When you buy a car on credit at a dealership, the dealer sells your loan contract to a bank, credit union or finance company. Contract funding is the lender checking that the signed contract is the deal it approved, then paying the dealer.
This guide is for anyone who funds, oversees or pays for that work: indirect lending managers, dealer funding teams and funding analysts, dealer relations, credit, compliance, and the leaders deciding whether to automate it. No background needed.
Imagine you are a funding analyst at a credit union. On Saturday, Andre Polk buys a used 2023 Honda Pilot at a dealer you work with, and signs the contract. On Monday morning his deal jacket is one of 64 waiting for you.
Most of it matches the approval. But look at the mileage: the contract says 21,406, and the odometer statement he signed says 24,106. The GAP price is over your cap, too.
The contract · what was signed
Buyer
Andre Polk
Vehicle
Honda Pilot
GAP
$1,195
Mileage
21,406
The approval and the file
Buyer
Andre Polk
Vehicle
Honda Pilot
GAP cap
$995
Odometer form
24,106
3 fixespended 9:12 a.m.mileage, GAP price, and an insurance card with no lienholder.
Swapped digits change the book value, and the book value sets how much you can lend. If nobody checks, you buy a contract that is not the deal you approved. If the pend is vague, the dealer calls, resubmits, and waits days for the money.
With one clear pend, the dealer re-signed by 10:40 a.m. The deal was funded at 10:44 a.m.
The contract, the funding review, the money in transit and the rules, explained without selling anything. If you are new to indirect auto lending, start here.
01 · In plain words
Indirect auto contract funding in five minutes
Most people who finance a car never talk to the lender. They sit with the dealer's finance manager, who sends their application to several lenders and picks an approval. The buyer signs a retail installment contract with the dealer. The dealer then sells that contract to the lender, and the lender pays the dealer. This is indirect auto lending, and the Congressional Research Service calls it "the vast majority" of US auto loans.
The approval takes minutes. The money takes days. In between, the dealer sends the lender a deal jacket: the signed contract, the buyer's order, proof of income and insurance, the forms for add-ons such as GAP, the odometer statement. That is often 15 to 25 documents.
Contract funding means checking the deal jacket against the approval and the lender's program, then paying the dealer. If something is wrong, the deal is pended: sent back to the dealer to fix. Until the lender pays, the dealer has handed over a car and not been paid for it.
The life of one deal, in five steps
The buyer applies at the dealerAndre Polk picks a used SUV on Saturday. The dealer sends his application to lenders.
A lender approvesThe credit union approves an amount, a rate, a term, and conditions called stips.
The buyer signs and drives awayThe dealer writes the contract, adds products like GAP, and sends the deal jacket.
The funder checks the jacketEvery term against the approval and the program. This is where the swapped mileage is caught, or missed.
Fund, or pendClean deals are paid in the next wire. Others go back to the dealer, who waits for the money.
Step 4 is the subject of this guide. Done slowly, the dealer waits. Done loosely, the lender buys a contract it never approved.
Four things that surprise people
Approval is instant; funding is not
"Many lenders advertise 'instant approvals,' but still take three to five business days to fund," one lender wrote in 2026.
Dealers choose lenders on experience, not rate
In J.D. Power's 2026 dealer study, a competitive rate was the top reason for 18% of dealers. Ease, approval speed and relationships made up 70%.
A vague pend costs a second call
Dealer satisfaction was 841 when an issue was solved in one contact, and 599 when it took a second (J.D. Power, 2026).
Regulators now look at the lender
A September 2026 settlement with 40 attorneys general over add-ons packed into contracts required the lender to monitor its dealers.
Four words you will see on this page
Deal jacket
Everything the dealer sends for funding: the contract and its supporting documents.
Stips
Stipulations: conditions on the approval, such as proof of income or insurance, cleared before funding.
Pend
A deal sent back to the dealer to fix before the lender pays.
Contracts in transit
The dealer's money for cars already delivered on credit but not yet funded.
The dealer sells the car on credit. The lender buys the contract.
In shortThe dealer is the creditor: it sells the car on a retail installment contract, then sells that contract to a lender. The contract carries the price, the add-ons, the rate and the federal disclosures, and the dealer is paid for arranging the credit.
In indirect auto lending the buyer never sits with the lender. The dealer is the creditor: it sells the car on a retail installment contract and assigns that contract to a bank, credit union or finance company that agreed, minutes earlier, to buy it on stated terms. Funding is the moment the lender checks that the signed contract is the deal it approved, and pays.
Indirect vs direct auto lending
Indirect · through the dealer
The dealer arranges the credit
The F&I manager sends the application to one or more lenders through a dealer platform, picks an approval, structures the deal, and has the buyer sign a retail installment contract (RIC). The lender buys the contract by assignment and pays the dealer. Most US auto loans start this way.
Direct · to the lender
The buyer borrows, then shops
The buyer applies to the bank or credit union, gets a loan or a pre-approval, and pays the dealer like a cash buyer. The lender owns the relationship from day one, and there is no dealer reserve and no deal jacket to fund.
Reviewing the retail installment contract (RIC): what is inside
cash price + doc fee + taxes + title fees + add-ons − cash down − net trade = amount financed
The RIC is the dealer's credit contract with the buyer. It carries the federal Truth in Lending box (annual percentage rate, finance charge, amount financed, total of payments and total sale price), the payment schedule, the itemization above, the optional products, the insurance terms, the FTC Holder notice, and the signatures. When it is assigned, the lender becomes the holder.
Taxes and fees follow state rules (for example, whether a documentary fee is taxable, or how a trade-in credit applies) check your states.
Dealer reserve and dealer participation: how the dealer is paid for the credit
Illustrative example · $30,000 over 72 months
Swipe sideways to see the full diagram.
The buy rate is what the lender will buy the contract at; the dealer may mark it up to the lender's cap, and the dealer agreement says how much of the spread the dealer earns, whether it is paid up front, discounted or capped, and when it is charged back on an early payoff or default. Some programs pay a flat fee instead of a markup. Arithmetic is simplified for illustration: use your dealer agreement.
Add-on products
GAP, service contracts and the back end
Optional products sold in F&I and financed in the contract. Lenders cap each product's price and the total "back end" (often as a share of book value), and require signed forms that show price and cancellation terms. Cox Automotive reports that over 61% of contracts include at least one aftermarket product, which could require one of 250,000 different forms (vendor data, January 2026).
Stipulations
The conditions on the approval
Proof of income, proof of residence, proof of insurance naming the lender as lienholder or loss payee, a valid driver's license, sometimes references and payoff letters. A "stip" is cleared by a document that names the buyer and meets the freshness and tolerance rules of the program.
03 · Funding and contracts in transit
Nine steps, six parties, and the dealer's money out the door until the wire lands.
In shortApproval takes minutes; funding takes days. Until the lender pays, the dealer has delivered a car and may already have paid its floor plan, the trade-in payoff and commissions. The calculator shows what that wait costs your dealers.
Approval takes minutes. What takes days is everything after the buyer drives away: the jacket travels, the funder checks it, the dealer fixes what's wrong, and the wire waits for its cut-off. The blue row is the lender's funding team.
Who ↓ · Step →
1Application
2Decision
3Structure and sign
4Submit the jacket
5Funding review
6Pend and fix
7Fund and wire
8Title and lien
9Book and audit
Buyer
Applies at the dealership
Signs the RIC and forms; takes the car
May re-sign
Starts paying the lender
Dealer F&I
Keys the application
Picks an approval
Sets rate, adds products
Sends the deal jacket
Waits
Fixes and resubmits
Sees "Funded"
Files the title application
Reserve; chargebacks
Dealer platform
Routes to lenders
Returns decisions
eContract and e-signature
eVault transfer
Carries the pend
Funding status
Lender credit
Approves, counters or declines; sets stips
Off-approval deals come back
Decides waivers
Lender funding
Checks contract, math, add-ons, stips
Writes the pend; re-checks
Funds; builds the batch
Tracks the lien
Books; post-funding QC
Treasury · state
Wire or ACH by cut-off
Records the lien (ELT or paper)
In transit $
The dealer has delivered the car and may have paid the floor plan, the trade payoff and commissions →until funding
Swipe sideways to see all nine steps.
