# Point of Sale Financing Underwriting at Checkout

- Author: TrueBankFlow (https://truebankflow.com)
- Published: 2026-10-17
- Category: Personal and consumer loans
- Canonical: https://truebankflow.com/blog/point-of-sale-financing-underwriting

> How point of sale financing underwriting decides in seconds at checkout: soft pulls, income estimates, account data and when bank statements are requested.
Point of sale financing underwriting decides on a loan while the customer is at a checkout. The lender runs a soft credit pull, weighs the stated income against the purchase price, and scores the merchant and ticket size. Most answers come in seconds. Larger or borderline applications are routed to a bank statement or account data review.

## What does point of sale financing underwriting check at checkout?

Point of sale financing underwriting checks three things in the seconds before a checkout decision: who the applicant is, what their credit file says, and whether the purchase amount fits what they report earning. Identity comes first, because a fraudulent application at a merchant counter costs the lender the full ticket.

A typical instant decision draws on these inputs:

- **Identity data**: name, date of birth, address, phone and the last digits of a Social Security number, matched against identity databases.
- **The credit file**: score, open accounts, recent inquiries, delinquencies and public records, usually through a soft pull that does not affect the score.
- **Stated income**: a single number the applicant types into the form, often gross annual pay.
- **The purchase**: the merchant, the item category and the amount being financed.
- **Device and session signals**: whether the phone, email and location look consistent with the applicant.

The lender's model weighs these together and returns one of three answers. An approval comes with a term and an amount. A decline ends the session. A third answer, a pending or "needs more information" result, is where documents enter the process. That middle answer is small in volume and large in risk, and it is where most of the actual underwriting work happens.

Checkout lending runs on speed because the merchant loses the sale if the customer waits. Every extra screen costs approvals. That pressure explains why lenders push as much of the decision as possible into data they can pull without asking the customer for anything.

## Soft credit pulls and stated income in checkout lending

A soft credit pull lets a checkout lender read the applicant's credit file without leaving a hard inquiry on it. Stated income is the figure the applicant types in, and on its own it is unverified. At checkout the soft pull carries most of the weight, and stated income acts as a ceiling on what the model will approve.

Lenders treat stated income cautiously for good reasons. People round up. Some report household income when the form asks for personal income. Some type monthly pay into an annual field. A model that trusts the number as entered will approve payments the applicant cannot carry.

Several checks keep stated income honest without slowing the checkout:

1. **Income estimation models**: some credit bureaus and data vendors offer an estimated income from the credit file. A stated figure far above the estimate is a flag.
2. **Payment burden from the file**: the minimum payments on reported tradelines show how much the applicant already owes each month.
3. **Plausibility rules**: a stated income of $250,000 with a thin file and two recent collections does not fit together.

When the stated figure and the other data disagree, the model can reduce the approved amount instead of declining. A customer asking for $4,800 might be offered $2,500 and told to pay the rest another way. That outcome often suits both sides better than a document request at the counter.

For a lender that wants to see the full sequence behind a consumer credit decision, the guide to [personal loan underwriting from application to offer](/blog/personal-loan-underwriting) walks through the steps a slower, non-checkout file follows.

## Merchant category and ticket size in point of sale loan decisions

The merchant and the size of the ticket shape a point of sale loan decision as much as the applicant does. A $300 pair of headphones and a $9,000 elective surgery carry different loss patterns, different return and dispute rates, and different chances that the customer stops paying when something goes wrong with the purchase.

Lenders usually build merchant category into the model in a few ways:

- **Category loss history**: home furnishings, dental work, fitness equipment, auto repair and electronics each perform differently over time.
- **Merchant performance**: a single store with high dispute or refund rates can drag down approvals for every customer who shops there.
- **Ticket size bands**: a small ticket may get an instant answer on the soft pull alone, while a large one triggers more checks.

Small tickets often use a short pay-in-four structure. The [CFPB study on buy now pay later lending](https://www.consumerfinance.gov/about-us/newsroom/cfpb-study-details-the-rapid-growth-of-buy-now-pay-later-lending/) describes these products as letting a consumer buy an item and repay over four installments, with the first payment typically made at purchase, and notes that such loans generally run from $50 to $1,000. Larger tickets move into longer installment loans with monthly payments and interest, and those need firmer evidence of income.

Ticket size also changes what a mistake costs. Approving a $200 loan that goes bad loses $200. Approving a $9,000 loan on an inflated income figure can lose many times the margin earned on dozens of good loans. That arithmetic is why the bank statement request clusters at the high end.

## When does a point of sale lender ask for bank statements?

A point of sale lender asks for bank statements when the instant data cannot support the amount requested. The usual triggers are a large ticket, a stated income the model cannot confirm, a thin or new credit file, or a self-employed applicant whose pay does not show up in the usual income estimates.

