Data products
Every product owned, contracted and classified
Open a product to see its field-level contract, its lineage, and a preview scoped to the persona you pick.
Raw authorization stream from the issuer switch. Immutable and append-only; the source of truth for approve and decline.
Settled, deduplicated transactions with merchant category and program code. The product analysts and data scientists build on.
Cardholder profile: segment, tenure, region, credit line, attrition flag and contact data. Contact fields are masked unless the persona and purpose allow.
Conformed merchant dimension with 4-digit MCC, category, risk tier and card-not-present flag.
Points earned and redeemed by category, keyed to cardholder and program. Feeds rewards liability and redemption analytics.
Dispute lifecycle with network reason codes and outcomes, keyed to cardholder and program.
Model-ready features for real-time fraud scoring: velocity, merchant risk, card-not-present, amount z-score.
Contracted metrics for the Aurora co-brand program by contracted dimensions. No cardholder-level rows; any cell under 25 cardholders is suppressed; every metric maps to a contract clause.
Product briefs
Three further products worked through from problem to success measures, on cited public inputs.
A card-scoped, de-noised, narrative-enriched product over CFPB complaints - emerging-theme detection before the regulator's annual report.
Credit & risk · real-timeA governed, reusable real-time risk score returned inside the auth window - published as a data product with contract, SLA, entitlements, and SR 26-2 monitoring.
Engagement · App Store / PlayA curated review corpus that mines the negative tail across 10M+ ratings to catch regressions in hours - curation is the moat that stops GenAI hallucinating on raw reviews.