What waits on whom
Dealer → lender
~5 daysPaper contract packages averaged five days in transit before eContracting (Dealertrack data, 2019).
Lender review
2–4 daysTypical funding cycle at bank indirect programs; 24–48 h is strong, 5+ days a warning (DefiSolutions, vendor, 2026).
Lender → dealer
One contactDealer satisfaction is 841 when an issue is solved on the spot and 599 when it needs a second contact (J.D. Power 2026).
Funding → cash
1–2 daysACH typically settles in one to two business days after funding (Lendbuzz, vendor).
Weekend → Monday
BacklogWeekends carry a large share of retail sales, while many lenders fund on business days (Lendbuzz, vendor).
Contracts in transit: why the dealer counts every day
The dealer's view
Contracts in transit (CIT) is the dealer's money for cars already delivered. NADA's cash-management guidance puts it plainly: "If you receive a payment of a contract in transit before you have to pay the floor plan, lien payoff on the trade-in, and sales commission—then your cash flow is OK." It recommends "a daily report of all contracts in transit."
Every day of delay adds floor-plan or working-capital interest, and a slow or confusing pend is the delay a dealer remembers when choosing where to send the next deal.
The lender's view
Dealers choose lenders on the deal experience. In J.D. Power's 2026 dealer financing study, a competitive rate was the top reason for 18% of dealers, while ease, approval speed and relationships made up 70% of the reasons.
An older J.D. Power study (2010) found satisfaction rose by as much as 64 points when lenders funded error-free contracts the same day.
Price a day in transit
Defaults are illustrative, not benchmarks. For scale, Experian puts the average amount financed in Q2 2026 at $43,610 new and $27,852 used.
Your dealers' money in transit
17contracts delivered a day, on average
$1.5Mowed to your dealers at any moment
$10,000dealers' carrying cost a month
$20carrying cost per contract
In transit = contracts a day × days × amount. Carrying cost = in transit × rate ÷ 12. Small per deal; large across a dealer's month, and felt every time a pend adds days.
04 · Why it matters now
A large, dealer-led market where speed wins the next deal, and the rules keep moving.
In shortAuto lending is huge, most of it starts at the dealer, and dealers send deals to the lenders that fund fast and clearly. Meanwhile regulators expect lenders to watch what dealers add to contracts.
Six numbers worth knowing, each with its source. Vendor figures are marked as such.
$1.71T
in US auto loan balances at the end of Q2 2026, up $28B in the quarter, with $211B of new auto loans made in the quarter.
New York Fed, Household Debt and Credit, Q2 2026 (Aug 11, 2026).
22.84%
of auto loans were held by credit unions in Q2 2026, behind banks at 29.50% and ahead of captives at 18.41%. Indirect was 55% of credit unions' vehicle balances in Q2 2024.
Experian, State of the Automotive Finance Market Q2 2026; Callahan & Associates (2024).
83%
of auto loans are obtained indirectly, "according to one paper"; CRS calls it "the vast majority".
Congressional Research Service, IF13179, Mar 10, 2026.
3–5 days
"Many lenders advertise 'instant approvals,' but still take three to five business days to fund."
Lendbuzz, Apr 9, 2026 (a lender that markets fast funding).
86%
of auto finance contracts are now eligible for digital submission, yet over 61% carry aftermarket products whose forms come in 250,000 variants.
Cox Automotive press release, Jan 23, 2026 (vendor data).
$10.4B
of auto lending fraud exposure in 2025, up from $9.2B; 69% of it was misrepresentation by applicants or dealers rather than stolen identities.
Point Predictive 2026 report, as reported by AFSA (Apr 9, 2026).
The pressure on funding teams
Dealer platforms push "auto-fundable" contracts and point-of-sale checks, so lenders that still review by hand fall behind on speed (Cox Automotive, 2026).
Regulators now look to the lender to police dealer add-on practices: the September 2026 multistate settlement required dealer monitoring (section 6).
Fair-lending exposure did not vanish with the 2026 Regulation B change: state laws may still reach disparate impact check your states.
A note on the numbers. Defect and cycle-time figures in this field mostly come from vendors and single institutions. Where a number describes one credit union, this guide says so. Published surveys put the share of defective deal jackets anywhere from under 10% to around 30%, depending on the lender: measure your own. Figures in the screens in Part B are fictional sample data.
05 · Inside the funding review
Deal jacket review: what a funder checks before the lender buys the contract
In shortA funder proves six things before paying: the jacket is complete, the contract is the deal credit approved, the math is right, the add-ons fit the program, every stip is cleared, and the deal can be funded, pended or sent to credit.
A funding analyst, also called a funder or contract processor, opens each deal jacket and proves that the signed contract is the deal credit approved, that the math and the disclosures are right, that the products fit the program, and that every stipulation is cleared. Six steps, then a decision.
1
The deal jacket arrives
An eContract arrives as structured data with its documents, and the authoritative copy sits in an eVault. A paper package is couriered or scanned, and the funder must also confirm the original arrived complete. Either way, the jacket holds 15 to 25 documents: RIC, buyer's order, application, approval, odometer statement, title application or MSO, add-on forms, stips, insurance and trade payoff.
Funding analystsContract processing
Dealertrack's 2019 data: one in four paper contracts contained an error (vendor, dated).
IneContract or paper package; stip documentsClockWeekend deliveries land on Monday; month-end spikes follow sales targetsWatch forMissing pages, unsigned forms, illegible scans
2
The contract is checked against the approval
Buyer and co-buyer names match the approval. Year, make, model, trim, new or used, and the VIN match across the RIC, buyer's order, title application, odometer statement and insurance. Amount financed is within the approved limit; term within the program; payment-to-income and loan-to-value within limits at the right mileage; the contract rate within buy rate plus the markup cap.
Funding analystsCredit (off-approval)
Anything off its approval (a higher amount, a different buyer or vehicle, income outside tolerance) is a credit decision, not a funding one.
ChecksNames, vehicle, VIN, amount, rate, term, payment, LTV, PTI, down payment, tradeWatch forA co-buyer on the approval missing from the contract; transposed odometer digits that move book value
3
The math and the Truth in Lending box are recomputed
The itemization must tie to the buyer's order, and the disclosures must recompute: APR within ⅛ of a percentage point in a regular transaction (12 CFR 1026.22), finance charge within ±$10 when the amount financed is over $1,000 (12 CFR 1026.18). The FTC Holder notice must be present (16 CFR 433.2).
Funding analystsCompliance
Why it matters to the buyer of the contract: under 15 U.S.C. 1641(a), an assignee is liable for violations apparent on the face of the disclosure statement.
RecomputeAPR, finance charge, amount financed, total of payments, payment scheduleWatch forState fee and tax rules; state form number and revision; a missing Holder notice
4
Add-ons and dealer reserve are held to the program
Each product's price is within its cap, the total back end within its limit, and each form signed, priced, with cancellation terms, on the right vehicle and term. Markup is within the cap and participation computed per the dealer agreement. A product must never be presented as required.
Funding analystsDealer relationsCompliance
Dealer-level add-on and markup statistics are the evidence of oversight that regulators and attorneys general now expect (section 6).
ChecksGAP, VSC and other product prices; total back end; markup; participation; product formsWatch forPrices over cap, packed products, flat-fee errors
5
Stipulations are cleared on documents that name the buyer
A pay stub within N days whose employer and income match within tolerance; a utility bill or statement at the application address; an insurance card or binder naming the lender as lienholder or loss payee (often "ISAOA/ATIMA") with deductibles within limit; a valid license that matches.
