to rank the market or name a winner.
Where did the numbers come from?
They come from 58 public sources, including lender pages, disclosures, filings, rate tables, and regulator records. The snapshot date is September 5, 2026.
Can this study tell me what rate or fee I will get?
No. It shows what public records say about these programs. Your current offer and loan disclosure control your rate, fee, payment, and cash received.
Why could only 1 funded rate be compared with an ad?
The 22 records cover different borrowers, terms, dates, and funding channels. Only 1 lined up closely enough with a posted rate to support a limited comparison.
What does “not confirmed” mean?
It means the public sources did not settle the question. It does not mean no, unavailable, or zero.
Can I download the data and check the work?
Yes. The data page links to the records, sources, method, and complete ZIP. CC BY 4.0 covers Rung-owned work. Third-party source clearance is still pending, so the full package is not cleared for redistribution.
Has the research been independently reviewed?
No independent expert review is claimed. Colin McCracken is assigned to review the math, source records, editing, and release scope as publisher. That review is pending.
Read the short report or check the numbers yourself
These publicly accessible working files are ready to check, but not ready for final release. Publisher review and third-party source clearance remain open.
This working report still needs publisher review
Loan versus card APR benchmarks
Added September 9, 2026. This appendix supports the large-purchase calculator. It is separate from the 17-program Collection 3 study. It adds no funded-loan observations.
The two sources use different score bins and dates. Their figures support example costs, not a claim that one product costs less for a person with a given score.
Download the benchmark data, sources, and method (JSON). Independent review is pending.
See how we got the numbers
Compare the cash you get, not only the amount you borrow.
Then compare the monthly payment, total dollars repaid, loan length, and who makes the loan. If the math does not improve your situation, walk away.