Verisoul
Verisoul, based in Austin, Texas, sells fake account and fraud detection to fintechs, payment and crypto companies and other online businesses. It scores sign ups, logins, checkouts and withdrawals from device fingerprints, network and proxy signals, email and phone intelligence, location spoofing checks and account history, and links accounts that look normal alone but belong to the same person or fraud ring. Where customers permit it, signals are pooled across Verisoul's customer base to spot repeat abusers who rotate cards, emails, wallets or IP addresses.
Each result comes with its reasons, and teams use no code rules in a unified dashboard to allow, block, review or step up a user, restrict withdrawals or freeze a wallet. AI fraud agents help with investigation. Add ons cover identity checks: ID Check validates a government document in about 30 seconds for $0.25 a check, Face Match compares a selfie in about 10 seconds, and phone verification checks a number. A fintech can run the account checks before its own know your customer step.
Gains, a payments heavy rewards and gaming platform, reports a 0.11% chargeback rate and a 7% lift in margins after adopting Verisoul. Copper and WeWard are among the fintech users, and Clay, PlanHub and lemlist report savings from blocking fake sign ups. Verisoul runs on Google Cloud, processes and stores data in the United States, and has a SOC 2 Type II report from 2025. The core platform is quoted after a demo.
Capability Axes
Capability grades
15 of 15 axes rated · 12 graded A or B
Risk models are the product. They turn device, session, network, email, phone and account history signals into a likelihood that an account is fake or an action fraudulent, and link related accounts that look normal on their own. Rules and workflows decide what happens next, but without the models there would be no score or reason for them to act on.
Each score and decision arrives with its reasons, and the customer's own no code rules decide whether to allow, block, send to review, step up verification, restrict withdrawals or freeze a wallet. Investigators can open the account history and linked accounts behind a flag, and the privacy policy sends an affected person to the customer for human review of an automated decision. Verification checks feed that policy rather than imposing a decision of Verisoul's own.
Each automated assessment carries its reasons and supporting signals, the API documentation sets out the response fields for verification and risk, and Chalk's feature platform tracks which signals fed each decision. Named case studies report detection and false positive rates for single deployments. Validation against independent data, drift monitoring and model version control are not described.
Gains, a rewards and social gaming platform with heavy payment flows, reports a 0.11% chargeback rate, a 7% improvement in margins and more than $30,000 saved across 50,000 transactions after adding Verisoul against promotion, payment and redemption abuse. PlanHub blocked 11.3% more spam and fraud while holding false positives at 0.5%. Clay and lemlist report savings from blocking fake sign ups. None of the named results comes from a bank or lender.
The customer agreement bars using customer data to train or improve third party models and requires zero retention or equivalent terms from any outside AI service. Verisoul may use customer data to maintain and improve its own service, and cross customer fraud intelligence pools signals where customers permit it, keeping fraud prevention signals as an independent controller. Retention periods for those pooled signals and model records are not stated, and biometric data has its own consent and retention terms.
The privacy policy separates the two roles Verisoul plays: processor for a customer's data, and independent controller for the fraud prevention signals it keeps. It covers access, deletion, transfers and limits on fraud prevention disclosures, the customer agreement refers to a data processing exhibit and subprocessor schedule, and a biometric consent form governs document and face checks. A financial institution addendum for GLBA duties is not offered.
The trust center lists a 2025 SOC 2 Type II report from the auditor A-LIGN, a 2026 penetration test and data protection impact assessments for real time fraud detection and for ID verification with Face Match, all released on request. Public controls cover encryption, restricted access to keys and security monitoring, and the audit's scope and control period come with the report.
Verisoul is a technology supplier to fintech, payments and crypto businesses and holds no banking or payments license. Its terms and privacy material cover data protection, international transfers and biometric consent, and its checks produce evidence for a customer's own onboarding and fraud procedures, with regulated decisions left to that customer.