Funding analystsFraud review
In an Informed.IQ survey of auto finance leaders, the most-named deal-jacket defects were employer mismatch (26.3%), income mismatch (24.2%) and missing signatures (23.2%) (vendor survey, 2024).
InPay stub, utility bill, insurance card or binder, license, referencesWatch forStale stubs, insurance with no lienholder, income that doesn't verify
6
Fund, pend, send to credit, or send to fraud review
Fund when every term is within approval and program, the math ties and every stip is cleared. Pend when the dealer can fix it (a re-sign, a missing form, a new insurance binder), and say exactly what to fix and by when. Send to credit when the deal is off its approval. Send to fraud review when income, identity or the vehicle's book-out doesn't hold up. Then build the wire batch before the cut-off, and track the title until the lien is recorded.
Funding analystsSupervisorTreasury
OutFunded deal and booking data, or a pend noticeClockWire cut-offs (for example 2:00 p.m.) push a late deal to the next batch or the next dayWatch forChanged dealer bank details on the wire
Post-funding audit and QC
After funding, a quality-control or audit team samples files, checks the decision against the program, and reconciles dealer reserve. Defects that slip through become audit findings, repurchase or chargeback disputes, unperfected liens, add-on refund liabilities or fair-lending exposure. NCUA's indirect lending guidance (Letter 10-CU-15) expects credit unions to keep approval authority in house, hold written dealer agreements, and monitor performance dealer by dealer.
Where it hurts in dealer contract funding
Monday and month-end peaks
Weekend deliveries and sales targets arrive at once; the team that is sized for Wednesday drowns on Monday.
Costs: overtime, missed cut-offs, aging jackets
Re-proving clean deals
Most of the day goes to confirming that deals with nothing wrong have nothing wrong.
Costs: analyst time; slower clean funding
Pends the dealer can't act on
A vague pend becomes a phone call, then a second submission, then a second review.
Costs: second contacts; dealer satisfaction (841 vs 599)
Stip hunting
Pay stubs, bills and insurance cards checked by eye for names, dates, addresses and lienholders.
Costs: time per jacket; stale or mismatched stips slip through
Add-ons over cap
Product prices, total back end and form contents vary by dealer and administrator.
Costs: refunds, findings, regulatory exposure
Missed wire cut-offs
A deal that is ready at 10:35 misses the 10:30 cut-off and waits for the afternoon or the next day.
Costs: contracts in transit; dealer calls
Audit findings after funding
Defects found in post-funding QC are expensive to cure once the money has moved.
Costs: repurchases, chargebacks, remediation
Fraud at the last control
Income misrepresentation, straw buyers and powerbooking (inflated options on the book-out) are cheapest to catch before funding.
Costs: early payment defaults; losses
No view by dealer
Without pend reasons, add-on prices and funding time by dealer, dealer relations can't coach and compliance can't show oversight.
Costs: dealer churn; weak oversight evidence
Capacity that doesn't scale
At one credit union, manual funding review ran 40 minutes to 3.5 hours per application (vendor-reported).
Costs: hiring for peaks; backlog
06 · The rules, October 2026
Indirect auto lending compliance: what changed, and what still applies at funding
In shortThe federal CARS Rule is gone, but California has its own, and a 2026 multistate settlement put dealer add-ons on the lender. Truth in Lending, the Holder notice, odometer and state caps still apply at funding.
Status as of October 2026. This is an overview for operations teams, not legal advice: confirm each point with your counsel and for your states.
Swipe sideways to see the full timeline.
Rule
Status as of October 2026
What it means at funding
FTC CARS Rule (Combating Auto Retail Scams)
Not in effect Vacated by the Fifth Circuit on Jan 27, 2025 and withdrawn from the Code of Federal Regulations by a Federal Register notice of Feb 12, 2026; not re-proposed. The FTC still acts under Section 5 of the FTC Act: on Mar 13, 2026 it warned 97 auto dealership groups about advertised prices that left out mandatory add-ons and fees.
No federal CARS obligations, but the conduct it targeted (add-on misrepresentation, hidden fees) is still pursued under other law.
California CARS Act (SB 766)
In effect Oct 1, 2026 Optional add-ons must be disclosed in writing as optional before signing; add-ons with no benefit are banned; add-on records are kept for 2 years.
For California contracts, look for the optional-product disclosure and keep the consent records with the jacket.
Multistate attorneys-general settlement on add-ons
Announced Sep 17, 2026 40 attorneys general (some states count 41) settled with a large subprime auto finance company for $694M in cash and debt relief (one law firm cites $709.5M). The allegation: dealers packed contracts with unwanted service contracts and GAP. The company must monitor dealers and cap certain vehicle prices at 109% of retail book for seven years.
Regulators expect the lender to police dealer add-on practices. Program caps checked at funding and add-on statistics by dealer are evidence of that oversight.
ECOA / Regulation B
Amended, effective Jul 21, 2026 The CFPB's final rule (published Apr 22, 2026) removes the "effects test" (disparate impact) from Regulation B and narrows the discouragement provision. Intentional discrimination remains unlawful, and state laws (for example New York and Illinois) may still allow disparate-impact claims.
Markup caps and dealer-level markup monitoring remain prudent, and the dealer program sets the caps anyway. Keep application records 25 months (12 CFR 1002.12).
Dealer markup history
Background CFPB Bulletin 2013-02 held indirect lenders responsible for markup disparities; Congress disapproved it (Public Law 115-172, May 21, 2018). 2015 consent orders set the template many lenders still use: American Honda Finance capped markup at 1.25 points for terms of five years or less and 1.00 point for longer terms; Fifth Third had similar caps.
Check the contract rate against buy rate plus your cap on every deal.
Truth in Lending / Regulation Z
In effect The dealer gives the closed-end disclosures (12 CFR 1026.18). APR tolerance ⅛ point; finance charge ±$10 over $1,000 financed. Records kept 2 years (1026.25). An assignee is liable for violations apparent on the face of the documents (15 U.S.C. 1641(a)).
Recompute the Truth in Lending box before buying the contract.
FTC Holder Rule (16 CFR 433.2)
In effect The contract must carry the Holder notice; the holder is subject to the buyer's claims and defenses against the seller, capped at amounts paid.
A missing notice is a funding defect.
CFPB supervision of add-ons
Reduced posture Supervisory Highlights (Oct 2024) found unfair, deceptive or abusive practices around add-on refunds and GAP. Press reports describe sharply reduced CFPB enforcement in 2025–26.
Treat earlier findings as a map of risk; states and attorneys general are filling the gap.
State rate and fee caps
Varies by state Examples: Texas publishes add-on rate charts by vehicle class; Florida caps add-on rates per $100 by class (§520.08); Ohio allows up to 25% APR (ORC 1317.061); Mississippi's 2026 chart runs from 18% to 28.75% by vehicle age. Many states also cap documentary fees.
Check rate, fees and the state form for the contract's state verify for your states.
Odometer disclosure
In effect 49 U.S.C. 32705 and 49 CFR Part 580 require a signed mileage disclosure on transfer; since Jan 1, 2021, model year 2011 and newer stay in scope for 20 years (NHTSA).
The contract's mileage must match the signed disclosure; mileage drives book value and LTV.
eContracts and eVaults
In effect A lender perfects its interest in electronic chattel paper by control of the single authoritative copy (UCC 9-105). The 2022 UCC amendments are being adopted state by state (New York, effective Jun 3, 2026).
Confirm the eVault transfer of control as part of funding.
Electronic lien and title (ELT)
Spreading More than 25 states run optional or mandatory programs (2023 count); Illinois requires ELT from Jul 1, 2026 for institutions processing five or more liens a year. AAMVA keeps the current list.
Track every title until the lien is recorded.
