Quick Answer
Lending PM is a balancing act between growth, risk, and compliance. You want to approve more loans (growth), to the right borrowers (risk), within legal boundaries (compliance). Every feature decision shifts this balance. The best lending PMs think in terms of default rates and origination volume simultaneously, because optimizing one at the expense of the other destroys the business.
What Makes Lending PM Different
Credit risk is the product. Your underwriting model determines who gets approved, at what rate, and at what terms. The model is not a backend concern. It is the core product decision. A PM who does not understand credit risk is flying blind.
The full lifecycle matters. Lending products have distinct phases: origination (application, underwriting, closing), servicing (payments, statements, escrow), and collections (delinquency, workout, recovery). Each phase is its own product surface with different users and metrics.
Fair lending laws are strict. ECOA, TILA, RESPA, and state-specific lending regulations govern what you can build. Disparate impact analysis is required. If your AI model denies loans at different rates across protected classes, you have a legal problem. Fairness testing is a product requirement.
Money has a time value. Every day your origination flow takes longer than necessary costs the business. A loan that closes in 3 days instead of 30 days reduces cost of capital and improves borrower experience. Speed is money, literally.
Core Metrics
| Metric | Why It Matters | Good Benchmark |
|---|---|---|
| Origination volume | Total loan dollars originated. The top-line growth metric. | Growth-dependent |
| CAC | Cost per funded loan. Includes marketing, processing, and underwriting costs. | $200-1,000 consumer |
| Approval rate | Percentage of applications approved. Balance against default rates. | 30-60% varies by product |
| Activation rate | Approved borrowers who accept and fund the loan. Drop-off here means friction or poor terms. | 60-80% |
| Default rate | Percentage of loans that go delinquent. The risk metric that determines viability. | Under 5% for prime |
| Time to close | Days from application to funded loan. Speed drives conversion and borrower satisfaction. | 1-3 days personal, 15-30 days mortgage |
Frameworks That Work
RICE with the calculator works well for lending product decisions. Origination improvements often score highest on Reach (every applicant) and Impact (direct revenue). But servicing improvements score high on Confidence because you have clear data on payment failure rates and support ticket volumes.
The Business Model Canvas is useful when evaluating new lending verticals. Each product (personal loans, auto loans, mortgages, BNPL, SMB lending) has different unit economics, regulatory requirements, and distribution channels. Map the full business model before expanding.
Recommended Roadmap Approach
Structure your product roadmap around the lending lifecycle: origination, servicing, and collections. Each area needs continuous investment. Over-indexing on origination while neglecting servicing creates operational debt that compounds with portfolio growth.
Size each lending vertical with a TAM calculator before expanding. The addressable market for personal loans looks different from mortgage or SMB lending. Check roadmap templates for lifecycle-based planning formats.
Plan around regulatory examination cycles. OCC and state examiners will request evidence that your product processes are compliant. Build audit-friendly features (logging, adverse action notices, fair lending reports) alongside growth features.
Tools PMs Actually Use
Lending PMs work with loan origination systems (LOS), credit decisioning engines, and servicing platforms daily. Understanding how these systems integrate determines what you can build and how fast.
Credit bureau data (Experian, TransUnion, Equifax) and alternative data sources feed your underwriting models. You need to understand what data is available, what it costs, and how predictive it is.
For competitive analysis, use the competitor matrix to map traditional lenders against fintech challengers across speed, rates, and approval criteria.
Common Mistakes
Optimizing approval rates without watching default rates. Approving more borrowers feels like growth. But if default rates climb, your portfolio bleeds money on a 12-24 month lag. Always track approval and default together.
Ignoring the servicing experience. Borrowers interact with your servicing product for years after origination. A confusing payment portal or unclear statement drives support costs up and refinancing retention down.
Treating adverse action as an afterthought. When you deny a loan, federal law requires specific disclosures about why. The denial experience is a product surface. Clear, respectful adverse action notices reduce complaints and sometimes convert denials into approvals with modified terms.
Shipping ML models without fairness testing. Machine learning credit models can encode historical bias. If your model approves different rates across race, gender, or age, you face regulatory action. Test for disparate impact before deploying any model change.
Career Path: Breaking Into Lending PM
Credit analysis, underwriting, or loan operations experience translates directly. If you understand credit scores, debt-to-income ratios, and loan structures, you have domain knowledge that generalist PMs lack.
Data science backgrounds also translate well. Lending is one of the most data-driven verticals in finance. Check the salary hub for lending PM compensation. The career path finder can help map transitions from credit analysis, banking operations, or general PM into lending tech.