Zarv
Zarv ID

Know the risk before you price it.

Behavioral scoring at onboarding — even with no history.

Static data answers who the client was. Zarv ID answers who they are right now — even with no history.

87%
Score accuracy
85%
Fraud reduction
<150ms
API latency
1–2 wk
Integration time

What Zarv ID does.

Block fraud at the source.

KYC & Identity Verification

Document verification (ID, driver's license, passport), facial biometry with liveness detection, deepfake and tampered document detection. Synthetic identities blocked. PEP screening and Zarv Restricted Profiles check.

Fraud at contracting costs more than preventing it. Zarv ID blocks before any contractual bond is established.

Fraud rings don't fool the graph.

Graph Intelligence — Relational Score

Every individual is analyzed in the context of their network. Detects fraud clusters: if someone is linked to 3+ flagged identities, their score rises. Identifies straw buyers and fraudsters invisible to bureaus.

Sophisticated fraudsters pass individual KYC. What gives them away is the network around them. The graph detects what documents don't show.

87% accuracy — validated with clients.

Behavioral Score (No Prior History)

Score calibrated for insurance and vehicle credit. Works for people with no prior policy or credit history. Detects geolocation shifts, atypical patterns, and incoherent history. Output: numeric score + risk band + flags.

Profiles without history get rejected or overcharged. With Zarv ID, you can price what was previously unknown risk — more volume, same loss ratio.

Underwriting-ready output.

Pricing Recommendation

Coming Soon

Not just a score — a premium band recommendation. Segmentation calibrated to your portfolio. API integration with your pricing engine. Compatible with existing actuarial models.

Underwriters don't need another score to interpret. They need a decision. Zarv ID delivers the decision — with data to back it.

Graph Intelligence — data sources

Zarv ID builds a relationship graph integrating these data sources:

Identity Data
Name, address, date of birth, SSN/EIN
Locations & Contacts
Phones, emails, historical addresses
Geolocation
Trajectories and location patterns
Corporate Relations
Partners, shareholdings, business links
Employment History
Jobs, positions, W-2/1099 records
Relationship Network
Connections with other individuals and entities
PEP
Politically exposed persons
Commercial Transactions
Purchases, sales, registered contracts
Financial Analysis
Income, assets, credit score
Debts & Liens
Defaults, tax liens, judgments
Legal Proceedings
Civil, criminal, labor lawsuits
Zarv Restricted Profiles
Proprietary high-risk profiles identified by Zarv

In practice.

Auto Insurance

New client underwriting — zero-km vehicle

Low bureau score due to lack of history → rejection or excessive premium → client leaves for competitor.

Behavioral graph reveals the driver lives in a low-risk area, works regularly, and has no links to suspicious profiles → competitive premium band → client accepts → policy issued with accurate pricing.

Auto Insurance

Identity fraud detection at renewal

Automatic renewal — no flags raised.

Relational risk flag: the holder's SSN is linked to 4 claims at other insurers in 18 months → underwriter reviews → fraud pattern identified → renewal denied.

Credit & Lending

Lending to a client with no vehicle credit history

Rejected due to unknown risk.

Graph shows consistent income patterns, stable employment location, active business links → positive risk scoring → financing approved with adequate guarantees → client stays current.

Credit & Lending

Dealership fraud detection (suspicious EIN)

Multiple applications processed without red flags.

Dealership EIN appears in the graph linked to other flagged EINs — fraud cluster pattern → automatic alert before credit release.

Fleet

Per-driver scoring for personalized fleet insurance

Flat premium for entire fleet, regardless of individual driver risk.

Each driver scored individually → fleet segmented into risk bands → insurer issues differentiated policy → company reduces total premium.

Proven results.

85% fraud reduction

Client performance (TagPro, B2C)

87% score accuracy

Validated in client operations

45% reduction in premiums lost to fraud

B2C — TagPro

18% fraud premium reduction

B2B — Arval / BNP Paribas

+3pp gross margin increase

TagPro — year over year after implementation

Proven at scale.

TagPro

Insurer · Auto Insurance (B2C)

$6M
in theft prevention saves in 2024
93%
vehicle recovery rate
+3pp
gross margin increase (YoY)
45%
reduction in premiums lost to fraud
1

New clients onboarded via Zarv ID (KYC + scoring)

2

Portfolio monitored in real time via Zarv Signal

3

Losses investigated via Zarv Lens with evidence generation

Arval / BNP Paribas

Leasing + Self-Insurance · Corporate Fleet (B2B)

85%
vehicle recovery rate
18%
reduction in fraud premiums
Auto
automated claims management
1

Fleet monitored continuously via Zarv Signal

2

Claims reconstructed automatically via Zarv Lens

3

Zarv ID integration underway for new lessee onboarding

Ready to integrate.

REST API · JSON response · SDK available
Average latency: < 150ms
Typical integration: 1–2 weeks
Webhooks for async alerts
Monitoring dashboard included

Per API request (pay-per-use). No minimums during pilot. Volume discounts available.

API Documentation
Security & Compliance

Your data, protected by design.

GDPR & CCPA compliant by design
SOC 2 Type II aligned infrastructure (certification in progress)
End-to-end encryption (TLS 1.3 + AES-256)
Role-based access control (RBAC) with MFA
Full audit trail for every API query
Data retention policies with auto-purge
Privacy by design — sensitive data never stored
Penetration-tested quarterly by third parties
Incident response SLA < 24 hours

See risk before it costs you.

GDPR & CCPA Compliant · No commitment · Integration in days