Military Lending Act: purchase-money vehicle loans are exempt, but whether financing credit-related products such as GAP affects the exemption has shifted over time: check with counsel. Record retention at a glance: Reg Z 2 years; Reg B 25 months; California CARS Act add-on records 2 years; plus state rules, the dealer agreement and the life of the loan.
07 · How teams fund today
Four ways lenders review dealer contracts before funding
In shortMost lenders mix these. The real questions are how much of each jacket gets read, how fast a clean deal is paid, and whether a pend tells the dealer exactly what to fix.
1 · Common
An analyst reads every jacket
A funder opens the documents, checks them against the approval and a checklist, and writes the pend by hand.
Catches
What an experienced funder notices
Misses
Speed at Monday and month-end peaks; consistent pend wording
2 · Built in
Rules in the LOS or dealer platform
Field checks run on the data the dealer keys or the eContract carries.
Catches
Fields outside limits: amount, term, rate
Misses
What only the documents show: product forms, insurance cards, stips
3 · Outside
An outsourcer or CUSO funds
A partner reviews some or all deals, or takes overflow and after-hours work.
Catches
Capacity at peaks
Misses
Your view of their error rate, unless you measure it
4 · Automated
Software reads the whole jacket
It reads every document, checks every term against the approval and program, and drafts the pend. Several vendors offer it.
Catches
Every term on every deal, with the source page
Watch for
Who decides, how it shows evidence, whether it follows your program rules
Manual review and full-jacket automation, side by side
Question
Manual review
Full-jacket automation
What is checked?
What the funder has time for, often by checklist
Every term on every deal, against the approval and program
Clean deals?
Re-proved by hand, one by one
Shown as clean, for a person to fund
The pend?
Written by hand, sometimes by phone
Drafted with the exact fix, for a person to send
Peaks?
Overtime on Mondays and at month-end
The weekend backlog read before the team arrives
Evidence?
A checklist and the analyst's notes
Each check linked to the document it came from
Oversight by dealer?
Pend reasons kept as free text
Pend reasons and add-on prices counted by dealer
Whichever way you review, a person should own each funding decision, and anything off its approval belongs with credit.
Who is involved, and what each cares about
Lender side · bank, credit union or auto finance company
VP
Indirect lending manager · head of auto finance
Owns volume, the dealer network, yield and cost per loan. Cares about dealers sending deals elsewhere, funding service levels and audit findings.
DF
Director of dealer funding · contract processing
Owns time to fund, contracts in transit, pend rate and QC. Cares about Monday and month-end peaks, vague pends and missed wire cut-offs.
FA
Funding analysts · funders · contract processors
Check jackets, clear stips and fund. Care about not re-proving clean deals and not hunting across 20 documents.
CR
Credit and underwriting
Decide off-approval deals and waivers. Care about seeing drift early, with context.
Around the funding desk
DR
Dealer relations
Own dealer satisfaction and retention. Care about pends a dealer can act on, and data to coach a dealer.
CO
Compliance, risk and audit
Own fair lending, add-on oversight, Truth in Lending, the Holder notice and retention. Care about evidence by dealer for examiners and attorneys general.
IT
IT and integration
Care about the LOS, dealer platform feeds, eVault, core and ELT links, security review and data residency.
FI
Dealer F&I manager
Submits the jacket and fixes pends. Cares about knowing exactly what to fix, and getting paid.
GM
Dealer principal or controller
Watches contracts in transit, floor-plan interest and the reserve statement.
Part B · See it working
Botminds Indirect Auto Contract Funding, one deal at a time
Part A described the work without naming a product. This part shows how Botminds does it at a fictional lender, Sablewood Federal Credit Union, on a Monday morning with 64 deal jackets in since Friday's cut-off. Every name, dealer and figure in the screens is sample data.
08 · One deal, start to finish
Follow Andre Polk's contract from the jacket to the wire
In shortThis is the example from the top of the page, inside the product. Agents read and check the jacket. A funding analyst decides, and nothing is paid without her.
1
The contract is laid over its approval
The agents read the signed contract from Hollister Ridge Auto Group and lay each part of it over the credit union's approval and program rules. Each box turns green when it matches, or red when it does not.
Seven boxes are green. Three are red. The odometer box says it plainly: 21,406 on the contract, 24,106 on the signed disclosure.
The contract over its approval. The analyst works only the red boxes.
2
Each problem is explained, with its source
GAP is priced at $1,195, and the program's cap for a used vehicle is $995. The mileage digits are swapped. The insurance card names no lienholder, and its collision deductible of $2,500 is above the $1,000 limit.
Everything else matches the approval. The rate sits at the cap, not over it. Each fact links to the document it came from.
Ask in plain words. "Why does Andre Polk's deal pend?" Three exceptions, three fixes, each cited.
3
The analyst sends one clear pend
The verdict: pend, fixable today. Two problems are fixed by one re-signed contract, the third by an insurance binder. Nothing needs credit, because the deal stays inside its approval.
Renee Okafor, the funding analyst, reads the drafted notice and sends it to the dealer at 9:12 a.m. The dealer knows exactly what to fix.
Fund or pend. Three fixes, and 14 other terms and 16 checks that pass.
4
Re-signed, re-checked, funded the same day
The re-signed contract and the binder arrive by 10:40 a.m. The agents check the whole deal again, not only the fixes: 10 green, 0 red.
Renee funds it at 10:44 a.m., in time for the 3:00 p.m. wire. The dealer is paid the same day.
10 green, 0 red. The same contract, laid over its approval again.
09 · Every screen
The rest of the work: add-ons, insurance, stips, the wire and the dealer view
In shortOne deal is the story. These screens show each check up close, how wires are released, and what the lender learns about each dealer.
Add-ons over cap. GAP at $1,195 against a $995 program maximum for a used vehicle.
Insurance. No lienholder named, and a collision deductible above the $1,000 limit.
Every box opens its evidence. The red odometer box opens the signed federal disclosure: 24,106 miles, the same as the buyer's order and title application.
Contract math. Each product against its cap, and the back end against 20% of book. The Truth in Lending box is recomputed on the same screen.
Wire batch release. Checks before the cut-off, including dealer bank details unchanged for 90 days. A supervisor co-releases any batch above $1,000,000.
Stipulations. Income, residence and license cleared, each on a document that names the buyer; insurance open. Waiving a stip goes to the credit analyst who set it.
GAP at $995. After the re-sign, within the program cap.
Binder. Names the lender as lienholder, deductibles within limit.
Program rules as settings. Markup cap, LTV, PTI, participation and product caps, with every version kept.
Why deals pend, by dealer. Pend reasons per dealer, for coaching and oversight. Fictional sample data, not a result.
Select any screen to open it full size. Names, the credit union, dealers, buyers and figures are fictional.
10 · Agents, and who decides
Six agents read and check. Three people make the decisions.
In shortEach agent has one job and one thing it never does. People fund, decide anything off its approval, and release the wires.
The six agents
Deal jacket reader
Sorts and reads every document: contract, buyer's order, application, approval, stips, add-on forms, odometer disclosure, title application.
Never edits a document or a figure.
Contract-to-approval checker
Lays every term beside the approval and the program: amount, rate, markup, term, LTV, PTI, vehicle, buyers.
Never changes a credit decision; anything off-approval goes to a credit analyst.
Never clears a stip on a document that doesn't name the buyer.
Contract math and add-on checker
Recomputes the itemization and the Truth in Lending box; checks product prices against caps, the Holder notice and every signature.
Never accepts an add-on priced above its cap.
Funding recommender
Turns the checks into a fund-or-pend verdict, drafts the pend notice in plain words with the exact fix, and computes participation and proceeds.
Never funds: the analyst decides.
Batch and booking agent
Builds the wire batches before each cut-off and the booking data for the loan system.
Never releases a wire: people release, two above $1,000,000.