Common routes into a document review include:

- **Ticket above the instant limit**: the lender sets a dollar amount above which it wants confirmed income.
- **Income mismatch**: stated income far above the estimated figure from the credit file.
- **Thin credit file**: too few accounts for the score to mean much, such as a young applicant or a recent arrival to the US.
- **Self-employment or gig income**: pay that arrives as transfers and platform payouts rather than payroll.
- **Fraud signals that need clearing**: a new phone number, an address change last month, or several applications in a day.

Lenders have two ways to get the evidence. The applicant can upload statement PDFs, or link the account so the lender reads transactions directly. Both have trade-offs at a checkout. Uploads take time and invite edited files. Account links are faster but depend on the applicant's bank being supported and the customer agreeing to connect. The guide to [open banking for lenders](/blog/open-banking-for-lenders) covers how account links work and what consent they require.

Some applicants cannot or will not produce statements at all. For those, the lender either approves a smaller amount, declines, or offers a different product. The guide to [loans without bank statements](/blog/loans-without-bank-statements) explains what other evidence lenders accept when statements are off the table.

## Reading a checkout applicant's bank statements

Reading a checkout applicant's bank statements means confirming three facts fast: that pay arrives regularly in roughly the amount stated, that the account is not already strained by overdrafts or other loan payments, and that the statements are genuine. The reviewer works under time pressure, because the customer may still be waiting at the merchant.

### A worked example

The figures below are an example of one fictional application. An applicant at a dental office requests $4,800 of financing for a treatment plan. She states an annual income of $62,000. Her soft pull shows a fair score, four open accounts and no delinquencies in two years. The lender's instant limit for this merchant category is $3,000, so the application goes to a statement review.

She links her checking account and the lender reads three months of transactions.

| Item (example) | Month 1 | Month 2 | Month 3 |
| --- | --- | --- | --- |
| Payroll deposits, two per month | $3,960 | $3,960 | $4,120 |
| Other deposits (transfers from savings) | $400 | $0 | $250 |
| Rent | ($1,450) | ($1,450) | ($1,450) |
| Auto loan payment | ($385) | ($385) | ($385) |
| Pay-later installments, two merchants | ($140) | ($210) | ($210) |
| Overdraft or NSF fees | 0 | 0 | 0 |
| Ending balance | $1,240 | $980 | $1,310 |

What the reviewer concludes from this example:

1. **Payroll supports the stated income.** Net pay of about $4,000 a month is consistent with a gross annual figure near $62,000 after taxes and deductions.
2. **The savings transfers are not income.** They come from her own account and are left out.
3. **Two pay-later plans are active.** Neither appears on the credit file. Together they add up to $210 a month.
4. **No overdrafts in three months**, and the balance never drops near zero.

A 36-month loan of $4,800 at an example APR near 20 percent costs about $178 a month. After rent, the car, the existing pay-later plans and the new payment, she still has room. The lender approves the full amount. Had the statements shown repeated overdrafts, the likely outcome would have been an approval at the $3,000 instant limit instead.

Statement reviews like this one take minutes when the transactions are already sorted. [TrueBankFlow](/) groups deposits by source, tags loan payments and flags NSF and overdraft events, which suits a review queue where the answer has to go back to the merchant the same day.

## Checkout loans a point of sale underwriter cannot see

Checkout loans that never reach the credit bureaus are the blind spot in point of sale financing underwriting. An applicant can carry several pay-later plans at once, and a soft pull shows none of them. The underwriter then approves a new payment on top of debts that the model does not know exist.

The CFPB study cited above reports that most buy now pay later lenders do not furnish data to the major credit bureaus, which means other lenders may not see a borrower's current pay-later balances when they decide on a new loan. Reporting practices are changing, but a lender cannot assume a file is complete.

Bank statements are the most direct way to see these plans. They show up as small, regular debits to the same handful of pay-later providers, often on a two-week cycle. A reviewer looking for them should:

- **Count distinct plans**, not just total dollars. Five small plans opened in one month say more about the applicant's situation than one plan at the same total.
- **Check the timing.** A cluster of new plans in the last 30 days points to a cash squeeze.
- **Add them to the payment burden.** Treat each active plan's installment as a monthly obligation until the statements show it ending.
- **Watch for failed installments.** A returned pay-later debit followed by a retry and a fee is a strong warning.

### When the instant decision should not stand

Some applications should not be approved on instant data even when the score looks fine. A large ticket at a merchant with a high dispute rate, a stated income that only works if every pay-later plan is ignored, or an applicant who opened three other checkout loans this week are all poor fits for a seconds-long answer. Sending those files to a short statement review costs some conversions. Approving them blind costs more.