Face Match compares a selfie with a document photo and ID Check validates government documents, tasks where error rates can differ across demographic groups. Verisoul has released no fairness testing, demographic error rates, bias policy or independent assessment for them. What it does publish concerns consent and review: a biometric consent form setting out why face and document data are collected and shared, human review through the customer, and reasons attached to each risk result.
A person affected by an automated result can ask for human review through the customer, and the reasons and signals behind a result help the customer examine a dispute. The customer agreement offers correction or termination for certain security failures, but no reimbursement for fraud losses, chargeback guarantee or indemnity for a wrong risk result.
Verisoul builds its own risk models and uses Chalk's feature platform to feed them current data and record which signals shaped each decision, a role Chalk describes in its own case study. Google Cloud hosts the service and Cloudflare handles network delivery. Outside AI services under the customer agreement must meet zero retention terms, but those providers, and the models behind the verification add ons, are not named.
Web and native iOS and Android SDKs capture session and device signals, and documented APIs return risk results and verification outcomes, including Face Match calls to verify or authenticate a user. Fintechs pass the results into onboarding before identity verification, and payment teams into deposit, checkout or redemption flows. Packaged connectors for a named banking core or payment processor are not offered.
Verisoul is a hosted service, and its privacy policy states that it processes and stores data in the United States, with standard contractual clauses for transfers from abroad. The trust center lists Google Cloud locations in the US, the European Union, Singapore and India, but customers are not offered a choice of region, and there is no on premises or customer cloud deployment.
One rate is published: ID Check costs $0.25 per check. The core fake account and fraud platform, Face Match and phone verification are priced after a demo, since the pricing page now leads to a demo request, and the customer terms allow either an order form or an online subscription. Volumes, unit charges for the rest of the platform and implementation costs are set in the proposal.
Financial services and fintech, and payments and crypto, each have their own pages covering onboarding fraud, account takeover, multi account abuse, checkout and payout risk, with screening before know your customer and account funding. Copper and WeWard are among the fintech users and Gains the payments example. Most named customers come from software, market research and gaming, and regulated bank deployments are not named.
Alternatives to Verisoul
Other products that do the same kind of work as Verisoul, and what each one does.
GeoComply supplies location, device and identity intelligence to banks, payment providers, crypto platforms, merchants and regulated gaming operators.
SEON combines fraud prevention and anti money laundering compliance in one platform, built around enriching a thin signup input such as an email address, phone number or address into a wide risk picture.
Incognia answers the identity question with geography. Its premise is that where a person is, over time, is a harder credential to fake than what device they are holding, because people handle important account actions from places they habitually occupy.
MaxMind sells minFraud to payment and commerce teams that need to assess transactions, new accounts, and logins. Founded in 2002 and the maker of GeoIP, MaxMind says more than 100,000 businesses use its data.
Transmit Security sells Mosaic, a cloud platform that combines customer identity, fraud detection and identity verification for banks, insurers and other large consumer brands.
iProov Authentication is the returning user side of iProov's Biometric Solution Suite. It replaces passwords and one time codes with a face check for account login, account recovery, device binding, step up checks before high risk actions and multifactor sign in.
Pricing
Vendor-published figures are labeled as such. Figures labeled “Estimated” are derived from third-party sources and have not been confirmed by the vendor.
| Entry Price | Pricing Basis | Data Protection Terms | Implementation | Source |
|---|---|---|---|---|
|
ID Check $0.25 per check. Core platform quoted after a demo.
$0.25 baseline
|
ID Check is priced per check. The fraud platform and other add ons are sold by subscription or order form after a demo. | A data processing exhibit and subprocessor schedule form part of the customer agreement, with separate biometric consent terms. | Implementation fees are not listed and are set in the proposal. | Vendor Published |
ID Check validates a government document in about 30 seconds for $0.25 a check. The fake account and fraud platform, Face Match and phone verification are priced in a proposal after a demo, and the pricing page now leads to a demo request. Volumes and unit charges for the rest of the platform are set in the order form or subscription.