People decide; everything is cited
Decides every dealFunding analystFunds or pends each deal and signs it, within a per-deal limit; sends the pend notice after editing it.
Off-approvalCredit analystDecides anything that moves off its approval, and every stipulation waiver.
Dual controlFunding supervisorCo-releases any wire batch above $1,000,000.
Every check cites the document and passage it came from. Program rules (markup cap, product caps, LTV, PTI, stip freshness, income tolerance, insurance limits, cut-offs) are settings, versioned and on the audit trail. Funding clean deals without an analyst is a setting that starts off.
The flow. Jacket to reader, three checkers in parallel, a recommender, then a person: the funding analyst, or the credit analyst when the deal is off its approval.
The people in the screens
Fictional, from the working demo
RO
Renee Okafor · funding analyst
Owns 12 dealers; Monday starts with 64 jackets from the weekend.
"Most of my day is proving clean deals are clean."
MA
Monique Archer · director of dealer funding
Owns time to fund, contracts in transit and the audit findings.
"Dealers don't leave us over rate. They leave over a pend they can't understand."
PR
Priya Ramaswamy · credit analyst
Decides any deal that moves off its approval.
"If the contract isn't the deal I approved, it comes back to me."
KB
Kyle Brandt · dealer F&I manager
Submits 40–50 deals a month; floor-plan interest runs until funding.
"Tell me exactly what to fix and I'll fix it before lunch."
The bridge: Monique's line matches what dealers tell J.D. Power: experience factors make up 70% of the reasons they choose a lender, against 18% for rate. A pend the dealer can act on is a funding control and a dealer-retention tool at once.
Design targets, not results. The solution is designed for clean deals funded the same day they arrive complete (against the 3–5 business days lenders are reported to take), every pend sent with the exact fix in plain words, and analyst time per jacket cut from manual review (40 minutes to 3.5 hours at one credit union, vendor-reported) to deciding the exceptions. Measure them against your own baseline.
11 · What goes in and out
Deal jackets in, funding decisions out, with a person and a citation in between
In shortIt reads what the dealer and your systems already hold and writes nothing into the contract. What comes out is a verdict, a pend the dealer can act on, wire batches and booking data, and the evidence for each.
From the dealer
eContract or scanned RIC
Buyer's order and credit application
Add-on product forms
Odometer statement, title application or MSO
Stip documents: pay stub, bill, license
Insurance card or binder
Trade payoff letter
From the lender
The credit approval and its stips
Rate sheet and program rules
Dealer agreement terms
Dealer account and bank details
Cut-off times and authority limits
Agents read, check, recompute, draftAgainst the approval and the program, with every finding linked to the page it came from
A person funds
Audit trail on every step
Outputs
Fund-or-pend verdict with every check
Pend notice with the exact fix and deadline
Credit or fraud referral with the evidence
Wire batches before each cut-off
Booking data for the loan system
Pend reasons and add-on statistics by dealer
Audit record, exportable for examiners
What changes for the team
Clean deals stop being re-proved
Dealers get one clear notice, not a phone chase
Cut-offs and aging jackets visible all morning
12 · With what you already run
It works alongside your LOS, dealer platforms, eVault, core and ELT
In shortKeep the systems you run. The new piece reads the whole jacket against your approval and program, and hands back decisions and data.
Each system you run does one part of the job well. Botminds reads what they produce, checks the whole jacket against your approval and program, and hands back decisions and data. It connects to them; it does not replace them.
Dealer platforms and eContracting
Dealer-to-lender networks (RouteOne and Dealertrack are the category examples) carry applications, decisions and eContracts, and validate some data before submission. Their checks are generic across lenders; your approval and program still decide.
Your LOS and decisioning
Holds the approval, stips and funding checklist. Rules run on fields; agents read the documents behind the fields and return the verdict and booking data.
eVaults
Hold the authoritative copy of each eContract and transfer control to you (UCC 9-105). Agents read the vaulted contract; the vault stays the system of record.
Core and servicing
Funded deals are booked through a file or API with the data already checked. No core replacement.
ELT and the state
Electronic lien and title programs record your lien. Agents track each title until the lien is recorded and flag the ones that stall.
Stip verification, document tools and fraud scores
Income and identity verification services, document-extraction tools and fraud scores are inputs: agents use their results as evidence alongside the jacket.
Treasury and wires
Batches are built before each cut-off with dealer account checks; people release them under your dual-control limits.
Outsourced or CUSO funding
If a partner funds some or all of your deals, the same checks can review their decisions or take overflow.
Earlier automation
Scripted bots follow fixed steps and break on a new form. Agents read the document, explain what they found, and ask a person when unsure.
Products named above are examples of public integration categories, not partners or claimed integrations.
Three principles
1
Check the whole jacket, every time
Contract, math, products, stips, odometer and title against the approval and the program: every term on every deal, not a sample.
2
Agents recommend; people fund
An analyst funds or pends every deal, credit decides anything off-approval, and large batches need a second release. Every check cites its document.
3
Every pend is the exact fix
The dealer gets one plain message: what is wrong, what to send, and by when to make the next wire.
Built on Botminds
One platform for agents, data and screens
Builder: describe the work, get a spec, build it, test it.
Runtime: agents, data, integrations, screens and models, in one place.
SecureIsolatedGovernedTestedScalableHighly available
Enterprise-ready
Roles and separation of duties; per-analyst funding limits; maker-checker on wire release
Program rules as versioned settings, with every change audited
Audit export an examiner can read
Integrations with LOS, dealer platforms, eVault, core, ELT and treasury
Service levels and support; human approval and citations everywhere
Part C · Are you ready to automate?
Should you automate contract funding? A buyer's view
You run the funding desk, own the dealer network, or pay for both. This part is written for you: what slow or loose funding costs today, what automating the review could return, whether your team is ready, and what to ask any vendor, including us.
13 · What slow funding costs
What slow and inconsistent funding costs, before you buy anything
In shortStart from your own numbers: minutes per jacket, pend rate, days to fund. Once you trust them, you can judge any solution against them, ours included.
40m–3.5h
of manual review per application at one credit union. One institution, vendor-reported, so measure your own.
Multimodal, September 2026
841 vs 599
dealer satisfaction when an issue is solved in one contact, against two. A vague pend is the second contact.
J.D. Power, 2026
$694M
in a September 2026 settlement with 40 attorneys general over add-ons packed into dealer contracts. It required the finance company to monitor its dealers.
California Attorney General, 2026
Find your own numbers
Cost
Where it hides
How to find your number in a week
Review time
Funders re-proving clean deals, term by term
Time 20 jackets, clean and pended; multiply by deals a month
Pend rework
Calls, second submissions, second reviews
Your pend rate and how many pends needed a second contact
Peaks
Overtime on Mondays and at month-end
Monday backlog and overtime hours for the last quarter
Dealer wait
Contracts in transit on the dealer's books
Median days from complete jacket to wire, in the calculator in section 3
After funding
QC findings, repurchases, add-on refunds, unperfected liens
Last year's findings, by type, with dollars
Costs no calculator counts
The next dealDealers choose lenders on deal experience. A slow or confusing pend sends the next deal elsewhere.
Missed cut-offsA deal ready at 10:35 misses the 10:30 wire and waits for the afternoon or the next day.
Oversight evidenceWithout add-on prices and pend reasons by dealer, compliance cannot show it monitors its dealers.
Fraud at the last controlIncome misrepresentation and inflated book-outs are cheapest to catch before the money moves.
Your best fundersExperienced analysts spend their day proving clean deals are clean instead of judging the hard ones.
14 · Your return
What automating contract funding could return for you
In shortPut in your own numbers. Every default is an assumption, not a benchmark or a promise: replace each one with yours, and set a lever to 0 if it does not apply.
The value levers
Analyst time
Decide the exceptions, not every term
Agents check every jacket; analysts look at what drifts.
