Our Research Methodology
How we collect, verify, and maintain source-traceable financial data without marketing bias or fabricated claims.
1. Source Hierarchy & Authority Rules
We classify all information into strict authority tiers. When two documents conflict, the higher authority source always overrides:
2. Handling Undisclosed Data
Many issuers do not publicly disclose fixed CIBIL score cutoffs or exact income criteria. We strictly distinguish between:
- KNOWN: The bank has published an exact figure in their schedule.
- NOT_DISCLOSED: The bank assesses applicants on holistic internal criteria without a public threshold. We never fabricate a fake minimum score.
- CONDITIONAL: Criteria depends on existing relationship, salary account, or fixed deposit.
3. Deterministic Arithmetic & Canonical Coverage
Audited: September 2026All GST calculations (18%), reward point conversions, fee waiver thresholds, and net yields are performed by deterministic application code using currency-safe Decimal arithmetic, never by speculative generative models.
4. CardIntel Score v1.0 — 100-Point Deterministic System
CardIntel strictly decouples objective product architecture from personal consumer spending. Every point is calculated from a published bank rule, raw verified data, or formula. Affiliate commissions and bank referral payouts have zero mathematical connection to card scoring or ranking.
Objective, standardized structural rating evaluated against the fixed ₹6,00,000/yr consumer basket.
Deterministic net annual return factoring your individual spending distribution, spend-waiver attainability, and lounge valuations.
Tailored alignment for category discovery (/best/cashback, /best/travel, /best/lounge, /best/low-fee, /best/fuel, /best/upi).
Calculates Effective Reward Return Rate across benchmark basket. Evaluated using lowest unrestricted published point valuation. 5.0% return serves as the 30/30 ceiling: MIN(30, Return / 5% × 30).
Recurring annual fee table: Lifetime Free = 12/12, ₹1–₹249 = 11, ₹250–₹499 = 10, ₹500–₹749 = 9, ₹750–₹999 = 8, ₹1k–₹1.5k = 7, ₹1.5k–₹2k = 6, ₹2k–₹2.5k = 5, ₹2.5k–₹3k = 4, ₹3k–₹4k = 3, ₹4k–₹5k = 2, ₹5k+ = 0.
Cap Retention Ratio = Actual Rewards After Caps / Uncapped Rewards. Uncapped cards receive 12/12. Cap Score = Cap Retention Ratio × 12.
Direct statement credit = 10/10, Bank-value = 9, Fixed-value vouchers = 8, Rewards catalogue = 7, Travel portal = 6, Transfer partners (no cash) = 5, Limited catalogue = 4, Merchandise = 2, Unclear = 1, None = 0.
Domestic visits (Max 5: Unlimited/12+ = 5, 8-11 = 4, 4-7 = 3, 2-3 = 2, 1 = 1) + International visits (Max 3: Unlimited/8+ = 3, 4-7 = 2, 1-3 = 1) + Spend condition (Max 2: None/≤₹10k = 2, ≤₹25k = 1, >₹25k = 0).
Stated issuer markup (excl. GST): 0% = 8/8, >0–0.5% = 7, >0.5–1.0% = 6, >1.0–1.5% = 5, >1.5–2.0% = 4, >2.0–2.5% = 3, >2.5–3.0% = 2, >3.0–3.5% = 1, >3.5% = 0.
Open retail / no published income = 8/8, ≤₹3L = 7, ₹3–5L = 6, ₹5–8L = 5, ₹8–12L = 4, >₹12L = 3, Relationship required = 2, Invite-only / private = 1, Discontinued = 0. Precedence rule: most restrictive applies.
5 fixed discrete pillars, exactly 1 point each: Travel (+1), Lifestyle (+1), Protection (+1), Fuel / Utility (+1), Digital / UPI / Payment (+1).
Base 5 points minus 1 point for each material friction: spend unlock hurdles, min transaction thresholds, manual opt-in/activation, redemption fees, or short expiry. (Caps are not double-counted).
5. Audit Case Study: SBI Cashback 1 April 2026 Revision
How CardIntel protects consumers from stale financial advice when terms change:
"Earn ₹5,000 monthly cashback purely on online shopping, making it a ₹1 Lakh/month card."
Monthly cashback is split into ₹2,000 max online (5%) + ₹2,000 max offline (1%), totaling a ₹4,000 aggregate cap. Digital gaming, tolls, and government payments are strictly excluded.
Provenance: Verified against SBI Cards & Payment Services circular ref SBIC/REV/2026/04.