Home / Fraud Library / Operational Techniques / Using BIN Data in Fraud Rules
Operational Techniques

Using BIN Data in Fraud Rules: Mismatches, Prepaid Cards and Thresholds

Operational Techniques ~6 min read The Fraud Practice Library

A BIN lookup gives you a handful of facts about a card. The value comes from comparing those facts with the rest of the order and acting proportionately. This guide covers the rules that work, how to set thresholds, and the mistakes that turn BIN data into lost sales. For the basics of what a BIN is, start with the credit card BIN checks guide.

The fields to work with

From the lookup you typically get the issuer country, brand, card type (credit, debit or prepaid) and category (consumer, business or corporate). Combine them with what you already have: billing country, shipping country, IP geolocation, and the product being bought.

Rule 1: country mismatch

Compare the BIN's issuing country with the billing address, the shipping address and the IP geolocation. A card issued in one country, billed in a second, shipping to a third and ordered from a fourth deserves a closer look. A single mismatch usually doesn't.

  • Many legitimate customers mismatch: travelers, expats, military, students abroad and people buying gifts for someone in another country.
  • Weight the combination. BIN country differing from IP country, plus shipping to a freight forwarder, is far stronger than either alone.
  • Prefer routing to manual review over declining, unless your own history shows a mismatch segment is almost all fraud.

Rule 2: prepaid and virtual cards

Prepaid cards are not fraud by themselves; many people use them for budgeting or privacy. They are riskier where the customer can get value without paying again, so apply the rule by product:

  • Subscriptions, free trials and recurring billing: a prepaid card can run dry or be abandoned. Ask for an alternative payment method or add friction.
  • High-value or easily resold goods: treat prepaid as an added risk signal.
  • Low-risk, one-time purchases: usually leave alone.

Rule 3: validate brand and type inline

Detect the brand from the BIN as the customer types and select it for them, or warn when the number doesn't match the brand they picked. This isn't a fraud rule so much as a conversion one: it catches typos early and reduces failed authorizations, which keeps your decline rate cleaner.

Rule 4: card category in context

A business or corporate card on a consumer order isn't suspicious in itself, but it can change what you expect: different spending patterns, different billing addresses and often higher limits. Use category as context for other rules, not a trigger on its own.

Rule 5: breach fallout

After a known data breach or an issuer alert, tighten rules for the affected BIN ranges for a limited time: more manual review, extra verification or lower velocity limits. Remove the extra friction once the window passes. See consumer alerts and hot, warm and positive lists.

If you use a card-testing intelligence feed, act on the first sighting and don't age entries out after 30 or 60 days. In a myNetWatchman analysis of 18.8 million tested cards, 81.9% were tested on a single day and never seen again, so a rule that waits for a second sighting never fires on most cards. Cards tested on two separate days averaged 66 days between them, so a quiet spell is not proof a card is dead. For prioritizing alerts, the number of distinct testing days is a better signal than raw attempt count: the 7% of cards tested on three or more days accounted for 33% of all testing.

Rule 6: velocity and performance by BIN

Track attempts, declines and chargebacks by BIN. A sudden burst of attempts from one BIN, especially many different cards from the same device or IP, points to card testing. Over time, you'll also see BINs that consistently produce more fraud for you. Treat that as a reason for more scrutiny, not a permanent block, because honest customers share the same BIN.

Setting thresholds from your own data

  • Segment your past orders by each BIN attribute (country match, prepaid, category) and measure chargeback rate and false-positive rate for each segment.
  • Add the signal to a rule or a fraud score only where it separates good from bad in your data.
  • Use three bands: auto-accept, review, and decline. Send mismatches to the review band first.
  • Recheck quarterly. Fraud patterns and BIN tables both change.

Pitfalls

  • Old tables. Use a provider that refreshes regularly and supports eight-digit BINs.
  • Hard country blocks. Declining every mismatch rejects good customers. Measure before you block.
  • Treating BIN data as identity. Pair it with AVS, CVV and device identification.
  • No feedback loop. Feed review outcomes and chargebacks back into the rules.

Want a second pair of eyes on your BIN and card-testing controls? Talk to a senior consultant: vendor-neutral, no junior hand-offs. See our consulting services.