Clean deals make the next wire; dealers get their money sooner.
funded $ × days saved × dealer rate ÷ 365
Dealer flow
More deals from the same dealers
Dealers choose on experience. Set your own uplift: there is no public benchmark.
dealers × deals each × uplift % × margin per loan
Audit and leakage
Fewer findings after funding
Caps, math, stips and liens checked before the money moves.
findings avoided × cost to cure
Peak capacity
Mondays without overtime
The weekend backlog is read overnight and sorted before the team arrives.
peak backlog ÷ analysts vs. your service level
The calculator
Every default below is illustrative, not a benchmark or a promise. Replace each with your own number, and set a lever to 0 if it doesn't apply to you.
Review timePend reworkYour dealers' side
Illustrative yearly value
Review time$0
Pend rework$0
Your total$0
Staff hours freed a year0
Dealers' carrying cost avoided$0
Formulas, per year (contracts = monthly × 12): Review time = contracts × (today − with) ÷ 60 × hourly cost Pend rework = contracts × pend rate × touches × minutes ÷ 60 × share removed × hourly cost Dealers' carrying cost = contracts × amount × rate × days ÷ 365 (your dealers' gain, not in your total)
Where the defaults come from: the median wage for loan interviewers and clerks is $24.05 an hour (BLS, 2025); $32 assumes a 1.35 loaded multiplier. Vendor benchmarks put typical pend rates at 10–25%. Not counted: dealer volume uplift, audit findings avoided, peak overtime and fraud caught. Weigh the result against the cost of the solution and your rollout.
Start here15 · Readiness check
Are you ready to automate? Nine questions, five minutes
In shortPick the answer closest to your team today. Eight questions score 0 to 2; the last is for reflection. The meter adds them up as you go. Each question says why it matters and what a strong answer looks like.
1
Your indirect channel
How much of your auto lending comes through dealers?
Why it mattersThe more of your book arrives as dealer contracts, the more deal jackets there are to review, and the more funding speed decides where dealers send the next deal.
A strong answer"Indirect is our main auto channel, through a few hundred active dealers."
2
Volume
How many dealer contracts do you fund in a month?
Why it mattersVolume turns minutes saved per jacket into hours and dollars. Monday and month-end peaks test capacity hardest.
A strong answer"About 1,500 a month, with Mondays at three times a normal day."
3
Your baseline
Do you measure time to fund, pend rate and the top pend reasons, by dealer?
Why it mattersA pain with a number gets a budget. Without a baseline, no improvement can be shown to dealers, leadership or examiners.
A strong answer"Median 2.8 days from complete submission to wire; 22% of jackets pend; insurance and stale pay stubs lead."
4
Current process
How is a deal jacket reviewed today?
Why it mattersAgents earn their keep reading mixed documents (contracts, product forms, stips, insurance cards), checking them against your approval and program, and writing the pend. If most deals already fund with no touch, there is less to gain.
A strong answer"eContracts and paper, an LOS checklist, then an analyst reads the jacket and writes the pend by hand."
5
Systems and data
Can your LOS, dealer platform feeds, eVault and core or servicing system be read and updated through APIs or file feeds?
Why it mattersIntegration sets the effort and the risk. Systems reached through screens can still work, but take longer.
A strong answer"Our LOS has APIs, jackets land in one document store, and booking takes a file."
6
Decision owner
Is there a named leader who owns time to fund and dealer experience, with a budget to change them?
Why it mattersFunding work crosses teams (funding, credit, dealer relations, compliance, treasury) and stalls without one owner.
A strong answer"Our VP of indirect lending owns it and has budget this year."
7
Timing
Is there a date that matters: a budget cycle, an LOS conversion, a dealer program change, an exam, a new state rule?
Why it mattersA real date turns interest into a plan.
A strong answer"Budget is set next quarter, and we add California dealers next year."
8
Governance
Do you have a way to approve AI for use: risk review, a person signing each funding decision, an audit trail?
Why it mattersAgents should recommend and people should fund, with every check on the record. A clear approval path shortens the road to production.
A strong answer"Model risk and vendor management have a framework, and funding authority stays with named analysts."
9
Vision · for reflection
If 64 deal jackets arrived over a weekend, what would have to be true for every clean one to be funded in Monday's morning wire, and every other one to be back with its dealer, with the exact fix, by 9:30?
Why it mattersYour answer lists the blockers. Each blocker is a candidate job for an agent: reading, checking, recomputing, drafting or batching.
A strong answerA short, specific list, e.g. "jackets read overnight, stips cleared automatically, pends drafted for us to send."
What your score means
0–6
Explore
Build the baseline: count jackets, time a few reviews, rank pend reasons, name an owner. Start with a narrow, low-risk job such as reading incoming jackets and saying what is missing.
7–11
Ready to pilot
Pick one program or a group of dealers and run a measured pilot on your own jackets, with your analysts funding or pending every deal.
12–16
Ready to implement
Plan the full flow: LOS and dealer-platform integration, program rules as settings, approval limits, wire batches, booking, audit export and service levels. Roll out dealer group by dealer group, measured against your baseline.
16 · Questions for any vendor
Sixteen questions to ask before you buy contract funding automation
In shortAsk every vendor the same questions, us included. The answers show how the work is really done: what gets read, what evidence you keep, and who is allowed to fund.
Ask
A good answer
A warning sign
What gets checked
Does it check every term on every deal, or a sample?
Every term, every deal, against your approval and program
Only the fields the dealer keyed
Which documents can it read?
eContracts and scans, add-on forms from many administrators, state contract forms, insurance cards and stips
Only structured eContract data
Does it recompute the Truth in Lending box?
Yes, with the Regulation Z tolerances, plus the Holder notice and state form
"The dealer platform validates it"
What happens when it is unsure?
It says so and a person decides
It picks the likely value
Evidence
Can I see the document behind every check?
One click from the check to the passage
A score or a flag with no page
Could we rebuild one funding decision for an examiner in an hour?
Every document, check and decision, exported from one place
Evidence lives only inside the vendor's screens
Rules
Does it follow our program rules?
Markup cap, product caps, LTV, PTI, stip freshness and cut-offs as settings you can read, with every version kept
One built-in rulebook for every lender
Does the pend tell the dealer exactly what to fix?
A plain notice with each fix and the deadline for the next wire
A list of error codes
Control
Can the software fund a deal on its own?
Not by default. A named analyst funds or pends every deal, within a limit
"Clean deals auto-fund" from day one
What happens to a deal that is off its approval?
It goes to credit, with the evidence
The funder decides
Who releases the wire?
People, with a second release above your limit, and a check on changed dealer bank details
Batches go out automatically
Your data
Where does borrower data live, and who at the vendor can see it?
Isolated to your institution, in a region you choose, with access logged
Shared environments, vague answers
Is our data used to train anything?
No, in writing
"Only to improve the service"
Fit and proof
Does it work with our LOS, dealer platforms, eVault, core and ELT?
It reads from them and replaces none of them
A migration comes first
Will you run a pilot on our own jackets, measured against our baseline?
Yes: time to fund, pend rate, second contacts, minutes per jacket, QC findings
Demo data only
What does the price depend on?
A clear unit, such as contracts funded a month, and what is included
A number that appears only after the contract
17 · How to start
How to start: five steps to a decision you can defend
In shortMeasure first, then pilot on your own jackets, then decide with numbers. None of these steps commits you to a vendor.
Time 20 jackets, clean and pended. Note the minutes and how many systems the funder opened. That is your baseline.
Rank your top pend reasons by dealer, and measure the median days from a complete jacket to the wire.
Gather 20 to 30 redacted deal jackets, with an approval path to share them.
Pilot one program or dealer group, with your analysts funding or pending every deal.
Decide with numbers: time to fund, pend rate, second contacts, minutes per jacket and QC findings, before and after.
The full working-session questions: 31 questions in 8 groups
A
Your indirect program
A1
How many active dealers send us contracts, and what is the mix of franchised and independent dealers?
