GeoAnalytics

When the geography doesn’t add up, that’s the signal

GeoAnalytics compares a customer’s address, phone, IP, counterparty and bank to find mismatches, scores fraud signals such as elder abuse risk and banking outside of jurisdiction, and lets you set the weights to fit your risk appetite.

Cross-attribute matching

Six geographic comparisons on every call

Each returns state and CBSA match flags and a distance in miles.

Address to phone

Does the customer’s zip sit in the same state and CBSA as their phone area code, and how far apart are they?

IP to address

Country and state match and distance between the customer’s zip and the geolocated IP address.

Phone to IP

State match and distance between the phone area code and the IP location.

Address to counterparty

State and CBSA match and distance between customer and counterparty.

Counterparty to bank

State, CBSA and zip match between the counterparty’s address and their financial institution.

Nearest bank branch

Miles from the customer to the counterparty bank’s closest branch, plus that bank’s branch footprint by state and CBSA.

Worked example

One Zelle payment, five mismatches

From the sample request in the API docs: a customer with an El Paso, TX zip, a 718 (New York) phone number and an IP in Cupertino, CA receives $8,718.12 from a New York counterparty whose bank is in Bennington, VT.

ComparisonValues comparedState matchDistance
Address to phone79922 El Paso, TX vs. area code 718, NYNo1,906 mi
IP to address79922, TX vs. IP in Cupertino, CANo959 mi
Phone to IP718, NY vs. Cupertino, CANo2,561 mi
Address to counterparty79922, TX vs. 10001 New York, NYNo1,903 mi
Counterparty to bank10001, NY vs. The Bank of Bennington, VTNo—
Nearest bank branchCustomer zip vs. closest Bank of Bennington branch (05201)—1,974.36 mi
Fraud signals

Fraud-specific geographic risk

Purpose-built indicators for elder exploitation, money mules, gang-driven fraud and impersonation scams.

Elder abuse risk

FinCEN FIN-2022-A002$27B EFE suspicious activity

Zip codes classified by concentration of residents 62 and older, so payments into retirement-heavy areas get enhanced monitoring. FinCEN’s trend analysis found 155,000+ elder financial exploitation reports totaling $27B in one year, with scams making up 80%. A wire from a distant, unknown sender into a high-elderly zip is a stronger signal than local activity.

0-N/aNone
1-VLwVery low
2-LowLow
3-MedMedium
4-HghMonitor
5-VHhFlag
SAR: “EFE FIN-2022-A002” in Field 2

Banking outside of jurisdiction

FFIEC Suspicious Activity

Flags when a counterparty’s bank has no branches near their stated address. Combining routing number enrichment with branch distances exposes shell companies, nominee accounts and money mule operations. A customer in rural Montana banking through a South Florida institution is a pattern associated with structuring and layering.

Fields: counterparty-to-bank state, CBSA and zip match; closest branch distance; branch distribution

Gang-driven fraud

“Fraud is the New Dope”FinCEN MN Child Nutrition Alert

When identity, check or benefits fraud clusters in gang-active zip codes, that spatial pattern is invisible to traditional transaction monitoring. The Minnesota child nutrition case ($250M+) showed how geographic concentration of fraud can be detected. Uses GeoEnrich gang territory data.

Impersonation, APP and check fraud

FinCEN Mail Theft AlertZelle Refunds (2023)

A geographic gap between victim and beneficiary adds a scoring dimension to authorized push payment and impersonation scams. Stolen checks deposited far from the originating address trip several independent mismatch checks at once.

Entity risk framework

Defensible entity risk scoring

A control framework that combines GeoEnrich entity data into five risk dimensions.

S

Sector risk

MSB/NBFI, cash-intensive, independent ATM, export/import and payments-adjacent. NAICS-driven tiering rather than a hard blacklist.

E

Entity risk

New formation, no web presence or a newly created domain, no verifiable operating footprint, opaque ownership.

A

Address risk

Shared with many unrelated entities, registered-agent-only, or inconsistent with state of formation and transaction geography.

I

Impersonation risk

Name close to an established business, near-match domain, altered branding, inconsistent tax or licensing data.

B

Behavioral risk

Early wire activity, high-velocity funds, beneficiaries unrelated to the stated business purpose.

Configurable scoring

Set the weights to match your risk appetite

A Southwest border bank weights border proximity differently from a Northeast community bank. Every factor’s default can be overridden.

Risk factorWhat you can changeDefault
HIDTAContribution to the composite scoreData Derivatives default
HIFCAContribution to the composite scoreData Derivatives default
Southwest BorderProximity weightingData Derivatives default
Geographic Targeting OrdersContribution to the composite scoreData Derivatives default
Drug traffickingScore for each tier, Very Low to Very HighData Derivatives default
Industry riskScore for each NAICS-based tierData Derivatives default
International nexusScore for each cross-border tierData Derivatives default
TBMLVulnerability scoringData Derivatives default

Catch what single-field checks miss

See GeoAnalytics run against your institution’s footprint.