Days 1–30
Map the decisions
Inventory the decision points (targeting, approval, treatment, retention) and their KPIs. Baseline account-level profitability by segment, and meet the Marketing, Risk, and Analytics partners.
JD Case Study · Citizens
Head of Decisioning Analytics & Optimization
- Where
- US · multiple sites
- Posting
- Read the job description →
- Shelf
- All briefings
A "Head of" role owning decisioning analytics and the optimization framework for a card portfolio — segmentation, statistical models, test-and-control, account-level profitability. This is exactly the decision-management and optimization engine I built (funnel, experimentation, LTV), now at portfolio scale.
One kind of number here. The analysis below runs on illustrative figures invented to show method. No Citizens data is used or implied; the operating logic is the deliverable.
Jun 22, 2026
What they’re actually buying
The decision layer that lifts portfolio ROI
Build the decision-optimization layer that lifts portfolio ROI. Model the decisions — who to target, approve, and how to treat them — prove them with test-and-control, and wire the whole thing into an optimization framework that improves campaign and channel returns. Optimize the funnel, not one campaign at a time.
The first 90 days
Map the decisions, diagnose them, run one 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 optimization diagnostic
Ship a decisioning dashboard that shows the targeting → approval → activation → profitability chain by segment, and pinpoints where the framework is leaving ROI on the table.
Days 61–90
Run the first optimization experiment
Design a test/control on a decisioning change with profitability guardrails, deliver the business case in dollars, and stand up the experimentation governance and optimization cadence.
Signature analysis · illustrative data
Signature analysis: the portfolio decision funnel
The optimization insight: the stages aren’t independent. Tightening approval to cut losses can starve activation and profitability; loosening targeting floods the funnel with low-value accounts. The job is optimizing the chain jointly against account-level profitability — which is what a real decisioning framework does, and a campaign-by-campaign view never will.
Which decision in the portfolio funnel is worth the most to optimize?Indexed to the first stage
Eligible populationindexed to 100
Targeted
Approved / accepted
Activated
Profitable at 12 monthsthe only stage that matters
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 Citizens asks for | What I’ve already shipped |
|---|---|
| Decisioning analytics & optimization framework | A full decision funnel diagnostic that locates and sizes the ROI leak. |
| Test & control, statistical / segmentation models | An A/B + incrementality readout with guardrails and a sample-size calculator. |
| Account-level profitability / LTV | An LTV / unit-economics model where the loss rate is the central optimization lever. |
| SQL / SAS, advanced analytical tools | Show-the-SQL blocks, channel-aware, warehouse-ready. |
| Card / consumer-credit portfolio domain | Competitive spotlights across the consumer-credit and BNPL field. |
The same engine
The engine behind this portfolio is a decision-optimization framework: a funnel modeled end-to-end, experiments with profitability guardrails, and an LTV model that prices every trade-off. I built it for a lease portfolio; on a card book the levers are targeting, approval, and treatment — same machine, same job.
Decisioning is optimizing the whole funnel against profit, not tuning one campaign — and I’ve shipped exactly that, in public. Let’s talk about your ROI frontier. — Paul Brown