Why it mattersDealer count and mix set the variety of forms, products and habits a funding team sees.
Strong answer · warning sign
A list with volume per dealer.
"We'd have to pull a report."
A2
How are our dealers paid: markup with a cap, flat fee, or both? Are reserve chargebacks and repurchase terms clear in the dealer agreement?
Why it mattersParticipation is computed on every funded deal, and errors here are dealer disputes.
Strong answer · warning sign
One written method, with caps and chargeback rules.
Exceptions negotiated dealer by dealer, undocumented.
A3
Who funds today: our own team, a CUSO or service bureau, an outsourced overflow or after-hours service?
Why it mattersDecides where agents help: alongside your analysts, or checking work that comes back from a partner.
Strong answer · warning sign
A clear split, with quality measured either way.
No view of an outsourcer's error rate.
A4
How many programs, tiers and rate sheets does a funder have to keep in mind, and where are those rules written down?
Why it mattersRules that live in people's heads can't be checked consistently or audited.
Strong answer · warning sign
A maintained program guide, versioned.
"Ask the senior funder; she knows the tiers."
B
Volume and timing
B1
How many contracts do we fund a month, and how many jackets are open right now, by age?
Why it mattersThe base for every other number, including the calculator in section 14.
Strong answer · warning sign
Tracked daily, with an aging view.
Needs a special report to answer.
B2
What does Monday morning look like after a busy weekend, and what happens at month-end?
Why it mattersPeaks, not averages, set staffing and decide whether dealers wait.
Strong answer · warning sign
A peak plan and a measured Monday backlog.
"We just do overtime."
B3
What are our wire or ACH batch times and cut-offs, and how many funded deals miss a batch?
Why it mattersA deal ready five minutes after the cut-off waits hours or a day.
Strong answer · warning sign
Batch times known; misses counted.
Nobody tracks which deals missed.
B4
What is our median time from a complete submission to the dealer being paid, and how does it compare with the 24–48 hours some programs reach?
Why it mattersThe headline number dealers feel. Vendor benchmarks put typical bank indirect funding at 2–4 days.
Strong answer · warning sign
A median and a trend, by dealer.
Measured from the wrong start, or not at all.
C
Current process and tools
C1
Take the last deal that pended: how did the jacket arrive, who touched it, which systems, and how long did each step take?
Why it mattersOne real deal shows the hand-offs, re-keying and waits better than any process map.
Strong answer · warning sign
We can trace it step by step.
Nobody can reconstruct it.
C2
Which checks does our LOS or dealer platform already automate, and which still need a person reading the documents?
Why it mattersRules engines check fields; the long tail of documents (product forms, state contract variants, insurance cards) is where people still spend their time.
Strong answer · warning sign
A list split into automated and manual.
"The system handles it," with no list.
C3
How are pend notices written and sent, and how many pends need a second contact with the dealer?
Why it mattersJ.D. Power's 2026 study: 841 satisfaction when an issue is solved in one contact, 599 when it takes two.
Strong answer · warning sign
Templated notices with the exact fix; second contacts counted.
Pends by phone, from memory.
C4
When a corrected contract comes back, is the whole deal re-checked, or just the fixed item?
Why it mattersRe-signs can introduce new errors; a full re-check is cheap when agents do it.
Strong answer · warning sign
A defined re-check.
Funded on the dealer's word.
D
Pain and cost
D1
What is our pend rate, and what are the top five pend reasons, by dealer?
Why it mattersVendor benchmarks call under 10% strong and 10–25% typical. Reasons by dealer show where coaching pays.
Strong answer · warning sign
A rate and ranked reasons per dealer.
Pend reasons kept as free text.
D2
How long does a funder spend on a clean jacket, and on one that pends? Has anyone timed it?
Why it mattersTime per jacket drives the ROI. At one credit union, manual review ran 40 minutes to 3.5 hours (vendor-reported).
Strong answer · warning sign
A short time study on 20 jackets.
"It depends."
D3
What did post-funding QC and audit find last year: repurchases, chargebacks, add-on refunds, unperfected liens?
Why it mattersFindings after funding are the most expensive defects to cure.
Strong answer · warning sign
Counts and dollars by finding type.
"No findings," without sampling.
D4
What do dealers complain about most, and have we lost dealers or volume over funding speed?
Why it mattersDealer experience decides where the next deal goes.
Strong answer · warning sign
Specific themes from dealer relations.
"Dealers always complain."
E
Documents and stipulations
E1
What share of our contracts arrive as eContracts, and what share on paper or as scans?
Why it matterseContracts bring structured data and an eVault; paper adds transit time, legibility and original-arrival checks.
Strong answer · warning sign
The split, by dealer.
Unknown.
E2
Which stips do we require, with what freshness and tolerance rules, and who decides a waiver?
Why it mattersClear rules can be checked on every document; a waiver changes the approval and belongs with credit.
Strong answer · warning sign
Written rules; waivers routed to credit.
Funders waive stips informally.
E3
How many states, contract forms and add-on administrators do our jackets come from?
Why it mattersVariety is where fixed rules and templates break and reading agents help most.
Strong answer · warning sign
An inventory of forms.
Nobody knows.
E4
Could we gather 20 to 30 redacted real deal jackets for a working session within two weeks?
Why it mattersReal jackets are the fastest way to see what automation would do with your work.
Strong answer · warning sign
Yes, with an approval path.
No way to share anything.
F
Integrations
F1
Can we read approvals, stips and deal data from our LOS through an API, or just through screens and reports?
Why it mattersThe approval is the yardstick for every check.
Strong answer · warning sign
APIs or an integration layer exist.
A conversion is planned with no date.
F2
How do jackets reach us from dealer platforms and eVaults, and where do pend notices go back?
Why it mattersA pend should land where the dealer works, not in an inbox.
Strong answer · warning sign
Known channels each way.
Pends by email and phone, nothing else.
F3
How do funded deals reach treasury for the wire and the core or servicing system for booking, and how do we track titles through ELT?
Why it mattersFunding isn't done until the money moves, the loan is booked and the lien is recorded.
Strong answer · warning sign
Files or APIs, with reconciliation.
Re-keyed by hand.
G
Compliance, fraud and risk
G1
Do we monitor markup and add-on prices by dealer, and could we show that monitoring to an examiner or an attorney general?
Why it mattersThe 2026 multistate settlement made dealer monitoring a remedy; states keep fair-lending tools after the Regulation B change.
Strong answer · warning sign
A monthly report by dealer, with actions taken.
Caps checked at funding, never reviewed by dealer.
G2
Do we recompute the Truth in Lending box and confirm the Holder notice and state form on every contract?
Why it mattersAssignees carry liability for errors apparent on the face of the documents.
Strong answer · warning sign
Every contract, recorded.
"The dealer platform validates it."
G3
How do we catch income misrepresentation, straw buyers and powerbooking before funding, and how do we protect wires against changed dealer bank details?
Why it mattersFunding is the last control before money leaves; fraud scores are a signal, not a decision.
Strong answer · warning sign
Defined checks, referrals to fraud review, account-change controls.
Fraud found at the missed payment.
G4
If an examiner asked, could we rebuild one funding decision, every document and check, in an hour? What does our AI policy require?
Why it mattersAn audit trail makes exams routine; knowing the AI approval path early avoids a late stop.
Strong answer · warning sign
Yes, from one place; a named AI reviewer.
Days across systems; "AI isn't allowed," with no path.
H
Decision, success and vision
H1
Who would sponsor this, and what would a pilot have to show: which numbers, measured how?
Why it mattersA measured pilot settles the decision.
Strong answer · warning sign
Time to fund, pend rate, second contacts, minutes per jacket, QC findings.
"Let's see what it does."
H2
Have we tried to fix this before, with rules, bots, an outsourcer or a new LOS module? What happened?
