Scoring Methodology

Strategic Alignment

Measures how well the vendor's solution aligns with your institution's strategic priorities and readiness.

(priority_match × 30%) + (member_relevance × 30%) + (timing × 20%) + (board_readiness × 20%)

Parameters

• •Priority Match: Your primary strategic priority vs. vendor category (Member Experience, Risk & Compliance, Operational Efficiency, Member Growth)
• •Member Relevance: Digital adoption rate, avg member age, product mix (e.g., auto loans %)
• •Timing: Your AI implementation experience vs. vendor complexity
• •Board Readiness: Board AI literacy, governance policy, vendor model transparency

Interpretation: ≥70 = strong alignment; 40–69 = partial fit; <40 = misalignment

Technical Fit

Assesses technical compatibility and implementation readiness.

(core_integration × 35%) + (infrastructure × 20%) + (data_readiness × 25%) + (team_capacity × 20%)

Parameters

• •Core Integration: Your core banking system vs. vendor compatibility (Production/Beta/None)
• •Infrastructure: Cloud posture, dedicated IT staff vs. vendor deployment model
• •Data Readiness: Data quality self-assessment, API gateway
• •Team Capacity: IT staff count vs. typical implementation weeks

Interpretation: Core integration can be a Blocker; other scores indicate readiness gaps

Regulatory Readiness

Evaluates regulatory and governance preparedness for AI deployment.

(MRM × 25%) + (fair_lending × 20%) + (vendor_compliance × 25%) + (exam_readiness × 30%)

Parameters

• •Model Risk Alignment: MRM framework, vendor SR 11-7 alignment (for lending AI)
• •Fair Lending: Adverse action support, bias testing (for lending vendors)
• •Vendor Compliance: Regulatory score, SOC2, NCUA examination ready
• •Examination Readiness: Your exam rating, governance policy

Interpretation: Critical for lending AI; governance gaps flagged as prerequisites

Risk Score

Overall risk level; lower is better. Combines vendor, implementation, regulatory, strategic, and member impact risks.

(vendor × 25%) + (implementation × 25%) + (regulatory × 20%) + (strategic × 15%) + (member_impact × 15%)

Parameters

• •Vendor Risk: Vendor risk score and flags
• •Implementation Risk: Core integration risk level (Blocker/High/Medium)
• •Regulatory Risk: Inverse of regulatory readiness
• •Strategic Risk: Lock-in, switching costs
• •Member Impact: Fair lending, bias risk (for lending)

Interpretation: >70 may block deployment; mitigations recommended

Expected 3Y ROI

Our estimate for your institution — not the vendor's marketing number. We adjust vendor case study results for your size, digital adoption, and data quality.

Parameters

• •Base effect from vendor case studies
• •Adjustments: institution size, digital adoption, data quality
• •Financial projection: operating expense, members, vendor cost

Interpretation: ≥50% = favorable; ≥25% = moderately favorable; <10% = unfavorable

Payback Period

Months to recover implementation and annual costs from net benefits. Shorter is better.

Parameters

• •Year 1–3 net benefits from our projection
• •Implementation cost and annual vendor cost

Interpretation: Target <24 months

Diversification Across Risk Drivers

How well your portfolio is spread across different risk sources, not just sectors. Higher = less hidden concentration — one macro shock is less likely to hit everything at once.

Parameters

• •We model which factors drive which assets
• •Compare observed links vs. links after stripping shared macro drivers
• •Scale 0–5

Interpretation: Higher = better diversification across risk drivers

Worst-Case Loss Under Stress

Estimated maximum portfolio loss when we simulate shocks (e.g., Fed raises rates 50bp, VIX spike). Illustrative, not a forecast.

Parameters

• •Simulated shocks: rate hike, volatility spike
• •Impact on your portfolio returns

Interpretation: Lower % = better resilience under stress

How Much We Trust the Vendor's Claims

Vendor case studies often lack control groups and may overstate results. 5 = RCT or quasi-experimental; 3 = pre/post without controls; 1 = marketing claim only.

Parameters

• •5: Gold standard — randomized or quasi-experimental
• •3: Pre/post comparison, no control group
• •1: Marketing claim only

Interpretation: Higher = more reliable basis for our adjustment