Days 1–30
Map the loss curve
Decompose fraud and credit losses by segment, channel, and vintage. Baseline the applications → approved → funded → first-payment-default → charge-off funnel, and meet Risk, Ops, and Analytics.
JD Case Study · Concora Credit
VP, Risk — Fraud Strategy & Analytics
- Where
- Beaverton, OR · hybrid
- Pays (est.)
- $130K–$180K (est.)
- Posting
- Read the job description →
- Shelf
- All briefings
A VP owning fraud strategy and the analytics behind it for a subprime credit-card book — the exact risk/decisioning world I lived in at Acima: income-and-bank-data underwriting, first-payment default, charge-offs, the approval/loss frontier. This is the closest fit in my portfolio to a hiring manager’s actual day.
One kind of number here. The analysis below runs on illustrative figures invented to show method. No Concora Credit data is used or implied; the operating logic is the deliverable.
Jun 22, 2026
What they’re actually buying
Protect the economics without strangling growth
Protect a subprime portfolio’s economics without strangling its growth. Own the fraud and risk-decisioning strategy, the models underneath it, and the test-and-learn that tunes the approval/loss frontier. Stripped down, the job is one trade-off, quantified: every basis point of fraud and credit loss against every approved account you turn away.
The first 90 days
Loss curve, decisioning frontier, first experiment
The order matters. Nobody should change a system they haven’t reconciled, and nobody should present a verdict before the numbers under it can be trusted.
Days 31–60
Stand up the decisioning frontier
Ship a risk-frontier dashboard — approval rate vs loss rate by segment — to show where tightening protects margin and where it’s quietly throttling good accounts. Identify the two or three highest-value moves.
Days 61–90
Run the first risk experiment
A/B a fraud-strategy or decisioning change with an explicit loss guardrail. Deliver the business case — loss reduction vs approval impact, in dollars — and set the model-governance and experimentation cadence.
Signature analysis · illustrative data
Signature analysis: the subprime decisioning funnel
Where strategy pays off: first-payment default is the earliest fraud/credit gate, and the cheapest to fix. A tuned fraud model cuts FPD with minimal approval drag — the prize is loss reduction without turning away good accounts. That’s the frontier I’d manage, not a blanket tightening.
Where on the approval-loss frontier does the book make its money?Indexed to the first stage
Applicationsindexed to 100
Approved
Funded
First payment cleared
Performing at 12 monthsthe book that pays
Requirement → proof
The posting, answered line by line
Each line of the posting against something already built and running on this site. Where the fit is a ramp rather than a match, the row says so.
| What Concora Credit asks for | What I’ve already shipped |
|---|---|
| Subprime risk / decisioning strategy | I modeled Acima’s funnel end-to-end: approval → delivery-verified funding → first-payment gate → charge-offs. |
| Experimentation with loss guardrails | An A/B / incrementality readout with a charge-off guardrail and a sample-size calculator. |
| Segmentation & loss / unit economics | An LTV / unit-economics model where the charge-off rate is the central lever. |
| SQL / SAS / Python | Show-the-SQL blocks, channel-aware, with the charge-off guardrail built in. |
| Subprime consumer-credit domain | A career in lease-to-own (deep-subprime), plus published spotlights across the subprime-credit field. |
The same engine
The engine behind this whole portfolio — approval/funding funnel, first-payment-default gate, charge-off economics, experimentation with a loss guardrail — is the same machine a subprime card book needs. I built it for lease-to-own; the merchandise changes, the risk/growth frontier doesn’t. I’d walk in fluent in the trade-off this role exists to manage.
Subprime risk isn’t a domain I’d ramp into — it’s where I’ve worked and what I’ve already modeled in public. Let’s talk about your loss curve. — Paul Brown