Why it mattersPast attempts show where the hard parts are.
Strong answer · warning sign
Clear lessons learned.
Nobody remembers why it stopped.
H3
Would we ever let software fund a clean deal without an analyst, and under what limits? Which steps would we never automate?
Why it mattersTells you where approvals belong, and lets you start with the analyst signing every deal.
Strong answer · warning sign
A clear position, revisited after a pilot.
"All of it," on day one.
H4
What if every dealer got its clean deals funded the same day, and every pend in one message with the exact fix?
Why it mattersLinks funding operations to dealer loyalty and volume.
Strong answer · warning sign
Leadership sees the growth case.
Nobody owns the dealer experience.
18 · Frequently asked questions
Indirect auto lending and contract funding: frequently asked questions
What is indirect auto lending?
Indirect auto lending is car financing arranged by the dealer. The buyer signs a retail installment contract with the dealer, and a bank, credit union or finance company that approved the deal buys that contract by assignment and pays the dealer. Indirect auto contract funding is the step where the lender checks the signed contract against its approval before paying. The Congressional Research Service reports that the vast majority of US auto loans, 83% according to one paper, are obtained this way.
What is a retail installment contract (RIC)?
A retail installment contract, or retail installment sales contract, is the credit contract a buyer signs with the dealer to buy a car over time. It carries the Truth in Lending disclosures (APR, finance charge, amount financed, total of payments and total sale price), the itemization, any optional products, the FTC Holder notice and the signatures, and it is assigned to the lender at funding.
What are contracts in transit and why do they matter?
Contracts in transit are the dealer's money for cars already delivered on credit but not yet funded by the lender. Until funding, the dealer may already have paid the floor plan, the trade-in payoff and commissions, so every day of delay costs the dealer interest and cash. Dealers notice which lenders fund fast, and J.D. Power's 2026 study found deal-experience factors outweigh rate in choosing a lender.
What are funding stipulations in auto lending?
Stipulations, or stips, are conditions attached to a credit approval that must be cleared before the lender funds: typically proof of income, proof of residence, proof of insurance naming the lender as lienholder or loss payee, and a valid driver's license. Each is cleared by a document that names the buyer and meets the program's freshness and tolerance rules.
What is dealer reserve, and are dealer markups capped?
Dealer reserve, or dealer participation, is the dealer's compensation for arranging the credit, usually a share of the spread between the lender's buy rate and the contract rate, or a flat fee, as set by the dealer agreement. Lenders cap the markup; 2015 federal consent orders set caps of 1.25 points for terms of five years or less and 1.00 point for longer terms, a template many lenders still follow.
Is the FTC CARS Rule in effect?
No. As of October 2026, the FTC's Combating Auto Retail Scams (CARS) Rule is not in effect: the Fifth Circuit vacated it on January 27, 2025, and the FTC withdrew it from the Code of Federal Regulations on February 12, 2026. California's own CARS Act took effect on October 1, 2026, and the FTC continues to act against deceptive add-on and pricing practices under Section 5 of the FTC Act.
What does the 2026 multistate add-on settlement mean for indirect lenders?
On September 17, 2026, 40 state attorneys general announced a $694 million settlement with a large subprime auto finance company over dealers packing contracts with service contracts and GAP. It requires dealer monitoring, among other remedies. The signal for every indirect lender is that regulators expect the lender to police dealer add-on practices, so product caps checked at funding and add-on statistics by dealer matter.
How long does dealer contract funding usually take?
It varies widely by lender and by how clean the deal is. A 2026 vendor benchmark for bank indirect programs calls 24 to 48 hours strong, 2 to 4 days typical and 5 or more days a warning sign, and one lender's 2026 article says many lenders still take three to five business days. Paper contracts add transit time, and a missed wire cut-off adds a day.
Can AI agents fund dealer contracts on their own?
They shouldn't by default. In a well-governed design, agents read the deal jacket, check every term against the approval and the program, clear stips, recompute the math and draft the pend notice, each with its source, while a funding analyst funds or pends every deal, credit decides anything off its approval, and large wire batches need a second release.
Does automated funding review replace our LOS or dealer platform?
No. It works alongside the loan origination system, dealer platforms and eContracting, the eVault, the core or servicing system and ELT: it reads what they hold, checks the whole jacket, and hands back verdicts, pend notices, wire batches and booking data.
19 · One-page summary
Indirect auto contract funding: one-page summary
Print this page
The readiness questions
How much of our auto lending comes through dealers?
How many dealer contracts do we fund a month?
Do we measure time to fund, pend rate and pend reasons, by dealer?
How is a deal jacket reviewed today?
Can our LOS, dealer feeds, eVault and core be read and updated through APIs or feeds?
Is there a named owner with budget?
Is there a date that matters?
Do we have a way to approve AI, with a person signing each funding decision and an audit trail?
What would it take to fund every clean weekend deal in Monday's morning wire?
Score 0–2 on 1–8: 0–6 explore · 7–11 pilot · 12–16 implement.
What the funding review checks
Contract vs approval (buyers, vehicle, VIN, amount, rate within markup cap, term, LTV, PTI) · math and Truth in Lending box · Holder notice and state form · add-on prices and forms · stipulations · odometer · title and lien · then fund, pend, credit or fraud review.
12 terms
Indirect lending
Dealer-arranged credit; the lender buys the contract.
RIC / RISC
The retail installment contract the buyer signs with the dealer.
Deal jacket
Everything the dealer sends for funding.
Stips
Conditions to clear before funding: income, residence, insurance, license.
Pend
A deal sent back to the dealer to fix.
Contracts in transit
Dealer's money for delivered, unfunded deals.
Buy rate · markup
Lender's rate; the contract rate above it, capped.
Dealer reserve
Dealer's share of the spread, or a flat fee.
Back end
Add-on products such as GAP and service contracts.
Truth in Lending box
APR, finance charge, amount financed, payments, sale price.
eVault
Holds the single authoritative eContract.
ELT
Electronic lien and title, recorded with the state.
20 · Sources · accessed October 2026
Sources
New York Fed, Household Debt and Credit Report, Q2 2026 (Aug 11, 2026). newyorkfed.org
Experian, State of the Automotive Finance Market, Q2 2026 (shares of financing and loans; average amount financed). experian.com
Congressional Research Service, IF13179, The Automobile Loan Market: Policy Issues for Congress (Mar 10, 2026). congress.gov
Callahan & Associates, "As Indirect Lending Falls, So Goes Auto Market Share" (Sep 10, 2024). creditunions.com
J.D. Power 2026 U.S. Dealer Financing Satisfaction Study, as reported by CBT News (Aug 12, 2026). cbtnews.com
Cox Automotive / Dealertrack press release (Jan 23, 2026): 86% digital-eligible, 61% with aftermarket products, 250,000 forms (vendor data). coxautoinc.com
DefiSolutions, Indirect auto lending for banks (Mar 20, 2026): funding-cycle and resubmission benchmarks (vendor). defisolutions.com
Dealertrack Digital Contracting launch (2019): one in four paper contracts with an error; five days in transit (vendor, dated). autosuccessonline.com
Informed.IQ survey of auto finance leaders, Auto Remarketing (Mar 27, 2024) (vendor survey). autoremarketing.com
Multimodal, credit union indirect auto lending (Sep 1, 2026): 40 minutes to 3.5 hours per application at one credit union (vendor-reported). multimodal.dev
NADA, Four Steps to Cash Management Success (contracts in transit). nada.org
Point Predictive 2026 Auto Lending Fraud Trends Report, as reported by AFSA (Apr 9, 2026). afsaonline.org
NCUA Letter 10-CU-15, Indirect Lending and Appropriate Due Diligence. ncua.gov
We use essential cookies to make this site work, and cookieless, anonymised analytics to understand how it's used. "Accept all" also turns on Google Analytics, which sets cookies. Privacy policy