Kaaj raises $3.8M in seed funding to power the future of small business lending 🎉Read more
← Resources

Credit policy guide

Underwriting exception management: tracking policy exceptions, overrides, and exception rates

Last updated · Kaaj editorial team

Underwriting exception management is how a lender approves, documents, and monitors loans that fall outside its credit policy. Classify each exception (credit or underwriting exception, policy or procedural exception, documentation exception, or pricing override), record the reason code, mitigants, and approver, and track the exception rate, which is the share of approved loans with at least one exception, by type, product, and approver. Review whether exception loans perform worse than the rest, and apply exceptions consistently so similar borrowers are treated alike. Kaaj shows each credit rule as met or not met, calls out exceptions at the top of the memo, and surfaces similar past exceptions so decisions stay consistent.

How to manage underwriting exceptions in 8 steps

The goal is not zero exceptions. It is exceptions that are deliberate, documented, consistent, and monitored.

  1. Write thresholds that can be tested

    Exceptions are only visible if policy has clear limits, such as minimum time in business, credit score, DSCR, or maximum soft costs.

  2. Classify every exception

    Separate credit exceptions (the borrower misses a guideline), policy or procedural exceptions (a required step or approval is missing), documentation exceptions (a required document is missing or stale), and pricing overrides.

  3. Use reason codes

    A short, fixed list of reasons makes exceptions countable and comparable. Free text alone cannot be analyzed.

  4. Record mitigants and approver

    Every exception should state what offsets the risk, such as a strong guarantor, collateral, or cash reserves, and who approved it under which authority level.

  5. Set approval authority by exception type

    Minor exceptions can sit with a senior underwriter; larger ones, or several on one deal, should go to credit leadership or committee.

  6. Flag exceptions automatically

    When rules run on every file, exceptions appear before an underwriter spends time on the deal, not after approval.

  7. Track the exception rate

    Measure the share of approved loans with exceptions by type, product, channel, and approver each month.

  8. Watch performance and consistency

    Compare delinquency on exception loans with the rest of the book, and check that similar borrowers receive similar treatment.

Types of underwriting exceptions

TypeWhat it isExample
Credit or underwriting exceptionThe borrower misses a credit guidelineOwner score 640 when policy minimum is 660
Policy or procedural exceptionA required step or approval did not happenDeal above an officer's limit approved without second signature
Documentation exceptionA required document is missing, unsigned, or staleBank statements older than policy allows
Pricing overrideRate or fees outside the pricing gridRate reduced to match a competing offer
Collateral exceptionCollateral outside policyEquipment older than the maximum age

How to calculate exception rate

Use the same period and population every time so the trend means something.

MeasureFormula
Exception rateApproved loans with one or more exceptions ÷ total approved loans
Exceptions per loanTotal exceptions ÷ total approved loans
Rate by typeLoans with that exception type ÷ total approved loans
Rate by approverException loans approved by a person ÷ loans that person approved
Exception loan performanceDelinquency rate on exception loans vs. non-exception loans

What to record for every exception

FieldWhy it matters
Rule and threshold missedShows exactly how far outside policy the deal is
Reason codeMakes exceptions countable
MitigantsExplains why the risk was acceptable
Approver and authority levelShows the right person approved it
Similar past exceptionsSupports consistent treatment of similar borrowers
OutcomeLets you measure how exception loans perform

Frequently asked questions

What is underwriting exception management?

The process of approving, documenting, and monitoring loans that fall outside credit policy, including reason codes, mitigants, approval authority, exception-rate tracking, and performance review.

What is the difference between a policy exception and an underwriting exception?

An underwriting or credit exception means the borrower misses a credit guideline, such as a minimum score or DSCR. A policy or procedural exception means a required step or approval was not followed. Both should be documented and tracked.

How do you calculate an exception rate?

Divide the number of approved loans with at least one exception by the total number of approved loans in the same period. Track it by exception type, product, channel, and approver.

What is a good exception rate?

There is no universal number. What matters is that the rate is stable, explained by documented mitigants, consistent across similar borrowers, and that exception loans do not perform materially worse than the rest of the book.

Why do exceptions matter for fair lending?

Inconsistent exceptions can mean similar borrowers are treated differently. Clear criteria, documented reasons, and regular monitoring help show exceptions are applied consistently.

How can automation help manage exceptions?

Rules that run on every file flag exceptions before review, record which threshold was missed, and keep the evidence. Kaaj shows each credit rule as met or not met, calls out exceptions at the top of the memo, and surfaces similar past exceptions.

See every exception before the underwriter does

Kaaj applies your credit rules to every file, calls out exceptions with evidence, and surfaces similar past exceptions for consistency.

Book a demoSee Kaaj credit memos
Automated credit decisioning with human review →How to automate credit memos →Policy exception (glossary) →Kaaj credit memo automation →